Siewert-Jan Marrink

The MARTINI Coarse-Grained Force Field: Extension to Proteins

Many biologically interesting phenomena occur on a time scale that is too long to be studied by atomistic simulations. These phenomena include the dynamics of large proteins and self-assembly of biological materials. Coarse-grained (CG) molecular modeling allows computer simulations to be run on length and time scales that are 2-3 orders of magnitude larger compared to atomistic simulations, providing a bridge between the atomistic and the mesoscopic scale. We developed a new CG model for proteins as an extension of the MARTINI force field. Here, we validate the model for its use in peptide-bilayer systems. In order to validate the model, we calculated the potential of mean force for each amino acid as a function of its distance from the center of a dioleoylphosphatidylcholine (DOPC) lipid bilayer. We then compared amino acid association constants, the partitioning of a series of model pentapeptides, the partitioning and orientation of WALP23 in DOPC lipid bilayers and a series of KALP peptides in dimyristoylphosphatidylcholine and dipalmitoylphosphatidylcholine (DPPC) bilayers. A comparison with results obtained from atomistic models shows good agreement in all of the tests performed. We also performed a systematic investigation of the partitioning of five series of polyalanine-leucine peptides (with different lengths and compositions) in DPPC bilayers. As expected, the fraction of peptides partitioned at the interface increased with decreasing peptide length and decreasing leucine content, demonstrating that the CG model is capable of discriminating partitioning behavior arising from subtle differences in the amino acid composition. Finally, we simulated the concentration-dependent formation of transmembrane pores by magainin, an antimicrobial peptide. In line with atomistic simulation studies, disordered toroidal pores are formed. In conclusion, the model is computationally efficient and effectively reproduces peptide-lipid interactions and the partitioning of amino acids and peptides in lipid bilayers.

Toroidal pores formed by antimicrobial peptides show significant disorder

A large variety of antimicrobial peptides have been shown to act, at least in vitro, by poration of the lipid membrane. The nanometre size of these pores, however, complicates their structural characterization by experimental techniques. Here we use molecular dynamics simulations, to study the interaction of a specific class of antimicrobial peptides, melittin, with a dipalmitoylphosphatidylcholine bilayer in atomic detail. We show that transmembrane pores spontaneously form above a critical peptide to lipid ratio. The lipid molecules bend inwards to form a toroidally shaped pore but with only one or two peptides lining the pore. This is in strong contrast to the traditional models of toroidal pores in which the peptides are assumed to adopt a transmembrane orientation. We find that peptide aggregation, either prior or after binding to the membrane surface, is a prerequisite to pore formation. The presence of a stable helical secondary structure of the peptide, however is not. Furthermore, results obtained with modified peptides point to the importance of electrostatic interactions in the poration process. Removing the charges of the basic amino-acid residues of melittin prevents pore formation. It was also found that in the absence of counter ions pores not only form more rapidly but lead to membrane rupture. The rupture process occurs via a novel recursive poration pathway, which we coin the Droste mechanism.

A Coarse-Grained Model for Polyethylene Oxide and Polyethylene Glycol: Conformation and Hydrodynamics

A coarse-grained (CG) model for polyethylene oxide (PEO) and polyethylene glycol (PEG) developed within the framework of the MARTINI CG force field (FF) using the distributions of bonds, angles, and dihedrals from the CHARMM all-atom FF is presented. Densities of neat low molecular weight PEO agree with experiment, and the radius of gyration R(g) = 19.1 A +/- 0.7 for 76-mers of PEO (M(w) approximately 3400), in excellent agreement with neutron scattering results for an equal sized PEG. Simulations of 9, 18, 27, 36, 44, 67, 76, 90, 112, 135, and 158-mers of the CG PEO (442 < M(w) < 6998) at low concentration in water show the experimentally observed transition from ideal chain to real chain behavior at 1600 < M(w) < 2000, in excellent agreement with the dependence of experimentally observed hydrodynamic radii of PEG. Hydrodynamic radii of PEO calculated from diffusion coefficients of the higher M(w) PEO also agree well with experiment. R(g) calculated from both all-atom and CG simulations of PEO76 at 21 and 148 mg/cm(3) are found to be nearly equal. This lack of concentration dependence implies that apparent R(g) from scattering experiments at high concentration should not be taken to be the chain dimension. Simulations of PEO grafted to a nonadsorbing surface yield a mushroom to brush transition that is well described by the Alexander-de Gennes formalism.

Combining an Elastic Network With a Coarse-Grained Molecular Force Field: Structure, Dynamics, and Intermolecular Recognition

Structure-based and physics-based coarse-grained molecular force fields have become attractive approaches to gain mechanistic insight into the function of large biomolecular assemblies. Here, we study how both approaches can be combined into a single representation, that we term ELNEDIN. In this representation an elastic network is used as a structural scaffold to describe and maintain the overall shape of a protein and a physics-based coarse-grained model (MARTINI-2.1) is used to describe both inter- and intramolecular interactions in the system. The results show that when used in molecular dynamics simulations ELNEDIN models can be built so that the resulting structural and dynamical properties of a protein, including its collective motions, are comparable to those obtained using atomistic protein models. We then evaluate the behavior of such models in (1) long, microsecond time-scale, simulations, (2) the modeling of very large macromolecular assemblies, a viral capsid, and (3) the study of a protein-protein association process, the reassembly of the ROP homodimer. The results for this series of tests indicate that ELNEDIN models allow microsecond time-scale molecular dynamics simulations to be carried out readily, that large biological entities such as the viral capsid of the cowpea mosaic virus can be stably modeled as assemblies of independent ELNEDIN models, and that ELNEDIN models show significant promise for modeling protein-protein association processes.

Molecular dynamics simulation of a phospholipid membrane.

We present the results of molecular dynamics (MD) simulations of a phospholipid membrane in water, including full atomic detail. The goal of the simulations was twofold: first we wanted to set up a simulation system which is able to reproduce experimental results and can serve as a model membrane in future simulations. This goal being reached it is then further possible to gain insight in to those properties that are experimentally more difficult to access. The system studied is dipalmitoylphosphatidylcholine/water, consisting of 5408 atoms. Using original force field parameters the membrane turned out to approach a gel-like state. With slight changes of the parameters, the system adopted a liquid-crystalline state. Separate 80 ps runs were performed on both the gel and liquid-crystalline systems. Comparison of MD results with reliable experimental data (bilayer repeat distance, surface area per lipid, tail order parameters, atom distributions) showed that our simulations, especially the one in the liquid-crystalline phase, can serve as a realistic model for a phospholipid membrane. Further analysis of the trajectories revealed valuable information on various properties. In the liquid-crystalline phase, the interface turns out to be quite diffuse, with water molecules penetrating into the bilayer to the position of the carbonyl groups. The 10–90% width of the interface turns out to be 1.3 nm and the width of the hydrocarbon interior 3.0 nm. The headgroup dipoles are oriented at a small angle with respect to the bilayer plane. The resulting charge distribution is almost completely cancelled by the water molecules. The electron density distribution shows a large dip in the middle of the membrane. In this part the tails are more flexible. The mean life time between dihedral transitions is 20 ps. The average number of gauche angles per tail is 3.5. The occurrence of kinks is not a significant feature.

Alternative Mechanisms for the Interaction of the Cell-Penetrating Peptides Penetratin and the TAT Peptide with Lipid Bilayers

Cell-penetrating peptides (CPPs) have recently attracted much interest due to their apparent ability to penetrate cell membranes in an energy-independent manner. Here molecular-dynamics simulation techniques were used to study the interaction of two CPPs: penetratin and the TAT peptide with 1,2-Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) and 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) phospolipid bilayers shed light on alternative mechanisms by which these peptides might cross biological membranes. In contrast to previous simulation studies of charged peptides interacting with lipid bilayers, no spontaneous formation of transmembrane pores was observed. Instead, the simulations suggest that the peptides may enter the cell by micropinocytosis, whereby the peptides induce curvature in the membrane, ultimately leading to the formation of small vesicles within the cell that encapsulate the peptides. Specifically, multiple peptides were observed to induce large deformations in the lipid bilayer that persisted throughout the timescale of the simulations (hundreds of nanoseconds). Pore formation could be induced in simulations in which an external potential was used to pull a single penetratin or TAT peptide into the membrane. With the use of umbrella-sampling techniques, the free energy of inserting a single penetratin peptide into a DPPC bilayer was estimated to be approximately 75 kJmol(-1), which suggests that the spontaneous penetration of single peptides would require a timescale of at least seconds to minutes. This work also illustrates the extent to which the results of such simulations can depend on the initial conditions, the extent of equilibration, the size of the system, and the conditions under which the simulations are performed. The implications of this with respect to the current systems and to simulations of membrane-peptide interactions in general are discussed.

Role of Lipids in Spheroidal High Density Lipoproteins

We study the structure and dynamics of spherical high density lipoprotein (HDL) particles through coarse-grained multi-microsecond molecular dynamics simulations. We simulate both a lipid droplet without the apolipoprotein A-I (apoA-I) and the full HDL particle including two apoA-I molecules surrounding the lipid compartment. The present models are the first ones among computational studies where the size and lipid composition of HDL are realistic, corresponding to human serum HDL. We focus on the role of lipids in HDL structure and dynamics. Particular attention is paid to the assembly of lipids and the influence of lipid-protein interactions on HDL properties. We find that the properties of lipids depend significantly on their location in the particle (core, intermediate region, surface). Unlike the hydrophobic core, the intermediate and surface regions are characterized by prominent conformational lipid order. Yet, not only the conformations but also the dynamics of lipids are found to be distinctly different in the different regions of HDL, highlighting the importance of dynamics in considering the functionalization of HDL. The structure of the lipid droplet close to the HDL-water interface is altered by the presence of apoA-Is, with most prominent changes being observed for cholesterol and polar lipids. For cholesterol, slow trafficking between the surface layer and the regimes underneath is observed. The lipid-protein interactions are strongest for cholesterol, in particular its interaction with hydrophobic residues of apoA-I. Our results reveal that not only hydrophobicity but also conformational entropy of the molecules are the driving forces in the formation of HDL structure. The results provide the first detailed structural model for HDL and its dynamics with and without apoA-I, and indicate how the interplay and competition between entropy and detailed interactions may be used in nanoparticle and drug design through self-assembly.

Solvent-Exposed Tails as Prestalk Transition States for Membrane Fusion at Low Hydration

Membrane fusion is a key step in intracellular trafficking and viral infection. The underlying molecular mechanism is poorly understood. We have used molecular dynamics simulations in conjunction with a coarse grained model to study early metastable and transition states during the fusion of two planar palmitoyl-oleoyl-phosphatidylcholine (POPC) bilayers separated by five waters per lipid in the cis leaflets at zero tension. This system mimics the contact area between two vesicles with large diameters compared to the membrane thickness at conditions where fusion may start in the core of the contact area. At elevated temperatures, the two proximal leaflets become connected via multiple lipid molecules and form a stalklike structure. At room temperature, this structure has a free energy of 3k(B)T and is separated from the unconnected state by a significant free energy barrier of 20k(B)T. Stalk formation is initiated by the establishment of a localized hydrophobic contact between the bilayers. This contact is either formed by two partially splayed lipids or a single fully splayed one leading to the formation of a (metastable) splayed lipid bond intermediate. These findings indicate that, for low hydration, early membrane fusion kinetics is not determined by the stalk energy but by the energy of prestalk transition states involving solvent-exposed lipid tails.

Sphere-to-Rod Transitions of Nonionic Surfactant Micelles in Aqueous Solution Modeled by Molecular Dynamics Simulations

Control of the size and agglomeration of micellar systems is important for pharmaceutical applications such as drug delivery. Although shape-related transitions in surfactant solutions are studied experimentally, their molecular mechanisms are still not well understood. In this study, we use coarse-grained molecular dynamics simulations to describe micellar assemblies of pentaethylene glycol monododecyl ether (C(12)E(5)) in aqueous solution at different concentrations. The obtained size and aggregation numbers of the aggregates formed are in very good agreement with the available experimental data. Importantly, increase of the concentration leads to a second critical micelle concentration where a transition to rod-like aggregates is observed. This transition is quantified in terms of shape anisotropy, together with a detailed structural analysis of the micelles as a function of aggregation number.

Concerted diffusion of lipids in raft-like membranes

Currently, there is no comprehensive model for the dynamics of cellular membranes. The understanding of even the basic dynamic processes, such as lateral diffusion of lipids, is still quite limited. Recent studies of one-component membrane systems have shown that instead of single-particle motions, the lateral diffusion is driven by a more complex, concerted mechanism for lipid diffusion (E. Falck et al., J. Am. Chem. Soc., 2008, 130, 44-45), where a lipid and its neighbors move in unison in terms of loosely defined clusters. In this work, we extend the previous study by considering the concerted lipid diffusion phenomena in many-component raft-like membranes. This nature of diffusion phenomena emerge in all the cases we have considered, including both atom-scale simulations of lateral diffusion within rafts and coarse-grained MARTINI simulations of diffusion in membranes characterized by coexistence of raft and non-raft domains. The data allows us to identify characteristic time scales for the concerted lipid motions, which turn out to range from hundreds of nanoseconds to several microseconds. Further, we characterize typical length scales associated with the correlated lipid diffusion patterns and find them to be about 10 nm, or even larger if weak correlations are taken into account. Finally, the concerted nature of lipid motions is also found in dissipative particle dynamics simulations of lipid membranes, clarifying the role of hydrodynamics (local momentum conservation) in membrane diffusion phenomena.