Biochemistry " Footprint " titrations yield valid thermodynamic isotherms ( DNase protection /

A central issue in gene regulation is the mechanism, and biological function, of the cooperative binding of regulatory protein ligands to specific sites on DNA. To elucidate the physical-chemical basis of these interactions we have developed a thermodynamically rigorous method for conducting DNase I "footprint" (protection) titration experiments. The intrinsic binding constants and also those for cooperative interactions between various sites can be resolved from the individual-site binding curves determined by this technique. Experimental studies of cl-repressor-operator binding have demonstrated that the method provides an accurate representation of the fractional saturation of a binding site. We present individual-site binding curves for a X operator with two competent sites that demonstrate the presence of cooperative interactions between the sites. These curves set a lower limit to the magnitude of the cooperative free energy without comparison to single-site mutant operators. The regulation of many prokaryotic and eukaryotic genes is dependent on specific, and often cooperative, interactions that control the binding of regulatory proteins to multiple sites on DNA. To identify the sites of interaction numerous investigators are currently using various forms of "footprinting" or "protection mapping" using DNase I or other cleavage reagents (1-4). As a titration method, footprinting has also been used to estimate the affinity of proteins for DNA (5, 8). Because many protein-DNA interactions occur with high affinity, necessitating nanomolar ligand concentrations, it is essential to demonstrate that systematic errors are sufficiently minimized so that thermodynamically valid binding curves can be obtained. Most techniques used to study DNA-protein binding measure only average binding properties over all of the sites, so that unique resolution of cooperative binding constants is impossible. Moreover, the most popular technique, nitrocellulose filter binding, does not correctly resolve even the macroscopic (average) binding constants for multisite systems (6). We have developed a footprint titration method that resolves thermodynamically rigorous equilibrium binding curves. Because footprint titration experiments distinguish binding at the individual sites, individual-site binding curves, which separately represent fractional saturation at each site, result. These can be used to resolve not only the intrinsic binding constants but also those for cooperative interactions between the various sites (6-9). This information is required to understand and predict (i.e., model) the biological roles of these interactions (8, 10). To ensure the thermodynamic validity of the footprint titration method we have developed (i) protocols for conducting the experiment and quantitating the data and (ii) a rigorous theory ofindividual-site binding that is necessary for analysis and correct interpretation of the resolved binding curves (7). Here we show that when these procedures are used the technique correctly measures the equilibrium distribution of occupied and vacant sites on the DNA. We will outline the experimental guidelines by which thermodynamically valid binding curves can be obtained. The footprint titration method is broadly applicable and can potentially be used to study control systems with different numbers of sites and control proteins. A more detailed description of the methods has been published elsewhere (9).