THERMODYNAMIC MODEL FOR LIGAND-INDUCED BIPHASIC PROTEIN DENATURATION

根据实验数据（Vox Sang. 47, 7-18(1984)）和J. Biol.；

On the basis of experimental data (Vox Sang. 47, 7-18 (1984) and J. Biol. Chem 263, 15392-15399 (1988)], which derive from thermally induced denaturation studies of human albumin, we concluded that ligand-induced biphasic denaturation of the protein, which does not relate to protein domain substructure can occur if the affinity of the protein for the ligand is great enough and if initially the protein is subsaturated with ligand. Approximate calculations indicated that such a description can account for these observations (Abstracts of the 42nd Calorimetry Conference (1987)]. We have developed a rigorous thermodynamic model, which incorporates the effects of ligand binding on the transition temperature for the unfolding of the protein and have demonstrated that a computation for a protein with a single binding site with three different values for the affinity constant can reproduce the salient features of the observed denaturation of albumin in the presence of low, intermediate, and high affinity ligands [J. Biol. Chem. 265, 50555059 (1990)). Currently we are carrying out more detailed calculations for albumin, utilizing binding data in the literature, in order to demonstrate that the thermodynamic model results in computed denaturation profiles that closely mimic those observed experimentally. In addition, we plan to illustrate that with ligand-induced biphasic denaturation, contributions to the denaturation profile by the different native protein species are in general not of a two-state nature thereby indicating that standard deconvolution procedures would result in a basis set of constitutive endotherms devoid of physical meaning. Furthermore, a theoretical investigation of the origins of bimodality demonstrates that multiphasic (i.e., triphasic etc.) denaturation can occur and reveals the factors responsible for this phenomenon.