Lipoproteins are large, heterogeneous assemblies, and the lipid component is highly disordered at biological temperatures. As a consequence, these complexes are impossible to study by x-ray crystallography and NMR. In the absence of direct high resolution structural information, only computer models will be able to provide the necessary level of detail for calculating both equilibrium and dynamic properties. The development of such models was begun during the previous grant period, and this proposal describes extensions of that research. In all the modeling proposed here, the peptides will be represented in atomic detail, but two different approaches will be used for modeling the lipid/solvent environment. The first approach uses molecular dynamics (MD) simulations, with explicit all-atom models for the lipid and solvent. This approach is the most rigorous and accurate method for modeling these systems, but it is computationally very demanding. The second approach uses continuum models for the lipid and solvent, and numerical methods are used to solve the Poisson-Boltzmann equation (the appropriate equation for describing electrostatic effects), allowing the determination of solvation free energies. This is a rapid, approximate method, facilitating the determination of the best starting geometries for the MD simulations. These methods will brought to bear on the study of amphipathic helices and beta strands interacting with planar bilayers and with discoidal and spherical HDL particles. The goals are to develop reliable models for peptide/lipid systems and for HDL particles, to assist in the design and interpretation of the experiments of projects 1,3, and 4, and, combining the theoretical and experimental results, to develop a comprehensive theory describing the interactions of amphipathic motifs with lipids.
Showing the most recent 10 out of 197 publications
