Multidrug-efflux membrane transporters, which are expressed in all cells, present one of the most puzzling and long standing problems of biochemistry. These proteins recognize dozens of structurally dissimilar toxic organic molecules, mostly hydrophobic cations, and extrude them from the cell to the outer medium. The phenomenon of multidrug recognition seemingly contradicts the existing views on the mechanism of enzyme-substrate interaction and remains unexplained. In this project the mechanism of multidrug recognition will be analyzed in a different, yet somewhat related protein, BmrR. BmrR is a transcriptional regulator of the Bacillus subtilis multidrug transporter Bmr. Upon binding a wide variety of structurally diverse hydrophobic cations, it activates the expression of Bmr, which in turn promotes their efflux from the cell and thus protects the cell from their toxic effects. The inducer-binding activity of BmrR is confined to its 159 residue-long C-terminal domain. This domain, designated BRC, has been individually expressed, crystallized and its structure was solved. Additionally, the structure of the complex of BRC with one of the inducers, tetraphenylphosphonium (TPP), has been solved. BRC forms a dimer each subunit of which contains an electronegative glutamate in the middle of the hydrophobic core of the protein. Substitution of electroneutral glutamine or alanine for this glutamate does not affect protein folding but completely abolishes the binding of the inducers. Furthermore, in the structure of the complex, TPP forms an electrostatic contact with this glutamate and multiple interactions with the surrounding hydrophobic residues. In order to allow TPP to gain access to the core of the protein, the surface amphiphilic a-helix completely unwinds. Therefore, the key to the bewildering inducer promiscuity of BmrR may lie in its flexibility, which allows each of the different inducers to penetrate into the protein and interact with the glutamate and the surrounding hydrophobic residues.
This hypothesis will be tested by using three approaches. First, attempts will be made to crystallize and solve the structures of several additional BRC-inducer complexes. Second, by using fluorescence measurements, the conformational changes occurring in the BRC molecule upon inducer binding will be assessed. Finally, mutational analysis will be used to verify whether different inducers interact with different sets of residues of the protein. If the hypothesis described above is proven largely correct, this will indicate the existence of a new principle of ligand recognition by proteins, which may be described as a "flexible induced fit", in which a protein achieves tight binding of structurally diverse ligands by acquiring a conformation dictated by the ligand itself. This principle is likely to be used in many situations requiring multi-ligand recognition, including the recognition of multiple drugs by multidrug-efflux transporters.
This project will investigate the question of how an enzyme can recognize structurally dissimilar substrates. The structures of a variety of substrates complexed with a bacterial protein that regulates gene expression will be determined. Since most enzymes recognize only a very limited set of substrates, this is a fundamentally important problem addressing the relationship between protein structure and function. Moreover, since multidrug resistance results when a diverse class of compounds are recognized and pumped out of the cell, this work has the potential of increasing our understanding of this clinically important process.
