Experimental design and estimation of parameters in complex radioligand binding systems.

A computer was used to simulate data from typical radioligand binding experiments with a 2-site competitive-allosteric model of a receptor. The 4 equilibrium parameters of this model cannot be estimated by fitting the model to equilibrium data. Data from simulations of association experiments give satisfactory estimates of the 9 competitive-allosteric model parameters. From the kinetic parameters, equilibrium constants may be calculated. Combining data from equilibrium simulations with data from association simulations provided estimates of the model parameters with smaller standard deviations. A further improvement in design was shown possible by including simulated experiments in which receptor was preincubated with inhibitor before adding ligand. This improvement was documented using Monte Carlo replications of parameter estimates using competing experimental designs. Replications also revealed certain biases in the parameter estimates and could provide a means of estimating those biases when parameter estimates are made using experimental rather than simulated data. Simulations offer a powerful tool in planning experiments designed to estimate kinetic parameters of a receptor system. This is especially true with complex systems that may require pooling data from different kinds of experiments in order to estimate the kinetic parameters.