Enhancements in peak motor performance have been demonstrated in response to intense stimuli both in healthy subjects and in the form of ‘paradoxical kinesis’ in patients with Parkinson’s disease. Might the latter phenomenon thus reflect a physiological process? The first study outlined in this thesis suggests this may be the case, as maximal effort grips in healthy subjects undergo dramatic enhancements when the imperative visual cue is accompanied by an intense auditory tone. Analogous enhancements in motor performance are demonstrated in a second study of patients with Parkinson’s disease and age-matched healthy controls. Remarkably, the facilitating effect of loud auditory tones is similar whether patients are off or on dopaminergic medication, suggesting a potentially non-dopaminergic basis for the phenomenon. A role of sub-cortical systems in the performance enhancements engendered by intense stimuli is next considered. Local field potentials recorded from the subthalamic nuclei of patients with Parkinson’s disease, whilst they undertake the above established paradigm, identify both theta/alpha (5-12 Hz) and high gamma/high frequency (55-375 Hz) activity as exhibiting remarkable scaling with maximal motor responses to the visual cue alone, but having little explanatory influence on performance enhancements beyond this. In the final study, a short-latency evoked potential in subthalamic nucleus local field potential recordings, which scales in amplitude with both stimulus intensity and corresponding enhancements in biomechanical measures of maximal handgrips, is identified. Interference with this potential through high frequency deep brain stimulation of the same nucleus, leads to a diminished behavioural effect of stimulus intensity. Recordings of a similar evoked potential in the related pedunculopontine nucleus – a key component of the reticular activating system – provide support for this neural signature as a physiological correlate of ascending arousal, propagated from the reticular activating system to exert an ‘energizing’ influence on motor circuitry through the subthalamic nucleus.

Huntington’s disease (HD) is an inherited neurodegenerative disorder caused by a CAG repeat expansion in the huntingtin (HTT) gene. The peripheral innate immune system contributes to HD pathogenesis and has been targeted successfully to modulate disease progression, but mechanistic understanding relating this to mutant (m)HTT expression in immune cells has been lacking. This thesis demonstrates that human HD myeloid cells produce excessive inflammatory cytokines due to cell-intrinsic effects of mHTT expression on the NFkB pathway, whereby mHTT interacts with IKK, leading to increased degradation of IkB and subsequent nuclear translocation of RelA. Transcriptional alterations in intracellular immune signaling pathways were also observed. Using a novel method of siRNA delivery to lower HTT expression, this thesis shows a reversal of disease-associated alterations in cellular function - the first time this has been demonstrated in human cells. Glucan-encapsulated siRNA particles (GeRPs) were used to lower HTT levels in human HD monocytes/macrophages, resulting in reversal of HTT-induced elevated cytokine production and transcriptional changes. These findings improve our understanding of the role of innate immunity in neurodegeneration, introduce GeRPs as a tool for studying cellular pathogenesis ex vivo in human cells and raise the prospect of HTT lowering in immune cells as a therapeutic in HD. Evaluating immune function in different mouse models of HD, blood and splenic monocytes replicated the hyper-reactive phenotype seen in HD patients demonstrating that HD mouse models can be of use to understanding HD immune pathology and to test immune modulatory therapies. Furthermore, human HD myeloid cells demonstrated a striking defect in migration towards different chemokines. Looking at the cell’s ability to form filopodia, it became apparent that actin-remodelling is reduced and causes decreased migration. Work performed in collaboration with Novartis revealed that mHTT levels in immune cell subsets differ significantly between disease stages. Monocyte and T cell mHTT levels were significantly associated with disease burden scores and caudate atrophy rates in HD patients. mHTT fragments detected in HD immune cells may explain the progressive increase in mHTT levels. These findings indicate that quantification of mHTT holds significant promise as a non-invasive disease biomarker.

Multiple sclerosis (MS) is an inflammatory demyelinating and degenerating disease of the central nervous system (CNS) that typically starts with a relapsing-remitting course of neurological deficits. Among the enigmas in the disease are 1) the cause of the neurological deficits, 2) the cause of the demyelination, 3) the cause of the degeneration, and 4) the cause of the disease itself. This thesis examines the novel hypothesis that tissue hypoxia might illuminate at least some of these enigmas. Tissue hypoxia can easily account for loss of function in a tissue as heavily dependent on oxidative phosphorylation as the CNS, and it can similarly selectively kill cells such as oligodendrocytes and neurons/axons if they are reliant on oxidative metabolism. Hypoxia can also promote inflammation in tissues and thereby reduce the threshold for the initiation of inflammatory disease. Three experimental models have been examined, namely experimental autoimmune encephalomyelilits (EAE, a common model of MS), an experimental model of the demyelinating Pattern III MS lesion, and animals rendered temporarily hypoxic due to placement in an atmosphere of 10% oxygen. We provide chemical, physical and therapeutic evidence that tissue hypoxia is, in part, responsible for 1) neurological dysfunction in EAE, 2) the demyelination in the model Pattern III lesion, in association with nitric oxide and superoxide, 3) by extension, perhaps neurodegeneration, and 4) a sensitization of the CNS to pro-inflammatory conditions, including evidence of the special sensitivity of oligodendrocytes to hypoxia. We conclude that true tissue hypoxia is a hitherto-unrecognised, but potentially important, factor in several of the cardinal characteristics of MS.

Excessive neural synchronisation in the beta frequency band (10-35 Hz) is a hallmark of Parkinson’s disease (PD) but the mechanisms underlying the generation and propagation of such oscillatory activity and its links with movement impairment remain partly unclear. Patients receiving deep brain stimulation of the subthalamic nucleus (STN DBS) provide a unique opportunity to address these issues by either recording the neuronal activity from the implanted electrodes or driving the network to study the behavioural and neurophysiological effects of stimulation. The studies outlined in this thesis bring further understanding into the mechanisms by which beta oscillations pervade every level of the basal ganglia – cortical network, how they impair movement and how they are modified by the usual treatments of PD such as dopamine or STN DBS. We performed a series of experiments involving PD patients either in the post-operative phase of DBS or chronically stimulated, and in an animal model of PD to study the effect of beta frequency STN stimulation on movement, the effect of high frequency STN stimulation on beta oscillations and the network properties underlying its vulnerability to beta frequencies. The findings of these studies suggest the existence of anatomically segregated subthalamo-cortical networks, with specific properties and susceptibility to resonate variably impaired in PD patients. We propose that excessive beta oscillations in PD are primarily due to a failing damping system rather than a net increase in their generation, and that dopaminergic drugs act to partially restore this damping system, while high frequency DBS suppresses pathological local beta activity. / Abstract in French: Les oscillations neuronales dans la bande beta (10-35 Hz) sont caractéristiques de la maladie de Parkinson (MP). Toutefois, les mécanismes qui sous-tendent la genèse et la propagation des ces activités et leurs liens avec les troubles moteurs restent obscurs. La chirurgie de stimulation des noyaux sous-thalamiques (NST) permet d’enregistrer l’activité neuronale par l’électrode implantée et d’étudier les effets moteurs et neurophysiologiques de la stimulation. Les travaux présentés dans cette thèse ont pour but d’améliorer la compréhension de l’origine de la diffusion des oscillations beta dans le réseau subthalamo-cortical, de leur retentissement sur la motricité et de l’influence des traitements de la MP sur celles-ci. Nous avons effectué une série d’expériences chez des patients parkinsoniens et chez un modèle animal de la MP pour étudir les effets de la stimulation du NST dans la bande beta sur la motricité, ceux de la stimulation à haute fréquence du NST sur les oscillations beta et les propriétés de réseau à l’origine de sa vulnérabilité aux fréquences beta. Les résultats de ces travaux suggèrent l’existence de réseaux subthalamo-corticaux distincts, présentant des propriétés et une susceptibilité à la résonance propres, et qui seraient atteints de façon variable selon les patients, possiblement en rapport avec la variabilité clinique de la MP. Nous proposons que les oscillations beta dans la MP sont principalement dues à la faillite du système d’atténuation plutôt qu’à une augmentation de leur genèse, que les traitements dopaminergiques restaurent partiellement ce système d’atténuation alors que la stimulation à haute fréquence du NST supprime cette activité.

Axonal loss is an important cause of irreversible disability in multiple sclerosis. Whilst recovery of neurological function due to restoration of neuronal conduction is usual after multiple sclerosis relapse, some relapses are followed by fixed disability. Quantitative measures of axons in the retina, and in their continuation in the optic nerve allow measurement of the effect upon axons of a prototypical multiple sclerosis relapse, optic neuritis. Using optical coherence tomography, axonal and neuronal loss were observable in the retina of patients with progressive multiple sclerosis, even when they were clinically unaf- fected by optic neuritis. With follow-up, these changes were found to be stable. After optic neuritis, 99% of retinal nerve fibre layer thinning was detectable by 4.75 months, and 99% of macular volume loss was detectable by 11 months. The earliest time to detect retinal nerve fibre layer thinning was 1.64 months, and the earliest time to detect macular volume loss was 0.99 months. Sample sizes for clinical trials generated for trials of neuro-protective agents, using multiple sclerosis thickness as an endpoint, showed that 35 subjects per arm of a randomised trial would be needed to detect an 50% effect, with 80% power. Early measures of colour vision and visual evoked potential delay were strongly related to the eventual degree of axonal loss. Comparisons of two methods of measuring multiple sclerosis thinning after optic neuritis suggested that optical coherence tomography would be superior to scanning laser polarimetry. Comparisons of optic nerve mean area and multiple sclerosis thickness showed that whilst initial swelling was similar, the degree of subsequent atrophy was less in the optic nerve. Measures of axonal loss in the retina will be an important method for assessing the efficacy of therapies that propose to prevent axonal loss in multiple sclerosis and optic neuritis.

Parkinson’s Disease (PD) is a common neurodegenerative condition characterised pathologically by progressive dopaminergic cell loss in the substantia nigra pars compacta, dopamine depletion and resulting cortico- basal ganglia circuit dysfunction. There is a considerable variation in symptoms and treatment response between patients and therefore a need to individualise treatments, such as dopamine replacement therapy, and deep brain stimulation (DBS). We therefore require a better understanding of how different motor and non-motor symptoms emerge from the cortico-basal ganglia dysfunction characteristic of PD. In this thesis, I investigated the hypothesis that distinct symptoms in PD may be due to the dysfunction of distinct cortico-basal ganglia circuits. I characterised cortico-basal ganglia coupling by simultaneously recording cortical activity with magnetoencephalography (MEG) and basal ganglia activity from intracranial electrodes placed during DBS surgery for PD. Coupling was measured in terms of coherence – a frequency specific measure of coupling. I found that resting cortico-basal ganglia networks had distinct cortical topographies at different frequencies. Frontal regions coupled to both the subthalamic nucleus (STN) and the pedunculopontine nucleus region (PPNR) in the beta frequency band whilst temporal, parietal and cerebellar areas coupled in the alpha range. I hypothesised that activity in the frontal beta network may relate to executive function, and found that local synchronisation in two frontal cortical hubs was related to stopping an on-going movement – a crucial executive function. In a related experiment in PD patients, transient frontal – basal ganglia coupling was again apparent during motor inhibition, but how this is related to behavioural performance needs further investigation. These results are useful in highlighting how cortico-basal ganglia networks can be separated both spatially and spectrally and how the function and dysfunction of these networks can be interrogated in PD patients. Future work should determine how different stimulation parameters differentially affect these distinct circuits.

To extract a meaningful visual experience from the information falling on the retina, the visual system must integrate signals from multiple levels. Bottom-up signals provide input relating to local features while top-down signals provide contextual feedback and reflect internal states of the organism. In this thesis I will explore the nature and neural basis of this integration in two key areas. I will examine perceptual filling-in of artificial scotomas to investigate the bottom-up signals causing changes in perception when filling-in takes place. I will then examine how this perceptual filling-in is modified by top-down signals reflecting attention and working memory. I will also investigate hemianopic completion, an unusual form of filling-in, which may reflect a breakdown in top-down feedback from higher visual areas. The second part of the thesis will explore a different form of top-down control of visual processing. While the effects of cognitive mechanisms such as attention on visual processing are well-characterised, other types of top-down signal such as reward outcome are less well explored. I will therefore study whether signals relating to reward can influence visual processing. To address these questions, I will employ a range of methodologies including functional MRI, magnetoencephalography and behavioural testing in healthy participants and patients with cortical damage. I will demonstrate that perceptual filling-in of artificial scotomas is largely a bottom-up process but that higher cognitive functions can modulate the phenomenon. I will also show that reward modulates activity in higher visual areas in the absence of concurrent visual stimulation and that receiving reward leads to enhanced activity in primary visual cortex on the next trial. These findings reveal that integration occurs across multiple levels even for processes rooted in early retinotopic regions, and that higher cognitive processes such as reward can influence the earliest stages of cortical visual processing.

Migraine can be conceptualised as a disorder of sensory processing, manifest by such symptoms as headache (pain), phonophobia and photophobia. Current models of migraine pathophysiology incorporate a significant role for the brainstem. Vestibular migraine (VM) is a subtype of the disorder in which significant brainstem dysfunction has been documented. The condition is known to have a significant effect on mental health. This study was designed to investigate disturbances in audiovestibular brainstem function in vestibular migraine in a four part study: 1. Otoacoustic emission suppression by contralateral noise, a test of auditory efferent pathway function, was measured in a group of 33 VM patients and compared with 31 healthy controls. Regression analysis showed a higher rate of abnormality amongst the VM group (p=0.03). 2. Vestibular evoked myogenic potentials were recorded in a group of 30 VM patients and compared with 35 healthy controls. Recordings showed a higher rate of abnormal responses in the VM group than amongst controls (p=0.008). 3. The potential for vestibular stimuli to act as migraine triggers was investigated by observing the effect of vestibular testing or a control condition on 148 individuals. Vestibular stimulation was associated with a significant increase in the probability of developing a migraine attack over the following 24 hour period (p=0.01). 4. Psychological symptoms of depression and anxiety were assessed using questionnaires 39 patients with VM and compared with a control group of 44 patients with dizziness of other causes. Although the VM group had a significantly higher load of symptoms of depression and anxiety, regression modelling showed that this effect was largely accounted for by an excess of dizziness symptoms. In conclusion, this study documents a number of audiovestibular sensory processing abnormalities using a variety of techniques. Vestibular migraine has a significant effect on psychological wellbeing, largely via the associated balance symptoms.

The five studies presented in this thesis consider different aspects of the long-term prognosis of epilepsy and febrile seizures. The studies were: 1). A systemic review to examine a) how the risk of premature mortality in an individual with epilepsy changes over time and b) whether population mortality rates due to epilepsy have changed over time. (Study 1) 2). An extension of the National General Practice Study of Epilepsy (NGPSE), a prospective community-based incident cohort study to examine a) long-term seizure prognosis and mortality in people with epilepsy and b) seizure prognosis in children with febrile seizures (Study 2) 3). Two retrospective hospital-based cohort studies examining prognosis of chronic epilepsy with regard to a) frequency of different seizure patterns (Study 3) and mediumterm seizure outcome following anti-epileptic drug changes (Study 4). 4). A systematic review examining the impact of aetiology and others factors on outcome in status epilepticus (Study 5). Amongst the findings were: 1): There is no conclusive evidence that either the overall standardised mortality ratio (SMR) or the mortality rate of people with epilepsy has changed significantly over time. The SMR is highest soon after diagnosis and subsequently decreases with a possible late increase after 10 years. In the NGPSE cohort the SMR remains significantly elevated after 20 years despite over 80% currently being in terminal remission. 2): 6.7% (95% CI 4, 11%) of children with febrile seizures developed epilepsy after 20 years. 3): Approximately one-third of people with chronic epilepsy have a history of at least one significant period of seizure freedom (two or more years) while a comparable number with apparent drug-resistant epilepsy attain at least one year of seizure freedom after medication change, although approximately half subsequently relapse. 4). Aetiology and, to a lesser extent, age are the primary determinants of prognosis in status epilepticus.

The aim of this project was to investigate the causes of spinal neural tube defects (NTDs), using the curly tail (ct/ct) mouse as a model system. The ct mutant allele corresponds to a hypomorphic allele of grainyhead-like-3 (Grhl3) gene. A two-dimensional protein gel electrophoresis (2-DE) based approach was used to compare the proteome profile of ct/ct embryos with a genetically matched wild-type strain at the stage of spinal neural tube closure. This analysis revealed a series of proteins whose abundance or 2-DE gel migration are abnormal in ct/ct embryos. Detailed follow-up analysis was performed on one protein, lamin B1. Differential migration of lamin B1 on ct/ct compared with wild-type 2-DE gels was found to result from a sequence change in Lmnb1, resulting in the deletion of a glutamic acid (E) in a region of 9 glutamic acids in the wild-type protein. Lamin B1 in ct/ct therefore only has 8 glutamic acids in this part of the protein. Further analysis showed that the lamin B1 variants functionally differ. Genetic crosses were performed to generate sub-strains of ct/ct mice carrying different combinations of the Grhl3 mutation and lamin B1 variants. These studies support the hypothesis that Lmnb1 can modify the risk of NTDs in the ct/ct strain. Finally, while ct/ct NTDs result from diminished Grhl3 expression, the effects of Grhl3 over-expression were also investigated by intercrossing curly tail Grhl3-transgenic mice (ct/ctTgGrhl3). High levels of Grhl3 expression were found to cause NTDs at high frequency, indicating that Grhl3 regulation is an important requirement for neural tube closure. Morphological and gene expression analysis in Grhl3 over-expressing transgenic embryos suggest that the cellular mechanism underlying NTDs differs from that in the ct/ct hypomorphic mutant.