Integrating ligand binding kinetics into screening using the IDBS Biobook™ Application

David Winpenny, Joanna Mulgrew, Paul Driscoll, Phillip Stanley. Pfizer Global Research and Development, Sandwich, Kent, United Kingdom.

Most in vitro assays performed during early stage drug discovery assume that the system is at equilibrium when measurements are taken. In many cases this is incorrect and can lead to errors in estimations of affinity and misinterpretation of observed differences in responses between compounds. Kinetic binding assays can mitigate some of these problems as the analysis makes no assumption of equilibrium. However, the integration of binding kinetics within an SAR generating programme requires a standardised workflow to cover compound ordering, processing, screening, analysis and reporting.

Data analysis and reporting is the biggest bottle neck in the process and to deal with this a bespoke analysis application was developed within IDBS E-Workbook package BioBook™. After data input the application models the kinetics of the tracer and subsequently the kinetics of the unlabelled test compounds, all fitted using global models with shared parameters. These parameters are reported automatically into corporate databases upon completion of the experiment. The time points, concentrations tested and plate layout are flexible and easily altered to suit user needs.

In order to test the process and application a 96 well kinetic filter binding assay was established for the muscarinic M3 receptor using 3H N-Methyl Scopolamine (NMS) as the tracer. For each time point three concentrations of tracer were used one of which was also tested in the presence of three concentrations of each test compound. In 20 mM HEPES pH 7.4 at room temperature. A range of compounds including four standards were tested and their kinetic parameters estimated.

pKDs were consistent with literature values (Dowling & Charlton 2006). For slow compounds, Koff <0.005, fitting failed if the concentrations and time course were insufficient to estimate equilibrium. This highlights the importance of choosing the correct concentrations and time course. The application has proven to be highly extensible and we are developing assays using repeated measurements from 96 and 384 well plates which will improve data quality and throughput, allowing the potential replacement of IC50 determination by kinetic KD estimation for certain assays.

Dowling, M. and Charlton, S. (2006). Br. J. Pharmacol. 148:927–937.