Generation of a blood-brain barrier (BBB)-on-a-chip model for modeling nanobody transcytosis

Abstract

The blood-brain barrier (BBB) and its selective transport mechanisms contribute to the brain homeostasis by preventing harmful substances from entering the brain and by tightly regulating nutrients exchange between the blood and the brain. The integrity and functions of the BBB are impaired during progression of various neurological diseases. As a central part in neurological diseases, BBB becomes the target during the development of brain-targeted therapies. Drugs can be designed to either directly target the BBB to restore its barrier functions and properties, or to bypass the BBB for treating diseased brain cells. In-vitro BBB models are important tools for fundamental brain research, disease modeling, and drug screening. Currently, there are unsolved problems in the conventional in-vitro BBB models. For example, the simplicity of transwell models fails to truly reproduce the complex pathophysiological environment of the BBB. The aim of this thesis is to generate an advanced BBB model that recapitulates key structures and functions of the in-vivo BBB, and could potentially improve BBB disease modeling and drug screening outcomes over the use of the transwell BBB models. In chapter 2, we generated two BBB-targeting nanobodies (Nbs) that bind to human heparin-binding EGF-like growth factor (HB-EGF). Using a transwell-based BBB model, we demonstrated that the HB-EGF targeting Nbs are able to cross the brain endothelial cells via receptor-mediated transcytosis (RMT). These Nbs could be further developed into BBB shuttle carriers for delivering therapeutical molecules into the brain. In chapter 3, we generated a BBB-on-a-chip model in a microfluidic chip platform. Our primary goal in this model was to create an in-vivo like BBB architecture. The brain endothelial cells formed a tight tube structure in the chip which can be then perfused with cell medium. The supporting neural cells (e.g., astrocytes and neurons) grew next to the endothelial tubes in a collagen-Matrigel double-ECM structure. Over time, the neural cells formed extensive networks and neural-vascular interactions. By using the HB-EGF targeting Nbs, it is shown that the microfluidic BBB chip model has better fidelity in revealing true BBB endocytosis and transcytosis efficiency of these Nbs than the transwell BBB model. The use of non-human derived cells for BBB modeling has difficulties in faithfully recapitulating human BBB physiology. In chapter 2, we differentiated hiPSCs into brain endothelial cells and used these cells to construct an induced BBB model in the transwell system. In chapter 4, we differentiated hiPSCs into functional neurons and astrocytes. We created an isogenic neural model by co-culturing these hiPSC-neural cells in the transwell system. We show that a co-culture environment is necessary in maturing both cell types. Besides, we demonstrate that the neuron-astrocyte co-culture model is suitable for modeling astrocyte-driven neuroinflammation. Bioprinting technology represents a promising approach in generating tissue-like components within a microenvironment that mimics biochemical and biophysical parameters of the brain. In chapter 5, we provide a review of neural bioprinting, which has the potential for advanced BBB engineering incorporating complex vascularized BBB structures. In chapter 6, the main findings of the thesis are summarized and discussed.