Multiple sclerosis is a chronic inflammatory disease of the CNS associated with widespread primary demyelination and axonal degeneration. The mechanisms underlying the expression of neurological deficits are incompletely understood. Recent histological findings are consistent with a view that active MS lesions may be hypoxic (i.e. suffer a low oxygen concentration). For example, inflammatory ‘Pattern III’ MS lesions express hypoxia inducible factor-1a (HIF-1a), a master regulator of genes whose function is to bias a cell for survival under hypoxic conditions. Hypoxia also results in a number of other consequences designed to limit energy expenditure, including the inhibition of protein translation by the phosphorylation of the eukaryotic initiation factor (eIF-2a), and the formation of stress granules (SGs). Both hypoxia and the inhibition of protein synthesis could cause neurological deficits and thus contribute to the neurological deficits of MS. The aim of the present study is to explore experimental models of MS for evidence of a) hypoxia and b) inhibition of protein translation. The experimental models of inflammatory demyelinating lesions were induced either by the intraspinal injection of lipopolysaccharide (LPS) into the rat dorsal column (LPS-DC, focal lesion), or by immunization of rats or mice to induce experimental autoimmune encephalomyelitis (EAE, disseminated lesions). Animals were examined daily for the expression of any neurological deficit, and tissue hypoxia was detected during life by the systemic administration of pimonidazole, a marker for hypoxia, several hours before termination. Tissue was taken at different stages of lesion development (1-28 days post injection) and examined immunohistochemically for the presence of hypoxia, determined by the expression of binding for pimonidazole, and for the inhibition of protein translation by examining the expression of eIF-2a and SGs. Other markers of disease activity were also examined, including a marker of microglial/macrophage activation (ED-1), HIF-1a, and glucose transporter-1 (GLUT-1). In LPS lesions, labelling for pimonidazole was most intense at the site of injection, 24 hours later. In EAE, labelling for pimonidazole was present as early as 2 days post immunization, but it was expressed more strongly when animals were exhibiting a neurological deficit, subsiding thereafter. In animals injected with LPS, eIF-2a and SGs were expressed most intensely 24 hours post LPS injection, localised to the spinal motor neurons. In EAE, eIF-2a and SGs were expressed in spinal motor neurons, and in cerebellar neurons, at the onset of neurological deficits. These findings reveal for the first time that inflammatory demyelinating lesions are associated with the presence of tissue hypoxia and markers of the inhibition of protein synthesis. It appears that these phenomena may contribute to the expression of neurological deficits, opening new opportunities for therapy.

Mutations in the DJ‐1 gene can induce the development of early‐onset Parkinson's disease (PD) through a loss of protein function. Currently any possible role for DJ‐1 in sporadic PD remains undetermined. To address this, we have studied the characteristics and activities of DJ‐1 in post‐mortem human brain tissue in order to gain insights into its contribution to the development of PD and other neurodegenerative disorders. Western blotting revealed DJ‐1 protein expression to be reduced in several brain regions associated with PD pathology including nigra, striatum and frontal cortex. Similarly levels of DJ‐1 mRNA were also shown to also be lower in PD striatum and frontal cortex suggesting a transcriptional regulation of protein expression in human brain. Further analysis of DJ‐1 gene expression showed PD related changes to be variable throughout the brain, with regions like the amygdala and entorhinal cortex displaying an up‐regulation. DJ‐1 protein was also shown to undergo increased oxidation in PD cases, highlighting the elevated oxidative stress conditions in PD. By using immunoprecipitation to investigate a possible role for DJ‐1 as an in vivo regulator of translation, we found DJ‐1 protein associates with RNA transcripts for selenoproteins, PTEN/Akt pathway components and mitochondrial subunits of complex 1. Protein levels for a number of these transcripts were altered in PD tissue without any parallel change in mRNA levels. DJ‐1 is reportedly involved in a diverse range of cellular activities and its proclivity to associate with multiple RNA species provides a simple biochemical mechanism for this. Moreover it demonstrates that under conditions of elevated oxidative stress, DJ‐1 can instigate a rapid and compartmentalised up‐regulation of pro‐survival proteins in a transcriptionally independent manner. Analysis of DJ‐1 in tauopathies showed co‐localisation with 3R and 4R tau, implicating a possible chaperone function for DJ‐1. Unlike in PD, no altered expression of DJ‐1 mRNA and protein was observed.

Multiple Sclerosis is the most common non-traumatic cause of neurological disability in young people. Until recently few treatments existed for Multiple Sclerosis and Interferon beta was the first and remains the most commonly prescribed. As a biological product it induces antibodies to the protein which may abrogate the efficacy of the drug (neutralising antibodies - NAbs). Testing for these antibodies has been problematic as biological assays are difficult to standardise, time consuming and expensive. In the experiments described here we sought to develop a novel cell-based reporter-gene assay to reliably test for Nabs, to explore the relationship between NAbs, treatment efficacy, biological activity and correlate NAb titres with in vivo biomarker induction to establish guidelines to interpret results. The work presented in this thesis describes the development and validation of the luciferase assay and has shown that subjects who develop NAbs experienced increased relapse rates, (which lag behind the appearance of NAbs). Evaluating the in vivo biological response to IFNβ injection it was established that in subjects with NAbs there was titre dependent loss of bioactivity with reduced myxovirus resistance protein A (MxA) level in those with titres 100 – 600 NU and absent response in those with titres > 600 NU. Opinion amongst neurologists (UK, USA, Canada and Austria) regarding NAbs was evaluated and revealed uncertainty of their significance in the clinical setting, and the reluctance of some neurologists to incorporate NAb testing into routine practice. Results were similar in the countries surveyed in that 90 – 100% were aware of NAbs and thought they abrogate clinical efficacy, yet few routinely tested for them, particularly in the UK. The validated luciferase assay has since been dissemintated to 9 countries. Since it’s launch in the UK in 2006 over 4,000 patients have been tested for NAbs. This work directly translates into clinical practice and provides a useful service to aid management of subjects with MS.

The key component of the Wnt/β-catenin pathway is β-catenin, a 780 amino acid protein which was originally identified as a component of E-cadherin junctions, where it links E-cadherin with α-catenin and consequently to the actin cytoskeleton. In the absence of Wnt signalling, cytoplasmic β-catenin is phosphorylated and subsequently degraded by ubiquitinylation. Activation of Wnt signalling stabilises cytoplasmic β-catenin, which translocates into the nucleus where it binds to nuclear transcription factors TCF/LEF and regulates the expression of over thirty genes, including cell cycle activating genes. The Wnt/β-catenin pathway is essential for nervous system development and it is aberrantly active in common malignant tumours of the central nervous system, such as glioblastoma in adults and medulloblastoma in children. We have generated a mouse model expressing a Cre-recombinase-inducible form of degradation-resistant (oncogenic) β-catenin. We generated compound mutant mice to activate the Wnt/β-catenin pathway in specific brain regions and during several developmental time points. Compound mutant transgenic mice expressed dominant active β-catenin in vivo, but did not show developmental abnormalities or brain tumours even in combination with inactivated p53 alleles. Similarly, neural stem cells expressing dominant active β-catenin did not show a growth advantage or altered self-renewal. We attributed these results to low level expression of dominant active β-catenin. In addition, we later acquired mice that carry an inducible deletion of the exon 3 of β- catenin (containing the phosphorylation site). Expression of this form of β-catenin did not increase neural stem cell growth and self-renewal. However, in vivo expression results in brain malformation and lethality depending on the cells in which β-catenin was activated, however, recombination in adult stem cells in vivo did not result in brain tumours.

If we are to approach rehabilitation of memory-impaired patients in a systematic and efficacious way, then it is vital to know if the human memory system has the propensity for plasticity in adulthood, the limiting factors on such plasticity, and the timescales of any plastic change. This thesis was motivated by an attempt to develop a body of knowledge in relation to these questions. There is wide agreement that the hippocampus plays a key role in navigation and memory across species. Evidence from animal studies suggests that spatial memoryrelated hippocampal volume changes and experience-related hippocampal neurogenesis takes place throughout the lifespan. Previous studies in humans indicated that expert navigators, licensed London taxi drivers, have different patterns of hippocampal grey matter volume relative to control participants. In addition, preliminary evidence also suggested there may be functional consequences associated with this grey matter pattern. Using licensed London taxi drivers as a model for learning and memory, the work undertaken centered on four key issues: (1) In a set of studies, I characterised the neuropsychological profile of licensed London taxi drivers in detail, which included devising a number of new table-top associational memory tests. This enabled me to assess the functional consequences of their expertise and hippocampal grey matter pattern in greater depth than previous studies. (2) In order to explore the effects of taxi drivers’ expertise in more naturalistic settings, I also examined how well they could learn the layout of an unfamiliar town compared with a group of non-taxi drivers, and how effectively taxi drivers could integrate a new district into their existing spatial representation of London. (3) I then conducted a study on experts whose knowledge was much less spatial than taxi drivers in order to examine if the effects on hippocampal grey matter and neuropsychology were general or whether they were specific to the spatial domain. (4) Given that previous taxi driver studies were cross-sectional, the question of whether the human hippocampus can exhibit spatial memory-related structural plasticity in adulthood was uncertain. I therefore conducted a longitudinal study which assessed participants both pre and post taxi driver training using structural MRI and neuropsychological measures. This enabled me to investigate, within subjects, whether hippocampal volume changes can be acquired in response to intense spatial stimulation. In addition, I explored whether ceasing to be a taxi driver (i.e. retiring after many years on the job) resulted in ‘reverse’ plasticity. I found evidence for hippocampal plasticity within individuals as a result of their intense acquisition of spatial knowledge over a number of years that was associated with qualifying to be a licensed London taxi driver, and preliminary evidence of reverse plasticity when taxi drivers retire. This suggests that hippocampal structure and memory ability can be modified in response to environmental factors and are not necessarily hard-wired. However, my results also provide some insights into the boundaries within which human memory operates, as I identified both positive and negative cognitive consequences of being an expert navigator, and also established that the MRI and neuropsychology effects of expertise on the hippocampus may be restricted to the spatial domain.

This thesis describes a series of studies, involving healthy subjects and a carefully selected stroke patient, in which the techniques of Transcranial Magnetic Stimulation (TMS) and Signal Detection Theory (SDT) were combined to explore processing in the posterior parietal cortex (PPC) and the phenomenon of unilateral spatial neglect: 1) A new SDT-based TMS 'hunting' technique was developed and then employed successfully over the right PPC. This revealed a cortical node which could be modulated with 'online' 10Hz-TMS to exert control over subjects' visuospatial perception. 2) By targeting this 'hotspot' and an equivalent area in the left PPC with disruptive 'cTBS' (continuous 'Theta Burst' simulation), neglect-like effects in the healthy brain could be induced and alleviated. These effects were quantified using a newly developed, fully balanced, bihemifield detection paradigm 3) By using TMS to map visuospatial function over healthy right PPC, the 'hotspot' could be enlarged after exposure to excitatory, intermittent TBS (iTBS). 4) cTBS was applied to the left PPC of a patient with a right sided stroke and visuospatial neglect. In doing this, neglect and its alleviation were described for the first time in fully balanced SDT terms. 5) The cerebellum was targeted in healthy subjects with 1Hz inhibitory TMS which induced a shift in their subjective midline for 'imaginary' but not 'real' space, as measured with number or physical line bisection respectively. These TBS studies lend support to Kinsbourne's hemispheric rivalry hypothesis and suggest that the extent of 'spatially eloquent' cortex in the right PPC can be increased. Both strategies could be useful in larger therapeutic studies of patients suffering from USN after right sided brain injury. The final study opens up an additional therapeutic target for patients with imaginal neglect and, for the first time, implicates the cerebellum in number line bisection.

The studies presented in this thesis employ neurophysiological outcome measures and the application of artificially induced cortical stimulation plasticity paradigms to study the effects of human genetic variation on human cortical neuroplasticity. The introductory chapter includes a review of illustrative models of neuroplasticity. I also cover the principles, physiology and pharmacology of TMS and rTMS. With this background, I set out the scope and principles of such an approach applied to the study of human genetic variation, and define the field of Stimulation Genomics. I set out the case for such an approach, highlighting previous studies that have employed neurophysiological outcome measures and the application of artificially induced cortical stimulation plasticity paradigms to study the effects of disease-causing human genetic mutations. In the subsequent introductory chapters I have focused on the rationale of selecting the Brain Derived Neurotrophic Factor polymorphism Rs6265 (BDNF Val66 Met) as the candidate polymorphism for our studies, covering the molecular biology and physiological roles of this highly conserved protein, and with a particular focus on its diverse roles in neuroplasticity. The 1st experiment presented here used established rTMS and TDCS paradigms to probe the effects of the BDNF Val66Met SNP on cortical plasticity and metaplasticity. The results generated from this study, and particularly the results suggesting an effect on metaplasticity, formed the basis for the studies in patients. We investigated the influence of this SNP on the rate of onset of Levodopa-Induced Dyskinesia (LID) in patients with Parkinson's disease and on the penetrance of DYT1 Dystonia. The final experiment presented here was designed to confirm the effects of the BDNF Val66Met polymorphism on the iTBS paradigm, and quantify its effects alongside other variables thought to influence the response to rTMS paradigms. This study also provides some crucial insights into the iTBS paradigm itself.

The data presented in this thesis were collected as part of a trial of neuroprotection with lamotrigine in secondary progressive multiple sclerosis (MS) and comprise clinical assessment and quantitative magnetic resonance imaging (MRI), specifically focussing on T1 hypointense lesion volume (T1LV) and magnetization transfer ratio (MTR). The aim of the study was to evaluate the usefulness of these MRI techniques for monitoring neuropathology in MS by examining the correlation of MRI measures with clinical measures both cross-sectionally and longitudinally and the responsiveness to change of these measures. The potential neuroprotective effect of lamotrigine was assessed by comparing clinical and MRI data in placebo and verum arms of the trial. The introduction briefly summarises: the clinical features of secondary progressive MS; some of the aspects of MS pathology and pathophysiology that may be ameliorated by lamotrigine treatment; some of the techniques for assessing clinical status in MS, including the impact of MS on quality of life (QoL); and the role of quantitative MRI in monitoring putative neurodegeneration in MS. The methods chapter details: the structure and conduct of the trial, including recruitment of subjects, acquisition of clinical and MRI data; and the post-acquisition analysis of the MRI data. The results chapter addresses four issues: the cross-sectional correlation of MRI with clinical measures; the responsiveness to change and longitudinal correlation of MTR measures with clinical measures; the responsiveness to change and longitudinal correlation of T1LV with clinical measures; and the cross-sectional and longitudinal correlation of a large number of MRI and clinical measures with the Multiple Sclerosis Impact Scale (MSIS-29), a measure of the impact of MS on QoL. The conclusions chapter summarises some of the more notable results presented, some of the limitations of the study and proposes some further work that could be done to clarify any areas of uncertainty.

Amyotrophic Lateral Sclerosis (ALS) is an untreatable and fatal neurodegenerative disease that leads to muscle atrophy, paralysis of voluntary muscles and death. ALS is characterized by the selective degeneration of upper and lower motoneurons in the spinal cord, brain stem and motor cortex. The maJority of ALS cases are sporadic, whereas 10 - 20Z of cases have a genetic component and are familial. The causes for most cases of ALS are unknown and the clinical course is highly variable, suggesting that multiple factors underlie the disease mechanism. Recent experimental evidence has illustrated that selective motoneurons degeneration in ALS does not arise solely within motoneurons. The involvement of non-neuronal cells, such as astrocytes, in the pathogenesis of ALS have recently been shown with convincing data supporting the hypothesis of non-cell autonomous mechanism in which motoneuron degenerations is influenced by the toxicity of non-neuronal cells surrounding the motoneurons. In view of this crucial interaction between astrocytes and motoneurons and the implication of astrocytes in ALS disease progression, the role of astrocytes will be examined. An in vitro approach is employed in order to examine the role of mutant SOD1 protein expression in primary astrocyte cells. Motoneurons have an unusually high threshold for induction of heat shock proteins (Hsps), which may contribute to their susceptibility in ALS compared to astrocytes. Thus by increasing the heat shock response (HSR), the cytoprotective mechanism in astrocytes, as a result of up-regulation of Hsp expression within these cells, is a potential therapeutic target. This neuroprotective effect of manipulating the HSR system is investigated in vitro in SOD1G93A transgenic mice. Furthermore, characterization of astrocytes was also carried out in vivo of the cortex and spinal cord in SOD1G93A transgenic mice. Finally, the effect of co- inducing the HSR system in astrocytes by pharmacological manipulation is assessed. The data of the present study clearly indicate that expression of mutant SOD1 protein in astrocytes causes a stress response, and successful manipulation of the increased HSR system in primary astrocytes causes an increase in primary motoneurons survival. Thus targeting the HSR in astrocytes in proximity of motoneurons may be a successful therapeutic approach by circumventing motoneuron degenerations in ALS.

Frontotemporal lobar degeneration (FTLD) refers to a group of diseases characterised by focal frontal and temporal lobe atrophy that collectively constitute a substantial source of clinical and social disability. Patients exhibit clinical syndromes that are dominated by a variety of nonverbal cognitive and behavioural features such as agnosias, altered emotional and social responses, impaired regulation of physiological drives, altered chemical sense, somatosensory and interoceptive processing. Brain mechanisms for processing nonverbal information are currently attracting much interest in the basic neurosciences and deficits of nonverbal processing are a major cause of clinical symptoms and disability in FTLD, yet these clinical deficits remain poorly understood and accurate diagnosis is often difficult to achieve. Moreover, the cognitive and neuroanatomical correlates of behavioural and nonverbal cognitive syndromes in FTLD remain largely undefined. The experiments described in this thesis aim to address the issues of improving our understanding of the social and behavioural symptoms in FTLD through the integration of detailed neuro-behavioural, neuropsychological and neuroanatomical analyses of a range of nonverbal functions (including emotions, sounds, odours and flavours) with high-resolution structural magnetic resonance imaging (MRI). A prospective study of emotion recognition in various domains including music, faces and voices shows that music is especially vulnerable to the effects of damage in FTLD. A profile of brain atrophy associated with impaired emotion recognition in music is identified, comprising a distributed bilateral cerebral network involving areas previously implicated in representing and evaluating the emotional content of stimuli including mesial temporal structures, insula and their connections in the mesolimbic system. Prospective studies of face and chemosensory processing provide further insights into the neuroanatomical framework and structural neuroanatomy for face, odour and flavour processing deficits in FTLD. A profile of cognitive deficits in different components of face processing is shown which correlate with distinct but partly overlapping brain networks. Deficits in flavor and odour identification are shown in FTLD with neuroanatomical correlates involving temporal and limbic areas which include entorhinal cortex, hippocampus and parahipocampal gyrus. A detailed systematic study of music knowledge in two expert musicians with different dementia diseases, sematic dementia (SemD) and dementia with Lewy bodies (DLB), involving a series of novel neuropsychological experiments probing various dimensions of music knowledge, yields new insights into the cognitive architecture of music knowledge and the brain organization of nonverbal knowledge systems. This thesis therefore provides neuropsychological and imaging data in relation to various nonverbal cognitive processes in FTLD that can offer greater insights into our understanding of behavioural symptoms in this group of diseases as well as the cognitive architecture of hitherto relatively poorly understood nonverbal cognitive modalities such as music knowledge, emotion and chemosensory processing.