Background: Neurodegenerative diseases are diseases of the nervous system with progressive course leading to death. Treatment remains symptomatic. Development of neuroprotective agents has been hampered for various reasons. This includes the inability of making the diagnosis accurately early in the course and the lack of reliable disease progression markers which could be used in future treatment trials. Transcranial magnetic stimulation (TMS) is a non-invasive and pain-free method for assessment of brain function. Methods: Here we evaluated TMS and its potential of serving as a reliable biomarker for neurodegenerative diseases with genetic cause. After clinical delineation of our patient cohorts with Huntington's chorea and young-onset Parkin-related Parkinsonism, we enrolled both patients as well as asymptomatic/presymptomatic gene-carriers. Patients, carriers and age-matched healthy controls were studied using TMS to establish an electrophysiological footprint of these conditions. Results: We found abnormalities in electrophysiological parameters which were present in manifesting patients and/or non-manifesting gene mutation carriers. In HD, both presymptomatic and early manifest patients had increased resting and active motor cortex thresholds. Short afferent inhibition (SAI), a measure of sensory-motor integration, was reduced in manifesting patients only. SAI changes were inversely correlated with clinical parameters like predicted years to onset and UHDRS motor score. Abnormalities in Parkin patients included prolonged central motor conduction time (CMCT), while thresholds and cortical inhibitory activity were normal. Asymptomatic carriers had increased motor thresholds and abnormal inhibitory measures (SICI recruitment) while CMCT was normal. Conclusion: We conclude that TMS may be a potential biomarker for neurodegenerative genetic diseases: 1) to detect changes early in the disease course and to monitor disease progression; 2) to help differentiating between clinically similar diseases on the basis of certain electrophysiological patterns; and 3) to give insight into underlying mechanisms of the disorders studied. Our findings suggest the potential for future research.

Huntington's disease (HD) is an inherited neurodegenerative disorder. An intriguing observation in the HD literature is that simple psychomotor tasks with low rather than high demands on attention and executive function are the most sensitive markers oflongitudinal chance in HD, and most consistently distinguish pre-manifest carriers of the HD mutation from non-carriers. It has been suggested that this reflects in HD a breakdown in the ability to execute simple tasks automatically. This hypothesis was explored in a series of studies that examined the performance of HD patients and pre-manifest carriers of the HD mutation on a variety of psychomotor tasks using dual-task and procedural learning paradigms. On a simple 'dual-task' paradigm, in which patients were required to perform a simple paced finger-tapping task with one or both hands, patients with early HD showed increased tapping variability and reported greater subjective difficulty for the bimanual 'dual-task' condition compared with controls, suggesting that the simple finger-tapping action was not executed automatically but rather, placed greater demands on HD patients than controls. On a visually cued reaching task in which stimuli were ordered in a highly simple, repeating sequence, HD patients showed slow and minimal motor-skill improvement and, despite extensive practice, failed to reach the level of rapid, efficient performance associated with skill automisation. In a further study it was shown that a subset of patients, who had greater cognitive impairment, failed to acquire any knowledge of the repeating sequence. It was suggested that, in these patients, the performance of the reaching task itself was so demanding that patients lacked the attentional resources required to abstract the sequential pattern. Subtle indications of a breakdown in automaticity could also be detected in premanifest carriers of the HD mutation (P-HD). On a paced finger-tapping task, P-HD participants exhibited greater performance variability, the degree of which was correlated with estimated proximity to disease onset. In a subset ofP-HD participants who were estimated to be close-to-onset, variability was greater still and was disproportionately increased by the performance of a secondary cognitive task, indicating impaired ability to execute the tapping task without directed attention. Taken together, the studies suggest that a breakdown in automaticity is a fundamental feature of HD, reflecting progressive striatal degeneration. The findings have implications for the understanding of attentional impairment in HD. If simple tasks cannot be automated then they will necessarily place greater demands on attentional resources in HD.

Anti-neuronal antibodies have been conclusively shown to be pathogenic in a handful of neurological disorders but their value in diagnosis is undeniable. Nevertheless, there are instances where diseases thought have an immunological component do not have clear antibody responses associated with them This thesis aimed to identify anti-neuronal antibodies and characterise their antigens in paraneoplastic and post-infectious neurological disorders. With regard to the latter. interest in the occurrence of anti-neuronal responses in post-streptococcal neurological disorders has culminated in the identification of four candidate auto-antigens. Recombinant forms of these proteins were produced and the frequency of an antibody response in patients determined. In addition, the antigen recognised by antibodies in the serum of patients with post-infectious opsoclonus-myoclonus was characterised using protein purification techniques and the frequency of an antibody response determined. Finally, a bacteriophage expression library was employed to study a novel antibody in a patient w ith paraneoplastic disease. Our findings were unable to provide support for an antibody response against the candidate auto-antigens in post-streptococcal disease. However we were able to characterise two target antigens, one in post-streptococcal opsoclonus-myoclonus and one in paraneoplastic neurological disease. Both antigens are thought to have specific roles in the nervous system and have provided interesting opportunities for further research into there roles in neuronal, and in the case of paraneoplastic disease, tumour biology. Further investigation is required to determine the importance of the antibody response in both post-infectious and paraneoplastic disease.

Objectives: Assess long term efficacy and safety of occipital nerve stimulation (ONS) for medically intractable chronic cluster headache (CCH) and hemicrania continua (HC). Assess the feasibility of a micro neurostimulator ("bion") for providing ONS. Investigate if improvement in headache during ONS is opioid and/or GABAA receptor mediated. Background: ONS has been reported to be effective for patients with medically intractable CCH and HC in small open label studies but more data on is long term efficacy and safety is required. ONS is currently performed by implantable pulse generators (IPGs) with electrodes attached. The mechanism of action of ONS is poorly understood, with very few studies performed. Methods: Three studies: Retrospective assessment of 14 patients with medically intractable CCH implanted with an IPG and bilateral electrodes. Prospective, crossover study for six patients with HC implanted with a bion ipsilateral to their HC. Prospective, double blind, randomised, controlled, crossover study of opioid and GABAA receptor mediated mechanisms for three HC patients with ONS. Naloxone and flumazenil drugs were used. Results: At median follow up of 17.5 (range 4-35) months, ten medically intractable CCH patients improved by 20-95% and nine of these recommend ONS. No patients were worse. Adverse events of concern were lead migrations and battery depletion. At median follow up of 13.5 (range 6-21) months, five HC patients improved by 30-95% and one was worse by 20%. Five patients recommend ONS. Mild adverse events occurred, mostly consisting of transient over-stimulation. No consistent changes in pain severity occurred at times related to naloxone and flumazenil administration for any group. Conclusions: ONS provides a safe therapeutic option for some patients with medically intractable CCH and HC. Opioid or GABA mediated ONS mechanisms were not demonstrated. Larger studies are required to further our understanding of ONS.

The causal association of variant Creutzfeldt-Jakob disease (vCJD) with bovine spongiform encephalopathy has raised significant concerns of a human epidemic. prolonged disease incubation, the possibility of asymptomatic carrier states, and the recognition of a transfusion-associated vCJD pose further public health concerns. Development of diagnostic tests is challenging as prions appear devoid of nucleic acid and comprised principally of an abnormal isoform (PrPSc) of host encoded prion protein (PrPC) to which there is immune tolerance. Detection of PrPSc can be used diagnostically but reliable detection in blood has yet to be achieved. Novel enrichment strategies have been developed for assaying PrPSc in clinical samples obtained from patients with vCJD and other forms of human prion disease. Firstly, a protocol for selective precipitation of PrPSc using sodium phosphotungstic acid has been developed. A second strategy involves optimisation of an immunoprecipitation protocol to detect PrPSc using a novel monoclonal antibody (ICSM33) with selectivity for disease-associated isoforms of PrP. This method obviates the need for protease pre-treatment of tissues and has been successfully applied to detection of PrPSc in vCJD brain homogenate and peripheral tissue samples. The method has also been successful in diagnosing other forms of human prion disease where conventional assays have failed. Both enrichment techniques have been optimised for recovery of PrPSc from up to 8mls of blood spiked to a dilution of 100ml 10% w/v vCJD brain. Assay of blood from patients with symptomatic vCJD using these techniques yields no immune signal; extrapolation from assays spiked with infected mouse brain dilution series suggests that levels of PrPSc present in symptomatic vCJD blood must represent levels of infectivity less than 2500 LD50 units ml-1. The techniques developed have immediate diagnostic applications and coupling with other high sensitivity methods may provide the increase in sensitivity required for a blood-based assay.

Alzheimer's disease (AD) is a multifactorial neurodegenerative disorder characterized by progressive cognitive impairments (Selkoe, 2001). Hippocampal place cells are a well understood candidate for the neural basis of one type of memory in rodents; these cells identify the animal's location in an environment and are crucial for spatial memory and navigation. This PhD project aims to clarify the mechanisms responsible for the cognitive deficits in AD at the hippocampal network level, by examining place cell physiology in a transgenic mouse model of AD. I have recorded place cells in tg2576 mice, and found that aged (16 months) but not young (3 months) transgenic mice show degraded neuronal representations of the environment. The level of place cell degradation correlates with the animals' (poorer) spatial memory as tested in a forced-choice spatial alternation T-maze task and with hippocampal, but not neocortical, amyloid plaque burden. Additionally, pilot data show that physiological changes of the hippocampus in tg2576 mice seem to start as early as 3 months, when no pathological and behavioural deficits are present. However, these changes are not obvious at the neuronal level, but only at the hippocampal network level, which represent hippocampal responses to environmental changes. Place cell recording provides a sensitive assay for measuring the amount and rate of functional deterioration in animal models of dementia as well as providing a quantifiable physiological indication of the beneficial effects of potential therapies.

Migraine pathophysiology is thought to involve activation of the trigeminal fibres which innervate dural structures. The nociceptive inflow from the meninges is relayed to the trigeminocervical complex (TCC), before ascending to higher brain areas, including the thalamus. Glutamate is implicated in the transmission of the nociceptive information and thus an increased understanding of the nature and effects of glutamate receptors activation has major implications in migraine pathophysiology and treatment. Here the role of kainate receptors, a member of the ionotropic glutamate receptors subfamily, was investigated in relaying sensory information upon activation of the trigeminovascular system. In order to study the role of kainate receptors on the periphery, we used the neurogenic dural vasodilation (NDV) model, in which electrical stimulation of the dura mater causes reproducible vasodilation, due to calcitonic gene-related peptide (CGRP) release. In this set of experiments kainate receptor activation but not blockade was effective in inhibiting NDV. Vasodilation induced by systemic administration of CGRP was not changed by administration of a kainate receptor agonist. In the TCC, local application by microiontophoresis of a selective kainate receptor antagonist on second order neurons which were excited by meningeal electrical stimulation, caused dual effects; 50% of the neurons tested were inhibited, whereas in a second subpopulation, activation in response to meningeal stimulation was facilitated. However, in all neurons tested, post-synaptic activation in response to kainate receptor agonists application was selectively inhibited. Microiontophoretic ejection of a kainate receptor antagonist in the ventroposteromedial thalamus (VPM) was able to inhibit cell firing in response to dural stimulation, as well as post-synaptic firing in response to kainate receptor activation. Both effects were reversed when the kainate receptor antagonist was co-ejected with a 5-HT1B receptor antagonist. We also carried electrophysiology studies in both the TCC and the VPM nucleus in order to compare the effects of the clinically active kainate receptor antagonist LY466195. Systemic and local application of LY466195 was able to inhibit cell firing in response to dura mater stimulation in both the TCC and VPM nucleus. Moreover, further to the kainate binding, a significant action of the compound on N-methyl-Daspartate receptors was observed.

Studies of the mechanisms underlying memory formation have defined central roles for CRE-dependent gene expression, which is mediated by the transcription factor CREB and the coactivator CBP. CBP creates a bridge between CREB and the basal transcriptional machinery and acetylates histones, which induces chromosomal changes and results in loss of chromosomal repression. This allows successful transcription of the underlying genes needed for synthesis of proteins underlying memory formation. CBP has been linked to neurodegenerative diseases and cerebral CBP levels were shown to reduce in mice lacking functional presenilins (PSs), a class of enzymes that has been associated with Alzheimer's Disease (AD). In this thesis it is shown that WT PS1 stimulates the transcriptional activating ability of CBP and its close homolog p300, whereas an Alzheimer's disease- associated N terminal mutant of PS1 did not produce this effect. Interestingly, PS1 C terminal mutants produced a reduction in CBP transcriptional activating ability, compared to control levels. Additionally, we showed that wild type PS1 increases the endogenous CBP level. Moreover, an increase in CBP endogenous levels was noted when the cells were transfected with the -M146L N-terminal mutant of PSI. However, these levels were still significantly lower when compared to cells transfected with wild type PSI. We were also able to show that knockdown of endogenous PSI leads to a decrease in endogenous CBP levels and a decrease in CBP activity. Hence, PSI can affect both the level and the activity of CBP. In addition, the activation of CBP by WT PSI involves the PI 3-kinase, p38 MAP kinase and p42/p44 MAP kinase pathways and targets primarily the C terminus of CBP. It is also shown that the effect of wild-type PSI is dependent on the histone acetyltransferase activity (HAT) of CBP. Moreover, it was demonstrated that WT PSI, but not its M146L mutant, could increase the promoter activity of c-fos, a CBP HAT dependent target gene. Additionally, we showed that application of the histone deacetylase inhibitor, TSA, rescued the long-term potentiation and long-term memory defects shown by APP/PS1 mutant mice. Moreover, it was shown that the acetylation level of histone H4 in APP/PS1 mice is lower than that of WT littermates and that TSA injection restores the acetylation of these histones. This is the first study to identify AD as a disease of epigenetic etiology and suggests that enhancing histone acetylation may have potential for the treatment of AD.

Huntington’s disease (HD) is a purely genetic neurodegenerative disorder affecting approximately 1 in 10,000 people. It is most commonly associated with excessive involuntary movement, or chorea, combined with varying degrees of other motor, psychiatric and cognitive disturbances. Identification of the mutation in the HD gene prompted the generation of several transgenic mouse models. HD is but one of a family of at least 9 triplet repeat disorders, all of which exhibit protein aggregation by a similar mechanism. The understanding of one disease is therefore of importance to the understanding of them all. This thesis aims to be a comprehensive comparative study of three very different mouse models of HD elucidating the pathological changes that precede and accompany the disease process. The work described in this thesis presents a detailed account of a longitudinal study of the pathological changes that occur within the brains of founder generations of mice transgenic for exon 1 of the HD gene, containing a highly expanded CAG repeat, the R6 lines. I have determined the intracellular sites for deposition and accumulation of the mutant protein huntingtin (htt), within both the neurons and glia of the central nervous system. The progressive accumulation of additional proteins within these aggregates has been described. The temporal evolution and spatial distribution of the neuronal intranuclear inclusion (NII) was determined using both immunohistochemical and morphometric analyses. The cellular consequences resulting from the aggregation of mutant htt were also investigated. I have conducted a detailed morphometric analysis of neurones within the cerebral cortex, striatum and cerebellum throughout the period of protein deposition, until the eventual degeneration of these cells. The dendritic and somal changes resulting from the cellular disruption associated with these NII are also described. In a further series of experiments I have investigated the changes that occur in a novel model of HD, namely the conditional, doxycycline inducible double transgenic mouse, HD94 model. It was interesting to find that the same construct when differently manipulated in two mouse lines can produce such contrasting symptoms and pathology. This was highlighted by the comparison of immunohistochemical and morphometric analyses between the HD94 and the R6 lines, where the pattern of mutant protein deposition was found to vary significantly. Lastly I have studied a more genetically accurate murine model of HD, the HD80 ‘knock-in model’. These mice develop a pathology broadly similar to that of the R6 lines but markedly different to that of the HD94, and over a much longer time frame This detailed comparative analysis of the molecular and cellular pathology of three transgenic mouse models of HD provides new insights identifying novel and unique neuropathology and suggests new approaches for therapeutic treatments for this disease.

The juvenile form of Batten Disease (JNCL) is an autosomal recessive neurodegenerative disease caused by mutations in the CLN3 protein, an integral membrane protein of unknown function. The hallmark of JNCL is the accumulation of storage material in lysosomes consisting predominantly of subunit c of mitochondrial ATP synthase. CLN3 protein is present primarily in endosomes and lysosomes in non- neuronal cells, whereas in neuronal cells it is also reported to be in presynaptic vesicles and along the neurites. The current best evidence is that it traffics through the endoplasmic reticulum and Golgi apparatus to the endosome and lysosome, sorted by a dileucine motif which binds the adaptor complexes AP-1 and AP-3. The effects of loss of CLN3 protein function on membrane trafficking were investigated in the Cln3Aexl'6 knockout mouse model and in human cell lines deficient for CLN3 due to siRNA-mediated gene silencing. Morphological analysis and immunostaining of CLN3-deficient fibroblasts and neurons revealed the presence of normal lysosomes but also an accumulation of aberrant organelles of autophagic origin. Endocytic tracer analysis indicated that the aberrant organelles accumulate due to a defect in fusion between autophagosomes and lysosomes. Phagocytic defects involving a failure to recruit late endocytic markers were also defined. Together these findings indicate a similar defect in both these cellular pathways, involving defective maturation of autophagosomes and phagosomes which is accompanied by accumulation of undigested material in CLN3-deficient cells. As well as impairments in the mechanism of endosome fusion with autophagosomes and phagosomes, defects in an additional cellular mechanism were detected in CLN3-deficient cells. The route of protein trafficking from the trans-golgi network (TGN) to endosomes is interrupted after CLN3 gene silencing, resulting in retention of the cation-independent mannose-6-phosphate receptor in the TGN accompanied by defective processing of lysosome hydrolases. This investigation of the cellular pathology underlying Batten disease reveals a role for CLN3 in two distinct but related cellular pathways: autophagyc and phagocytosis and it also plays a role in TGN to endosome trafficking. This contributes to the understanding of the cellular phenotype underlying CLN3/Batten disease and most likely other forms of neuronal ceroid lipofuscinosis.