Thermodynamic and computational studies on the binding of p53-derived peptides and peptidomimetic inhibitors to HDM2 Abstract Helix power: The binding interactions of linear and constrained beta-hairpin-shaped peptides with HDM2 were compared by using experimental and theoretical methods. The entropic advantages enjoyed by the constrained peptides were found to be largely offset by reduced enthalpic contributions to binding of the cyclic mimetics. Formation of hydrogen bonds upon helix folding could contribute significantly to the enhanced enthalpy observed in binding of the linear peptides.The human double minute 2 protein (HDM2) binds a short peptide derived from the N terminus of the tumor-suppressor protein, p53. This peptide (p53 residues 15-29) is flexible in free solution, but upon binding to HDM2 it folds into an amphipathic alpha-helical conformation. Three residues along one face of the p53 helix (Phe19, Trp23, and Leu26) dock into hydrophobic pockets on the surface of HDM2. A conformationally constrained cyclic beta-hairpin peptidomimetic of p53, with residues Phe1, 6-chloro-Trp3, and Leu4 in one strand of the beta-hairpin, was shown earlier to dock into the same pockets on HDM2. Here, we show by isothermal titration calorimetry that the entropy loss upon binding of the constrained peptide to HDM2 is, as would be expected, much lower (TDeltaS approximately 10 kcal mol(-1) at 300 K) than that for the linear peptide. However, the entropic advantage enjoyed by the constrained peptide is largely offset by a reduced enthalpic contribution, relative to the linear peptide, to binding of the cyclic mimetic. To explore the electronic nature of the interactions between the energetically important residues in each ligand and HDM2, hybrid quantum mechanical and electrostatic Poisson-Boltzmann computational studies were performed. The calculations reveal that significant stabilizing van der Waals interactions and polarization effects occur between the Trp side chain in each ligand and aromatic and aliphatic residues in HDM2. These stabilizing interactions are enhanced when a 6-chloro substituent is incorporated into the Trp, in agreement with the experimental studies. In addition, the calculations suggest that at least one stabilizing hydrogen bond is formed, between the Trp indole-NH in both ligands and HDM2. Other hydrogen-bonding interactions also arise, however, along the alpha-helical backbone of the linear peptide upon binding to HDM2, but are not mimicked in the constrained inhibitor-HDM2 complex. The formation of these hydrogen bonds upon helix folding could contribute significantly to the enhanced enthalpy observed in binding of the linear peptide to HDM2.

Mass spectrometry, and especially electrospray ionization, is now an efficient tool to study noncovalent interactions between proteins and inhibitors. It is used here to study the interaction of some weak inhibitors with the NCoA-1/STAT6 protein with K D values in the M range. High signal intensities corresponding to some nonspecific electrostatic interactions between NCoA-1 and the oppositely charged inhibitors were observed by nanoelectrospray mass spectrometry, due to the use of high ligand concentrations. Diverse strategies have already been developed to deal with nonspecific interactions, such as controlled dissociation in the gas phase, mathematical modeling, or the use of a reference protein to monitor the appearance of nonspecific complexes. We demonstrate here that this last methodology, validated only in the case of neutral sugar-protein interactions, i.e., where dipole-dipole interactions are crucial, is not relevant in the case of strong electrostatic interactions. Thus, we developed a novel strategy based on half-maximal inhibitory concentration (IC 50 ) measurements in a competitive assay with readout by nanoelectrospray mass spectrometry. T here is a strong interest in the study of noncovalent complexes between biomolecules, which are playing key roles in life. Numerous solutionphase analytical techniques were developed to determine the specificity and the strength of these types of interactions [1]. Mass spectrometry (MS), and especially electrospray ionization (ESI) [2], has become an efficient tool to study specific noncovalent complexes between various species (protein-protein, protein-small molecules, protein-DNA, DNA-DNA . . .) [3][4][5][6][7][8]. In fact, ESI is a very soft ionization technique, i.e., noncovalent complexes can be transferred intact from solution into the gas phase. Quantitative information such as stoichiometry, binding constants, or reaction kinetics can be obtained by ESI-MS, and values are often in good agreement with data coming from well-established solution phase techniques. Nevertheless, the study of noncovalent [protein-ligand] complexes require careful control of experimental parameters. Buffer, pH, pressure, and voltages applied to the different stages of the mass spectrometer have great influence on spectral characteristics and on the information gained. Moreover, electrochemical reactions and desolvation/ionization mechanisms involved in ESI can also complicate the analysis, thus giving rise to the so-called nonspecific interactions (i.e., interactions with nonspecific binding sites) that alter the solution phase stoichiometry. To study weak complexes with dissociation constants (K D ) in the M range or higher in solution, high ligand concentrations are employed, leading to an increase of nonspecific complex ions signals and to underestimate K D values, which might not reflect the solution-phase equilibria anymore [5, 9 -16].Three strategies have been developed to determine affinities of weak [protein-ligand] complexes by ESI-MS, even when nonspecific gas-pha...