Towards a multiscale model of P-glycoprotein efflux

The blood-brain barrier (BBB) remains a major roadblock to the delivery of drugs to the brain. While in vitro and in vivo measurements of permeability are widely used to predict brain penetration, very little is known about the mechanisms of passive and active transport. Detailed insight into interactions between solutes and cell membranes could provide new insight into drug design and screening. Multidrug resistance (MDR) is a problem that affects treatment of many diseases, and is often associated with P-glycoprotein (P-gp, ABCB1 or MDR1), one of the ABC family of transporters. These efflux pumps transport substrates that have crossed the blood-brain barrier back out into circulation, thereby limiting bioavailability in the local tissue. The mechanism of how P-gp is inhibited is not well understood, in particular how the ligand reaches its binding site. Here, we build a multiscale model of transport across the BBB. Firstly, we demonstrate our BBB membrane model of the brain microvascular endothelial cell (hBMEC) membrane. We perform simulate unbiased atomistic MD simulations to visualize translocation of a library of 23 solutes across a lipid bilayer representative of brain microvascular endothelial cells (hBMEC). Secondly, we perform simulations of the active transport of P-glycoprotein (P-gp) efflux out of the cell for a substrate and a potential third-generation inhibitor. Then, we demonstrate results using a novel coarse-grain model of P-gp with MARTINI 3.0, which enables us to access much longer timescales.