An AQP4-focused, HTS-compatible, BBB-on-a-chip model

"Background: Aquaporin-4 (AQP4) is the main water channel protein in the brain, where it is expressed in astrocytes and enriched at the blood-brain barrier, and allows water to move both between the blood and the brain tissue and through perivascular spaces as part of the recently-described glymphatic system. Following stroke or head injury, water homeostasis is disrupted, which can lead to influx of water into the brain.

This excess water causes the brain to swell, increasing intracranial pressure, which can be fatal or lead to long-term disability.

AQP4 is therefore established as a drug target for cerebral oedema following stroke and head injury. Despite this, little progress has been made on development of direct AQP4 inhibitors. A recent discovery by my supervisor, Philip Kitchen, has shown that AQP4 can rapidly relocalise from intracellular vesicles to the plasma membrane, and that targeting this relocalisation is a viable therapeutic strategy to prevent or minimise brain and spinal cord oedema in rodent models of CNS injury (Kitchen et al., 2020).

However, the drugs used in this study have clear potential for side-effects if used in patients. Further exploitation of this discovery for patient benefit will therefore require a screening campaign to discover novel AQP4 trafficking inhibitors.

2-dimensional (2D) astrocyte monocultures have been used to make some progress in understanding the molecular biology of AQP4. However, a key weakness is that monocultured astrocytes are not polarised in the same way as astrocytes in vivo, do not develop endfeet, and AQP4 is not localised to specific vasculature-facing membrane sub-domains. Therefore, an intermediate between simple 2D astrocyte monocultures and physiologically relevant in vivo experiments, using human cells in a high-throughput screening (HTS)-compatible system, would be an ideal platform for AQP4 drug discovery.

Such an in vitro system does not currently exist for the study of AQP4. The goal of this project is to develop this system.

Research Plan/Methods: We will develop a BBB model using the HTS- compatible microfluidic organ-on-a-chip platform developed by Mimetas BV (Wevers et al, 2018), who are collaborators on my BBSRC Discovery fellowship. Unlike most organ-on-a-chip platforms, the Mimetas system is compatible with existing high-throughput imaging, microplate and robotics technologies, making it ideal for screening projects.

In addition, it incorporates a simple, gravity-driven perfusion system, which is crucial for development of full endothelial barrier function. The Mimetas system's "phaseguide" technology means support membranes are not required and that adjacent microfluidic channels allow direct cell-cell contact (here between astrocytes, endothelial cells and pericytes).

Cell seeding densities, extracellular matrix composition, and media compositions will be optimised for formation of astrocyte endfeet and endfoot localisation of AQP4. This will be measured in chemically fixed co-cultures using primary astrocytes by immunofluorescence, or by live-cell imaging using iPSC-derived astrocytes which have been edited using CRISPR/Cas9 to have a 3' (C-terminal) eGFP tag on one copy of the AQP4 gene.

These cells have already been produced and partially characterised with support from the Joint Research Group Fund. Full characterisation will form part of this project.

Expected outcomes: Several publications describing our AQP4-eGFP iPSC astrocytes, and our optimised BBB-on-a-chip model. More importantly, we will have a model to use for screening for novel inhibitors of AQP4 trafficking. My supervisor is a founding shareholder of Estuar Pharmaceuticals which has received significant VC investment to fund several screening projects. There is therefore a clear path to exploitation for any discoveries made during this studentship.

References Wevers et al., Fluids Barriers CNS. 2018; 15: 23. Kitchen et al., Cell. 2020; 181(4): 784-799.e19. "