Global (Q)SAR models on substrates for human Cytochrome P450 3A4

The Cytochrome P450 (CYP) is a superfamily of enzymes which catalyze the metabolism of a wide range of endobiotics and xenobiotics. The latter category comprises drugs and about 75% of marketed drugs are metabolised by CYP enzymes. Besides drugs, CYP enzymes detoxify environmental compounds but paradoxically they also have the ability to form reactive intermediates which can damage DNA, lipids and proteins. It is therefore important to gain knowledge on which substrates that can potentially be metabolised by CYP. The CYP 3A4 isoenzyme plays a dominant role by the metabolic elimination of up to 35% of prescribed drugs. Literature on substrates and non-substrates primarily based on in vivo human data on the CYP 3A4 isoenzyme was collected and a training set of 863 chemicals was used to create (Q)SAR models. The modeling systems used were MultiCASE, Leadscope and MDL QSAR and the developed models cross-validated (Leave-groups-out) with concordances of 61%, 72% and 63%, respectively. 40,375 discrete organic EINECS chemicals (European INventory of Existing Commercial chemical Substances) were screened to predict an approximate percentage of CYP 3A4 substrates and the percentage of EINECS chemicals in the model domain. Domain coverage of EINECS chemicals and number of predicted substrates are discussed. Reference: C.W. Yap and Y.Z. Chen, Prediction of cytochrome p450 3A4, 2D6, and 2C9 inhibitors and substrates by using support vector machines, J. Chem. Inf. Model. 45 (2005), pp. 982–992.