Projects

Glucocorticoid Receptor

Endocrine disruption: Estimation of the binding affinity of small molecules towards the Glucocorticoid receptor.

The glucocorticoid receptor is a member of the nuclear receptor superfamily that affects immune response, development, and metabolism in target tissues. Glucocorticoids are widely used to treat diverse pathological conditions but their clinical use is still limited by side effects. A prediction of the binding affinity towards the glucocorticoid receptor would therefore be beneficial for the design of new glucocorticoids with minimal adverse side effects, but also for identifying possible glucocorticoid-mediated toxic mechanisms triggered by drugs or chemicals. By combining flexible docking (software Yeti) to the glucocorticoid receptor and multi-dimensional QSAR (software Quasar), we validated a model based on 110 compounds, representing four different chemical classes. The good correlation with experimental data (cross-validated r² = 0.702, predictive r² = 0.719) suggests that our approach is suitable for predicting the binding affinity of novel compounds towards the glucocorticoid receptor. The model was challenged by a series of scramble tests, a consensus approach using the Raptor software as well as the prediction of eight external compounds.

Binding of R-Lorazepam to the glucocorticoid receptor (stereo view).

Our QSAR study was based on a total of 110 compounds. 88 molecules thereof were assigned to the training set and the remaining 22 used as test compounds. The model for the glucocorticoid receptor converged at a cross-validated r² of 0.702 and yielded a predictive r² of 0.719 for the test compounds.

Comparison of experimental and predicted Ki values for the glucocorticoid receptor
(training set = black, test set = red).

Reference: ChemMedChem 2009, 4, 100–109. View abstract