Developing a Porcine Model to Study the Glymphatic System

Abstract: The glymphatic system is a brain-wide solute clearance system that has developed in the brain to clear metabolic waste during sleep. This clearance is mediated by advective fluxes of cerebrospinal fluid (CSF) along perivascular spaces (PVS) and through the brain. The movement of CSF from the PVS and through the brain is dependent on the polarised expression of aquaporin-4 (AQP4) at astrocyte endfeet, that project to form the outer PVS boundary. The capacity of the glymphatic system to clear proteins like amyloid-beta (Aß) and tau has generated great interest in exploiting this system therapeutically in the context of Alzheimer’s disease (AD). However, much of the knowledge on the glymphatic system, more specifically, the microscopic machinery and processes, have been studied and described exclusively in rodents. To this end, apart from several magnetic resonance imaging studies confirming the existence of macroscopic glymphatic phenomena in humans i.e. advective movement of gadobutrol from the CSF into the brain, the microscopic aspects of the glymphatic system remain largely unexplored in the large gyrencephalic brain. Thus, the aims of this thesis were to develop a large animal model to study the glymphatic system and further use this model to study the glymphatic system in the context of neuropathology, which in this case took the form of sub-dural haematoma, and amyloid-beta (re)-circulation in the context of AD. To achieve this, a cisterna magna (CM) cannulation surgery was translated from rodents to pigs, such that it would be possible to introduce fluorescent tracers into the porcine CSF in vivo and explore the glymphatic pathways. Initially, to characterise these pathways whole brains were extracted intact after fluorescent tracer circulation and processed using several advanced imaging readouts, including macroscopy, along with confocal, light sheet and electron microscopy. Imaging data revealed: 1) The folded architecture of the gyrencephalic brain helped direct upstream CSF distribution, 2) PVS-mediated CSF influx into the brain is steadfast in the pig brain, as in rodents, and could be traced in deep sub-cortical structures and down to a capillary level, 3) PVS influx sites are 4-fold more extensive in pigs than in rodents. Taken together these data indicate not only a conservation of the glymphatic system and its machinery from rodents to pigs but a more developed system in the large mammal. In the context of suspected sub-acute subdural haematoma (SDH) a brain-wide impairment in glymphatic influx amounted, raising important questions concerning the consequence of undiagnosed SDH for glymphatic function and brain clearance. In the context of AD the acute introduction of Aß1-42 into the CSF was found to impair glymphatic function, highlighting the consequences of Aß recirculation for brain clearance. Interestingly, upon closer examination it was found that Aß did not penetrate into the brain as was the case with inert protein tracers, but instead remained localised to pial and penetrating arteries. This localisation was elastin specific and only occured with Aß (1-40/1-42) but not dextran or bovine serum albumin. This outcome appears to reflect an Aß entrapment system that prevents the recirculation of Aß into the brain, but further work is necessary to unravel its potential as a clearance pathway for Aß from the CSF to protein transporters at the endothelial cell layer. In an attempt to study this in vivo a porcine cranial window model was generated in order to image PVS transport in the large gyrencephalic brain. While low resolution imaging was indeed possible, brain motion proved a challenge not yet overcome. We hope that in the future, through mitigating brain motion, it will be possible to study PVS transport, the glymphatic system, and this potential Aß entrapment pathway, in the brain of a large mammal in vivo.