Simulating Transport Across the Blood–Brain Barrier Endothelial Membrane: Comparing Unbiased Molecular Dynamics Methods to Enhanced Sampling Protocols

The blood–brain barrier (BBB) is a highly selective membrane that restricts the delivery of therapeutics into the brain. In this review, we consider a number of computational techniques that can be used to assess BBB transport and central nervous system (CNS) delivery. We outline the details of human brain microvascular endothelial cells (hBMECs) and how passive transport is the main mode of transport. We then outline details of molecular dynamics techniques for predicting in silico drug permeability across an atomistic hBMEC model. Due to the very slow timescales required to sample CNS transport a priori, we compare transition-based counting (TBC) methodology to enhanced sampling techniques. These advanced methods overcome traditional sampling limitations and include umbrella sampling (US) with the Weighted Histogram Analysis Method (WHAM) and the Dynamic Histogram Analysis Method (DHAM). Additionally, we consider constant velocity-steered molecular dynamics (cv-SMD)/US combined with the Inhomogeneous Solubility–Diffusion (ISD) model, as well as Metadynamics and the Adaptive Biasing Force.