A Method of Curve Fitting Analysis for Cooperative-type Scatchard Plots and Its Application to Dexamethasone Binding to Hepatic Cytosol Receptors of Mice, Strains A/J, C57 B10, and B10A

Abstract
A method of regression analysis for Scatchard plots of the cooperative type (1–15) was derived from Deming's method (16) which is applicable to implicit functions, f(X, Y), in which both variables, X and Y, are subject to error. Deming's method involves the L coefficient, (f/X)2/Wx+(f/Y)2Wv whose value varies with the unit of concentration. When the observations fit the theoretical curve well, the variation with the concentration unit is negligible, whereas the variation is appreciable when the observations fit badly. In order to standardize the effect of the unit, we introduced a hypothetical unit of concentration, which makes the mean values of the observations, X and F, identical. Assuming two binding sites in one receptor protein, we applied the method to dexamethasone binding to the hepatic cytosol receptor protein of mice, strains A/J, C57 BIO and B10A. The Kd values of the first binding site of A/J and B10A mice were calculated to be significantly smaller than that of BIO mice. However, the Kd value of the second site of A/J mice was significantly larger than the corresponding values of BIO and B10A mice, and the value of B10A was larger than that of BIO mice. The differences in the Kd values may explain the genetic differences in the susceptibility to glucocorticoid-induced cleft palate among the three strains, since the affinity of the first binding is higher in the A/J and B10A strains, which have higher susceptibility than BIO.