The global population is predicted to increase to around 11 billion by 2100. By 2050, the average age in the most populous age group will be over sixty. The ageing population (over sixty-five) is projected to exceed the number of children by 2047. These demographics imply that as the ageing population section increases, there will be a greater need for long-term care services. In order to adequately prepare against this trend, medical experts and evidence-driven policymakers are realising that personalised healthcare can help alleviate the burden related to the planning and commissioning of services allied to long-term care. Central to this picture is conditions that affect the brain - the most important organ of the human body. Dementia, stroke, and other conditions have a tremendous impact on loss of life, quality of life and healthcare cost. The challenge regarding brain disease is exacerbated further due to the difficulty regarding accessibility of this organ, but also due to the immense complexity regarding its morphology and functionality. In this context, advanced biophysical modelling is considered a promising option for studying brain pathophysiology and becomes a priority investment regarding routes for brain research. Simulations offer the promise of improved, clinically relevant, predictive information, acceleration for the pipeline of drug discovery/design and better planning of long-term care for patients. Within this paradigm, a particular model of water transport in the cerebral environment is essential. Numerous brain disorders arise from water imbalance in the cerebral environment, such as hydrocephalus (HCP), oedema and Chiari malformations to name a few. In this research, a novel multiscale model of fluid regulation and tissue displacement in the cerebral environment is developed, arising from the use of Multiple-network Poroelastic Theory (MPET). Characteristics of a four-network poroelastic model (4MPET) are first explored. Then, this model is extended to a fully dynamic (transient) six-network model (6MPET) via the addition of two new compartments, namely the glial cells compartment and the glymphatic system compartment. The introduction of these two compartments in the MPET paradigm reflects recent seminal findings in cerebral physiology, namely the extent and importance regarding transport/clearance of the perivascular spaces of the brain vasculature. We develop and present a numerical implementation of the 6MPET model, and we utilise this framework to analyse acute HCP and cerebral oedema in a variety of settings, in order to show the enhanced capability of the proposed 6MPET model compared to the classical 4MPET. Investigations of acute hydrocephalus through the fully dynamic 6MPET reveal compensatory trans-ependymal pressure behaviour in the glymphatic compartment. It was also shown that aquaporin-4 (AQP4) deficient expression exaggerates ventriculomegaly, and this too is demonstrated in acute hydrocephalus. Additionally, using the 6MPET model, one is able to witness three mitigating factors for cytotoxic oedema. Specifically, these are: reducing water mobility in the glial cells compartment, increasing the compliance of the glial cells compartment and finally AQP4-deficient expression.

L'élaboration de nouvelles stratégies pour la détection de l'activité de la sclérose en plaques (SEP) est importante pour améliorer le diagnostic et le suivi de cette pathologie. Pour cela, nous avons utilisé des microparticules d'oxyde de fer (MPIO) couplées à un anticorps spécifique de la protéine P-sélectine ou de MAdCAM-1. Durant cette thèse, nous avons démontré que l’IRM moléculaire spécifique de la P-sélectine est capable de détecter les événements pathologiques qui se déroulent dans la moelle épinière de modèles murins de SEP chronique et récurrente-rémittente. De façon intéressante, nous montrons ici que cette technique d'IRM peut prédire l'apparition des poussées et des récupérations dans l’EAE. De plus, nous avons démontré que l’IRM moléculaire de MAdCAM-1 est capable de détecter l’inflammation intestinale dans des modèles de pathologies intestinales et de SEP. Les techniques novatrices d'IRM développées dans cette étude pourraient apporter de nouvelles avancées dans le diagnostic et le pronostic des rechutes de la SEP en ciblant l'activation vasculaire. Enfin, nous rapportons dans la dernière partie de cette thèse que le système glymphatique existe également dans le parenchyme de la moelle épinière de la souris. Dans l’EAE, l’activité de ce système est réduite dans la moelle épinière mais pas dans le cerveau ou le cervelet. Cette altération est associée à l'accumulation de cellules inflammatoires dans l'espace péri-vasculaire, à la désorganisation de l'AQP4 et entraine une forte augmentation du volume ventriculaire. Ces perturbations pourraient contribuer à la physiopathologie de la SEP. Nos résultats sont très prometteurs pour l'élaboration de nouvelles stratégies thérapeutiques. / Developing new strategies to detect disease activity in multiple sclerosis (MS) is essential to improve the diagnosis and follow-up of this pathology. To this aim, we used microparticles of iron oxide (MPIO) coupled to an antibody specific to the P-selectin or MAdCAM-1 protein. In this thesis, we establish that molecular MRI specific to P-selectin protein is able to detect the pathological events that take place in the spinal cord of chronic and relapsing-remitting models of MS in mice. Interestingly, we show here that this MRI technique can predict the apparition of relapses and recoveries in EAE. Moreover, we demonstrate that MRI specific to MAdCAM-1 protein is able to detect the gut inflammation that takes place in models of bowel diseases or MS. The innovative MRI techniques developed in this study could bring new advances in the diagnosis and prognosis of MS relapses by targeting gut inflammation. In the last part of this work, we report that the glymphatic system also exists in the spinal cord parenchyma of the mouse. In EAE, the activity of this system is reduced in the spinal cord but not in the brain or cerebellum. This alteration is associated to inflammatory cell accumulation within the perivascular space, AQP4 disorganization and leads to a large increase of ventricular volume. These disruptions could contribute to the MS pathophysiology. Our results hold significant promise for the development of new therapeutic strategies.

The glymphatic system (GS) is primarily a neural waste clearance system that relies on cerebrospinal fluid (CSF) to transport neuronal byproducts and nutrients. Studies demonstrate that sleep facilitates movement within the GS to clear metabolites and maintain cerebral homeostasis. However, functions of the GS during sleep and its implications have predominantly been examined in animals, clinical/at-risk, and ageing populations. Our understanding of the neural mechanisms underlying GS during sleep in typically developing human infants is limited. The objective of this study was to investigate the relationship between GS imbalance (characterized by extra-axial CSF [EA-CSF] from MRI structural images) and sleep problems in early development. Data from 75 infants were obtained from the Baby Connectome Project. Sleep was indexed with the Brief Infant Sleep Questionnaire. Multilevel models were utilized to explore the associations of EA-CSF volumes and EA-CSF/total cerebral volume (TCV) ratios with age and sleep. We replicated previous findings on lower TCV and overall CSF volumes in infants with dysregulated sleep compared to infants with regulated sleep. Results also demonstrated a decline in EA-CSF/TCV ratios from 9 to 34 months of age (b = -0.0005, <i>t</i> = -2.19, <i>p</i> = .032). Sleep problems were not associated with differential developmental trajectories of EA-CSF volumes or EA-CSF/TCV ratios. Findings from the present study do not support sleep problems as a mechanism through which CSF disbursement within the GS is altered. Although elevated EA-CSF is associated with developmental and neurodegenerative pathology, in early typical development, its links with sleep dysregulation are not robust.