The analysis of tight protein-ligand binding reactions by isothermal titration calorimetry (ITC) and thermal shift assay (TSA) is presented. The binding of radicicol to the N-terminal domain of human heat shock protein 90 (Hsp90aN) and the binding of ethoxzolamide to human carbonic anhydrase (hCAII) were too strong to be measured accurately by direct ITC titration and therefore were measured by displacement ITC and by observing the temperature-denaturation transitions of ligand-free and ligand-bound protein. Stabilization of both proteins by their ligands was profound, increasing the melting temperature by more than 10 ºC, depending on ligand concentration. Analysis of the melting temperature dependence on the protein and ligand concentrations yielded dissociation constants equal to 1 nM and 2 nM for Hsp90aN-radicicol and hCAII-ethoxzolamide, respectively. The ligand-free and ligand-bound protein fractions melt separately, and two melting transitions are observed. This phenomenon is especially pronounced when the ligand concentration is equal to about half the protein concentration. The analysis compares ITC and TSA data, accounts for two transitions and yields the ligand binding constant and the parameters of protein stability, including the Gibbs free energy and the enthalpy of unfolding.

Protein thermal denaturation by following fluorescence of a reporter probe has developed into an important method to measure protein-ligand binding and characterize protein stability. The assay, usually termed, the thermal shift assay, the differential scanning fluorimetry, or ThermoFluor®, determines protein melting temperature in the presence of various ligands, inhibitors, excipients, buffers, salts, or osmolytes. The compound dosing curves enable determination of the protein denaturation thermodynamic parameters and the ligand binding constant at various temperatures, including physiological temperature. The ThermoFluor method, when the thermal shift assay is carried out in 384 well plates, is used in high-throughput screening of drug leads in the pharmaceutical industry. The general methodology and several recent developments of the method are presented in this review.