Improved curve fitting procedures to determine equilibrium binding constants

For ligand–biomacromolecule titration experiments it has been traditional practice to extract parameters such as the equilibrium binding constant K and the number of bases per ligand binding site n with relatively labour intensive methods, usually based on single wavelength data, such as the difference method by Rodger and Nordén coupled together with a Scatchard plot. Presented in this paper are both the theory and a least squares fitting method to derive parameters such as K and n more directly from all spectral non-linear experimental data. Both the case of non competitive binding of a metal complex ligand to DNA and the case of displacement by a metal complex ligand of an ethidium marker attached to the DNA are considered. This work may be applied directly to reduce experimental data produced by a spectropolarimeter (for circular or linear dichroism) or a spectrophotometer (for fluorescence or UV-Vis spectroscopy).