Analysing molecular polar surface descriptors to predict blood-brain barrier permeation.

Molecular polar surface (PS) descriptors are very useful parameters in prediction of drug transport properties. They could be also used to investigate the blood-brain barrier (BBB) permeation rate for various chemical compounds. In this study, a dataset of drugs (n = 19) from various pharmacological groups was studied to estimate their potential properties to permeate across the BBB. Experimental logBB data were available as steady-state distribution values of the in vivo rat model for these molecules. Including accurate calculation of the electrostatic potential maps, polar surface descriptors, such as a two-dimensional polar surface area (2D-PSA), topological polar surface area (TPSA) and three-dimensional polar surface area or polar area (3D-PSA; PA) were measured and analysed. We report the strong correlation of these descriptors with logBB values for the prediction of BBB permeation using the linear partial least squares (PLS) fitting technique. The 3D-PSA descriptor showed the best fit to logBB values with R² = 0.92 and RMSD = 0.29 (p-value < 0.0001). The obtained results demonstrate that all descriptors bear high predictive powers and could provide an efficient strategy to envisage the pharmacokinetic properties of chemical compounds to permeate across the BBB at an early stage of the drug development process.