Reverse engineering a human 3D cell-based model of the neurovascular unit to investigate blood-brain barrier dysfunction in Alzheimer’s disease

Abstract

Neurovascular dysfunction is a major component of Alzheimer’s disease (AD) pathology, yet the aetiology is relatively unknown. It has been hypothesised that there is a “dual hit” of vascular dysfunction either working in conjunction with other AD pathology, or as an instigating factor. The blood-brain barrier (BBB) is the interface between the neural and vascular components of the brain, which when combined, form the neurovascular unit (NVU). BBB breakdown during NVU dysfunction reduces clearance and aids build-up of toxic amyloid-β (Aβ) peptide, as well as causing a reduction in the selective permeability to circulatory compounds and cells that exacerbate AD. Current in vitro research models for investigating the NVU typically utilise stiff substrates, such as tissue culture plastic (TCP) or semi-permeable transwell membranes, in 2D culture with a single cell type in isolation. This is very different from the in vivo situation where NVU cells are mostly in a 3D, soft environment, with other cell to cell interactions at a biochemical and physical level. In this thesis, a physiologically relevant 3D model of the NVU was created using specialised biomaterials called hydrogels in combination with NVU cells differentiated from human induced pluripotent stem cells (iPSCs). In particular, brain microvascular endothelial cells (BMECs) were differentiated from iPSCs to create the main basis for a highly functional BBB within the 3D-NVU model. Oscillatory rheological characterisation determined the suitability of collagen type 1 (ColI) hydrogel (PureCol®), and alginate and ColI blended hydrogels (alg-col). Alg-col hydrogel was unable to support the growth of a fully functional BMEC monolayer independent of stiffness, whereas on PureCol® BMECs formed a fully functional BBB as measured by trans endothelial electrical resistance (TEER). By encapsulating iPSC-derived neurons or pericytes, or human primary astrocytes, within the PureCol® hydrogel, it was possible to co-culture the BMECs with another NVU cell type with appropriate biochemical and physical interactions. This co-culture enhanced the functional properties of the BBB within the 3D-NVU model, with higher and prolonged TEER recordings. The development of both the BMEC monoculture, and BMEC and neuron co-culture, in a 3D-NVU model was used to investigate the effect of AD relevant compounds on BBB integrity. Aβ did not cause any significant changes in BBB integrity in either monoculture or co-culture models. Hydrogen peroxide (H2O2) at high concentration caused BBB disruption in BMEC monocultures, but enhanced BBB integrity in co-culture 3D-NVU models. The protective effect of the anti-ageing protein klotho and of the neuroprotective soluble amyloid precursor protein-α (sAPPα) were both tested as a preventative measure to H2O2 disruption. However, neither klotho nor sAPPα affected BBB integrity. The data presented in this thesis demonstrates the optimisation and development of a 3D-NVU model that has a highly functional BBB, recapitulating the physical properties of the brain. This model can be used in future for the investigation of how AD pathology affects the NVU and BBB, and for more effective drug testing.

Date of Award	1 Aug 2021
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Tao Wang (Supervisor), Nigel Hooper (Supervisor) & Marco Domingos (Supervisor)