We have developed high-throughput mass spectrometric methods for mapping the structure and the location of active sites of proteins. These methods rely on measuring changes in solvent-accessibility and stability in the presence and absence of a small or large molecule ligands, as determined from the exchange rates of solvent deuterons with amide NH and sidechain CH groups of the protein; collectively we term this hydrogen-deuterium-exchange LC-MS, or H/D-LC-MS. Although our methods are being extensively developed to assess the binding of drugs to water-soluble proteins, they have not yet been applied to membrane proteins. Here we propose to extend our technology to study a relatively simple, but medicinally important membrane protein, the proton channel from influenza A virus (M2). This M2 proton channel is essential to the infectivity of the virus, and is the target of the anti-influenza drugs, amantadine and rimantadine. We have three specific aims in this proposal: 1) Using M2 as a model, we will extend H/D-LCMS to characterize the structures and binding sites of membrane proteins. 2) The methods will then be used to map the solvent accessibility of amides and sidechains within the M2 proton channel. Specifically we will determine the extent to which we can map the structured water-soluble regions of the protein, the membrane-contacting regions, and the solvent-filled channel of the M2 protein. 3) We will determine the feasibility of using these methods to assess structural changes associated with ligand binding and alterations in pH; the M2 proton channel is activated at the low pH (near 5.5) where it functions. We therefore will examine whether our technology can be used to evaluate differences in its H/D exchange at pH 7 versus 5, in the presence and absence of amantadine. The results will have an impact on our knowledge of this important therapeutic target, and will have broad impact on the future analysis of membrane-protein structures. With methods in place, we will be able to define structure perturbations of M2 upon binding other amantidine-related ligands and new pharmacophores. One long-term goal will be to aid the discovery of new antiviral agents. In addition, the methods developed herein will fill the gaps in the ability to explore structures of membrane proteins. We hope to develop H/D exchange into platform for the study of important membrane-associated drug targets, such as G-protein coupled receptors (GPCRs).
