A Three-Dimensional Cell Culture System To Model RNA Virus Infections at the Blood-Brain Barrier

The blood-brain barrier (BBB) comprises the foremost protective barrier in the brain and is composed in part of a layer of microvascular endothelial cells that line the capillaries surrounding the brain. Here, we describe a human three-dimensional (3-D) cell-based model of the BBB microvascular endothelium that recapitulates properties of these cells in vivo, including physiologically relevant transcriptional profiles, the capacity to induce potent antimicrobial innate immune signaling, and the ability to resist infection by diverse RNA viruses, including members of the enterovirus (coxsackievirus B, echovirus 11, enterovirus 71, poliovirus) and flavivirus (dengue virus, Zika virus [ZIKV]) families. We show that disruption of apical tight junctions by proinflammatory cytokine tumor necrosis factor alpha (TNF-α) sensitizes 3-D-cultured BBB cells to ZIKV infection and that 3-D derived BBB cells can be used to model the transmigration of ZIKV-infected monocytes across the endothelial barrier to access underlying astrocytes. Taken together, our findings show that human BBB microvascular endothelial cells cultured in 3-D can be used to model the mechanisms by which RNA viruses access the central nervous system (CNS), which could be used for the development and screening of therapeutics to limit this event. IMPORTANCE Neurotropic viral infections are significant sources of global morbidity and mortality. The blood-brain barrier (BBB) is composed in part of a layer of microvascular endothelial cells and functions to restrict viral access to the brain. In vitro models that recapitulate many of the properties of the human BBB endothelium are lacking, particularly with respect to the unique cellular and immunological mechanisms by which these cells restrict viral infections of the brain. Here, we developed a three-dimensional cell culture model that recapitulates many of the morphological and functional properties of the BBB microvasculature and apply this model to the study of RNA virus infections. The model we describe can therefore be used to study a variety of aspects of BBB physiology, including the mechanisms by which viruses might access the CNS, and could be used for the development and screening of antiviral therapeutics to limit this important step in viral pathogenesis.