Our understanding of the diverse biological roles played by membrane proteins lags far behind our knowledge of soluble proteins, which are simpler to express, purify, and crystallize. The study proposed here is a systematic exploration of the extent to which the hydrophobicity of the surface amino acid residues determines whether a 2-barrel protein will remain soluble or insert into a membrane. We will perform a statistical analysis of the surface amino acids of known 2-barrel membrane protein structures, from which we will develop a computational method to evaluate the energetic cost of inserting a 2-barrel protein into a membrane. We will apply this method to redesign the surface of the E. coli outer membrane protein OmpA with "skins" of different membrane-insertion propensities. Expression and evaluation of the membrane incorporation of each of these designs will provide data to optimize our computational energy function. Using our validated system, we will undertake the computational redesign of the soluble 2-barrel protein GFP into a transmembrane protein. Designs will be experimentally tested for membrane insertion and fluorescence. Directed evolution and high-throughput screening by fluorescence-activated cell sorting will be used to reverse any loss of fluorescence in the transmembrane GFP. Success will represent a major achievement in protein design and engineering and provide a powerful tool for the study of membrane-associated processes and membrane protein folding.
Membrane proteins are among the most common drug targets, yet our understanding of their properties is limited. Our computational and experimental studies will provide insights on the defining characteristics of membrane proteins and provide new tools for the study of membrane-associated processes.