Computationally Intensive Research Project
First Principles Multiscale Analysis of Biochemical Processes: Signal Transduction and Spectroscopic Analysis of Local Structure-Folding Relations in Membrane Proteins
MSC News
Additional Information
John H. Weare,1 Joe Adams,1 Scott B. Baden,1 Eric J. Bylaska,2 Phillip A. Cole,3 Weinan E,4 Judy E. Kim,1 Karol Kowalski,2 Peter W. Langhoff,1 Andrew McCammon,1 John H. Miller,5 Julie C. Mitchell,6 Tjerk P. Straatsma,2 Susan S. Taylor,1 Wei Wang,1 Marat Valiev2
1University of California, San Diego, 2Pacific Northwest National Laboratory, 3Johns Hopkins University, 4Princeton University, 5Washington State University Tri-Cities, 6University of Wisconsin-Madison
FY07 Allocation - 100,000
Abstract
We propose computational work in two application areas: first principles simulation of posttranslation modification in signal transduction in eukaryote cells and the use of spectroscopic methods to probe the mechanisms of protein folding in membrane proteins. Substantial effort will also be devoted to the necessary development of the multiscale high performance computational tools to address problems encountered in the course of our project. The proposed computational work will be carried out in close collaboration with experimental groups (Adams, Cole, and Taylor, signal transduction program; Kim, protein folding work).
The most important posttranslational modification affecting cellular function is the phosphorylation of residues in target proteins carried out by the kinase enzymes. The structure-function relationship among these enzymes has been an active topic of research for many years. The structure of their active site is highly conserved implying that the mechanism for phosphorylation is similar across the family. Despite intense experimental work the nature of the reaction process and the role of the conserved elements in the active site are not well understood. These issues will be addressed in this projects using recently developed QM/MM module(1) in the NWChem computational chemistry package to simulate the phosphorylation reactions. The atomic level accuracy provided by QM/MM methods can provide much needed interpretation for the considerable amount of structural, kinetic, chemical modification and mutagenesis data accumulated for these systems. The opposing process of dephosphorylation is carried out by phosphatases. Analysis of biochemical pathways with temporal and spatial resolution has become a major objective of "systems biology". Ab-initio methods are needed to supplement the sparse experimental data on rate constants to enable kinetic simulations that involve the actual molecular species of signaling networks. As part of this program we will investigate the modulation of cellular signaling through the interaction of phosphatases with reactive oxygen species (ROS); a system well-suited to this approach because (1) the fundamental reaction, oxidation of a catalytic cysteine residue, can be examined by QM/MM methods and (2) kinetic signaling data are available to test simulations results based on calculated rate parameters.
Another part of our project involves a combined experimental and computational investigation of membrane protein folding via fluorescence spectroscopy. Membrane proteins constitute around 30% of all cell proteins and are major targets of biological research. They facilitate the transfer of materials and information between cells and their environment and are essential to the integrity of biological interfaces. Despite their significance, the fundamental assembly mechanism of these ubiquitous biomolecules remains a mystery. Our program will focus on one of the best characterized membrane protein folding reactions, that of Outer Membrane Protein A (OmpA). This bacterial integral protein is the most abundant protein of the outer membrane of E. Coli with proposed functions that include acting as a nonspecific pore and receptor, and providing structural integrity to the biological membrane. In addition, OmpA belongs to a large family of -barrel membrane proteins and hence, elucidation of the assembly?pore-forming mechanism of OmpA may shed light on the important porin proteins implicated in transport of toxic substances during bioremediation. OmpA is an ideal model system; it is one of the few membrane proteins that reversibly folds from the fully denatured state. High resolution x-ray and NMR structures are available identifying the residues and their environments in the protein. Because the protein folding can be reversibly controlled and the fluorescence is sensitive to the local folding environment we will obtain direct spectroscopic information about the local structure in various folded environments. Our computational approach is unique in providing the highly accurate finite temperature excited state properties of chromophores in large proteins that are required for the analysis of the fluorescence data. This new theoretical tool is based on the integration of coupled cluster methods with molecular dynamics simulations in the NWChem program.
In order make progress in these research areas our development of new multiscale computational tools for application to problems requiring high performance computing will have to continue. We believe that are entering a new era of discovery in which the information from necessarily high level experimental observations can be greatly expanded by appropriate multiscale theoretical interpretation beginning at the atomic first principles level. The applications proposed will also drive the development of new methods that will be included in NWChem software package. The availability of such tools benefits other areas important to the mission of the Department of Energy including biogeochemistry, subsurface science, waste isolation, interfacial chemistry and catalysis.
