This project focuses on the study of membranes, proteins and carbohydrates by molecular dynamics computer simulation. Progress is reported under each Aim listed above Aim 1. Understand Model Membranes. A unified treatment of curvature of membranes has been developed wherein a lipid mixture is simulated both in the inverse hexagonal phases (where experimental data are directly available) and the lamellar phase (as present in cell membranes). Simulations on POPC and POPE demonstrated the that bending free energy is the same for the two phases of each lipids, differences between the two lipids, and confirmed experimental assumptions regarding the location of the pivotal plane (ref 4).
Aim 2. Develop Simulation Methodology. The correct treatment of the inverse hex phase noted in Aim 1 required development of a new algorithm for evaluating the pressure in the interior of a curved interface. A paper describing the theory and tests on simple systems (pure water and a water tube in octane) was published in the Journal of Chemical Physics (ref 3). Adjustments in the Lennard-Jones interaction parameters between sodium ions and different oxygens on charged lipids were developed. Assessing the impact of the new parameters on neutral lipids required the development of a method to determine the effective charge on the bilayer, analogous to electrophoresis. Simulations of phosphatidylserine and phosphatidylglycerol bilayers show excellent agreement with experiment, and have expanded the reach of membrane simulations to charged bilayers (ref 5).
Aim 3. Simulate Complex Membranes Three papers related to this Aim were published, each covering a different topic. Continuum elastic models of bilayers with gramicidin A (gA) were shown to break down in the first shell of lipids surrounding the peptide, highlighting the necessity of molecular dynamics simulations to describe this region (ref 2). All-atom microsecond simulations of transmembrane helix dimers ErbB1/B2 and EphA1 in lipid bilayers show good agreement which experimentally derived helix tilt, crossing angles, and dimer contacts, although the detailed contact surface remains offset for one of two helices in both systems. This results indicate that both implicit membrane models require improvement, and that even microsecond simulations are not sufficient to anneal incorrect starting structures. Nevertheless, the alternate structures can be rationalized with reference to interaction motifs and may represent still sought after receptor states that are important in ErbB1/B2 and EphA1 signaling (ref 6). Molecular dynamics simulations were used to refine a theoretical model that describes the interaction of single polyethylene glycol (PEG) molecules with -hemolysin (HL) nanopores. The simulations yielded excellent agreement with experimental ion conductivities and current blockage by a 29-mer PEG, and indicated that on average 1.5 K+ bind to this polymer, not 5 as previously assumed. This work is part of an overall effort with NIST to nanopore sequencing methods for DNA (ref 1).
Showing the most recent 10 out of 52 publications
