Novel method for analyzing spectroscopic ligand binding data

CHED 911

Adam G. Georgas, georgaag@rose-hulman.edu and Mark E. Brandt, brandt@rose-hulman.edu. Department of Chemistry, Rose-Hulman Institute of Technology, 5500 Wabash Ave., CM2262, Terre Haute, IN 47803
The estrogen receptor protein plays a key role in the development of breast cancer. In order to learn more about the disease and its treatment, the effect of ligands, such as tamoxifen, on the receptor protein must be understood. By measuring quenching of intrinsic tryptophan fluorescence due to resonance energy transfer, the amount of ligand bound to the receptor can be determined. Attempts at using least squares non-linear regression techniques for standard equations of ligand binding yielded fitting parameters that were highly dependent on the initial values chosen. Furthermore, this method frequently resulted in physically impossible fitting parameters. By employing a highly iterative least squares optimization method written in Matlab, unique and physically possible solutions for the Hill coefficient and K0.5 fitting parameters were obtained. Analysis using this method showed that the human estrogen receptor exhibited positively cooperative binding to tamoxifen. The Hill coefficient observed was 1.41 ± 0.07 and the K0.5 was 0.024 ± 0.006 μM. The method described can be used to analyze data from any receptor binding assay in which only the concentration of bound ligand can be measured directly.