An award is made to the University of Michigan in Ann Arbor to develop a new computational method, associated web server, and a database for structural studies of transmembrane (TM) alpha-helical complexes. Motivated by the lack of systematic analysis of single-spanning (bitopic) proteins, the most abundant class of membrane proteins, we propose to create web resources for modeling alpha-helix association in membranes and to apply these resources to study bitopic proteins from six organisms representing all kingdoms of life. These proteins frequently self-associate via TM alpha-helices to perform their functions as receptors, channels, auxiliary transport proteins, enzymes, structure/adhesion proteins, or regulators. A proposed physics-based method will implement our anisotropic solvent model of the lipid bilayer, template-driven helix docking, and new interatomic potentials that reproduce free energy of helix association in membranes. The new web tool, TMDOCK, will calculate three-dimensional structures and stabilities of TM helices, their dimers, and symmetric homo-oligomers. The Membranome database (http://membranome.org) will organize the obtained models and provide structural classification, intracellular localization, topology, domain organization, and functional annotation of proteins. Analysis of TM helical oligomers will be performed to reveal evolutionary trends in self-association of helices in different functional categories of bitopic proteins.
The project will have a broad impact for researchers studying membrane-active peptides and proteins by providing the first easy-to-use web server that is practical for high-throughput modeling of helix complexes in membranes. The availability of this tool to the scientific community will accelerate structural and functional studies of membrane proteins, facilitate understanding of mechanisms of pore and channel formation by antimicrobial peptides and polypeptide toxins, enable modeling of bihelical TM proteins, identify folding intermediates of polytopic proteins, and assist in de-novo design of self-associating peptides with potential use in pharmacology and bio-nanotechnology. The Membranome database will be an important resource with multiple purposes ranging from education to cutting edge academic research. It will consolidate our knowledge on bitopic proteins and facilitate comparative analysis of these proteins. Rich graphics, interactive visualization tools and tangible molecular models based on protein structures from the database will provide a valuable educational resource.
