Insight Into the Stability of the Hydrophobic Binding Proteins of Escherichia coli: Assessing the Proteins for Use as Biosensors Branka Salopek-Sondi, Matthew C. Skeels, Derrick Swartz, and Linda A. Luck * Department of Chemistry, Clarkson University, Potsdam, New York ABSTRACT Spectroscopic methods were used to monitor the unfolding of the leucine specific (LS) and the leucine-isoleucine-valine (LIV) binding pro- teins. Our studies indicate that ligand-free protein undergoes a simple two-state unfolding, whereas the protein-ligand complex undergoes a three-state unfolding model. Ligand binding causes significant stabilization of both proteins. There is correlation between ligand hydrophobicity and protein stabili- zation: the most hydrophobic ligand, isoleucine, causes the most significant stabilization of LIV pro- tein. A disulfide bond present in N-domain of both proteins makes a large contribution to the protein stability of these periplasmic binding receptors. Proteins 2003;53:273–281. © 2003 Wiley-Liss, Inc. Key words: leucine-specific binding proteins; leucine- isoleucine-valine binding proteins; pro- teins stability INTRODUCTION The potential use of periplasmic binding proteins as biosensors has come to light because of their ability to reversibly bind a variety of small ligands such as sugars, amino acids, and inorganic ions. These proteins comprise a large family of functionally similar proteins with a hinge cleft binding mode for substrates. The soluble, monomeric proteins located in the space between the cell walls of gram-negative bacteria serve to function as an escort in the high-affinity active transport systems or they are initial receptors in chemosensory pathways. 1,2 The mecha- nism by which these proteins function in transport and signaling involves ligand binding in a pocket between the two domains and subsequent global conformational change. 3–6 This structural change allows recognition of the substrate-loaded receptor by the membrane compo- nents in either system. Key to these interactions is the stability of the receptor-ligand complex. Bound substrate lifetimes must be substantial to allow the receptor to diffuse long distances in the activated conformation in search of a docking site. Once docked to the membrane components, rapid release of ligand into the assembly expedites the addition of the nutrient to the cytoplasm. Hence, the stabilizing factors for the ligand-protein com- plex need to be tight enough to hold the complex together but flexible enough to release the substrate on demand. These two properties of the periplasmic binding proteins allow them to be extraordinary candidates for regenerat- able biosensors. Our laboratory is interested in the use of these periplas- mic binding proteins from E. coli, which have the capabil- ity of being used as biosensors for small substrates. We focus this article on the hydrophobic amino acid receptors, the leucine-specific receptor (LS) and the leucine-isoleucine- valine receptor (LIV) for several reasons. First, there is a curiosity for the reason why there are two proteins associ- ated with the transport system. They are 80% identical in amino acid content but have differing specificities for hydrophobic amino acids. These two proteins could provide a scaffold on which to mutagenize a binding pocket for a specific hydrophobic ligand. Second, our laboratory has found the leucine-specific protein is not so specific; it sequesters phenylalanine and other fluorinated com- pounds. 7,8 This increases the likelihood that the binding site may be altered to accommodate small molecules used in biological warfare. Hence, our biological system may be ideal for the development of biosensors. Third, these proteins have been used as models for a number of neuroreceptors including the metabotropic glutamate re- ceptor and the N-methyl-D-aspartate receptor. 9 –12 Insight into the function of the bacterial receptors will directly impact the study of Eukaryotic receptors for neurological signaling. In this article we address the stability of the two hydrophobic amino acid receptors from E. coli. With use of UV and intrinsic tryptophan fluorescence we have moni- tored the urea denaturation of each receptor with and without ligand. Using a two-state or three-state model we have obtained thermodynamic parameters of unfolding. The stabilizing function of the disulfide bond is discussed. Understanding the structure and functionality of the receptors are needed for the development of the biotechno- logical applications of these proteins. Grant sponsor: Petroleum Research Fund; Grant number: ACS- PRF#36825-AC4; Grant sponsor: National Institutes of Health; Grant number: R03-CA 89705-01; Grant sponsor: National Science Founda- tion: Grant number: DUE-9979509. Branka Salopek-Sondi’s present address is Rudjer Bos ˇkovic ´ Insti- tute, Zagreb, Croatia. *Correspondence to: Dr. Linda A. Luck, Department of Chemistry, Clarkson University, Potsdam, NY 13699. E-mail: luckla@clarkson.edu Received 12 February 2003; Accepted 11 April 2003 PROTEINS: Structure, Function, and Genetics 53:273–281 (2003) © 2003 WILEY-LISS, INC.