Development of a physiologically based in vitro model of the blood-brain barrier

Several in vitro models of the blood-brain barrier (BBB) have been studied for isolated studies of cell biology and barrier permeability. As BBB endothelial cells tend to lose their differentiated barrier characteristics when isolated in cell culture, most of these models have been only moderate replicas of the true BBB. We propose that by increasing the physiological basis of the model, the barrier properties of the model will improve. Our work is based on two hypotheses: 1) increasing the degree of direct contact between endothelial cell and astrocyte-like cells will increase the tightness of endothelial cell intercellular junctions and 2) adding shear stress will lead to rearrangement of endothelial cell junctions more like in vivo. We have fabricated an ultra-thin membrane from low-stress silicon nitride that allows for intimate interaction of endothelial and astrocyte-like cell types grown on opposing sides of the membrane. This membrane is easily incorporated into a flow chamber for controlled shear stress experiments. Initial studies indicate the fabricated membranes support endothelial and astrocyte-like cell growth on opposing sides for several days but that attachment is poor when the cells are subjected to hydrodynamic shear stress