Parkinson’s disease (PD) is a common neurodegenerative disorder which usually occurs sporadically, but in 5-10% of cases is genetically inherited. Many of the causative mutations underlying these familial forms have been identified, and studying the functions of the proteins encoded by these genes has highlighted several common pathogenic mechanisms. In particular, mitochondrial dysfunction has been shown to play a major role in the pathogenesis of both familial and sporadic forms of the disease. This thesis investigates the importance of the proteins encoded by these PD-associated genes in mitochondrial function, focusing on two proteins in detail. Firstly, live cell imaging techniques were used to investigate the mitochondrial physiology of cells derived from HtrA2 knockout mice, an animal model in which the loss of a mitochondrial protein gives rise to severe neurodegenerative phenotype. Similar approaches were then applied to an RNAi screen to investigate the effects of other PD-associated genes on mitochondrial function, while a separate study specifically investigated the putative mitochondrial localisation and function of the PD-associated protein Fbxo7. Results from these studies revealed that HtrA2 has an important role in maintaining the function of the ATP synthase, as HtrA2 deficient cells exhibited a severe uncoupling combined with an increase in proton translocation through the ATP synthase but a reduction in ATP synthesis. Furthermore, Fbxo7, a protein with no reported link to the mitochondria, was found to partially localise to the mitochondria under basal conditions and to further accumulate on depolarised mitochondria. Further work indicated that this protein interacts with two other PD proteins, PINK1 and Parkin, and together with these proteins functions in a previously described pathway to mediate the selective autophagic clearance of damaged mitochondria. These results contribute to our understanding of the functions of these proteins and further emphasise the relevance of mitochondrial dysfunction in PD pathogenesis.

Estimation of uncertainty is an important aspect of perception and a prerequisite for effective action. This thesis explores the implementation of uncertainty estimation as precision modulation within a predictive coding hierarchy, optimised within a neurbiologically-plausible message-passing scheme via the minimisation of free-energy. This thesis consists of six chapters. The first presents a new model of a classic visual illusion, the Cornsweet illusion, which demonstrates that the Cornsweet illusion is a natural consequence of Bayes-optimal perception under the free-energy principle, and demonstrates that increasing contrast can be modelled by increasing signal-to-noise ratio. The second chapter describes dynamic causal modelling of EEG data collected from participants viewing the Cornsweet illusion, demonstrating that a reduction in precision, or superficial pyramidal cell gain, in lower visual hierarchical levels, is sufficient to explain contrast-dependent changes in ERPs. The third describes a model of a simple attentional paradigm – the Posner paradigm – recasting attention as the optimal modulation of precision in sensory channels. The fourth describes an MEG study of the Posner paradigm, using Bayesian model selection to explore the role of changes in backwards and modulatory connections and changes in local superficial pyramidal cell gain in producing the electrophysiological and behavioural correlates of the Posner paradigm. The fifth chapter recasts the Posner paradigm in the motor domain to investigate the level (intrinsic vs. extrinsic) of precision modulation by motor cues. The sixth describes a new model of sensory attenuation based on using precision modulation to balance the imperatives to act and perceive. I hope to demonstrate that precision modulation within predictive coding hierarchies, under the free-energy principle, is a flexible and powerful way of describing and explaining both behavioural and neuroimaging data.

Oscillatory brain activity is a ubiquitous feature of neuronal dynamics and the synchronous discharge of neurons is believed to facilitate integration both within functionally segregated brain areas and between areas engaged by the same task. There is growing interest in investigating the neural oscillatory networks in vivo. The aims of this thesis are to (1) develop an advanced method, Dynamic Causal Modelling for Induced Responses (DCM for IR), for modelling the brain network functions and (2) apply it to exploit the nonlinear coupling in the motor system during hand grips and the functional asymmetries during face perception. DCM for IR models the time-varying power over a range of frequencies of coupled electromagnetic sources. The model parameters encode coupling strength among areas and allows the differentiations between linear (within frequency) and nonlinear (between-frequency) coupling. I applied DCM for IR to show that, during hand grips, the nonlinear interactions among neuronal sources in motor system are essential while intrinsic coupling (within source) is very likely to be linear. Furthermore, the normal aging process alters both the network architecture and the frequency contents in the motor network. I then use the bilinear form of DCM for IR to model the experimental manipulations as the modulatory effects. I use MEG data to demonstrate functional asymmetries between forward and backward connections during face perception: Specifically, high (gamma) frequencies in higher cortical areas suppressed low (alpha) frequencies in lower areas. This finding provides direct evidence for functional asymmetries that is consistent with anatomical and physiological evidence from animal studies. Lastly, I generalize the bilinear form of DCM for IR to dissociate the induced responses from evoked ones in terms of their functional role. The backward modulatory effect is expressed as induced, but not evoked responses.

This work is about understanding the dynamics of neuronal systems, in particular with respect to brain connectivity. It addresses complex neuronal systems by looking at neuronal interactions and their causal relations. These systems are characterized using a generic approach to dynamical system analysis of brain signals - dynamic causal modelling (DCM). DCM is a technique for inferring directed connectivity among brain regions, which distinguishes between a neuronal and an observation level. DCM is a natural extension of the convolution models used in the standard analysis of neuroimaging data. This thesis develops biologically constrained and plausible models, informed by anatomic and physiological principles. Within this framework, it uses mathematical formalisms of neural mass, mean-field and ensemble dynamic causal models as generative models for observed neuronal activity. These models allow for the evaluation of intrinsic neuronal connections and high-order statistics of neuronal states, using Bayesian estimation and inference. Critically it employs Bayesian model selection (BMS) to discover the best among several equally plausible models. In the first part of this thesis, a two-state DCM for functional magnetic resonance imaging (fMRI) is described, where each region can model selective changes in both extrinsic and intrinsic connectivity. The second part is concerned with how the sigmoid activation function of neural-mass models (NMM) can be understood in terms of the variance or dispersion of neuronal states. The third part presents a mean-field model (MFM) for neuronal dynamics as observed with magneto- and electroencephalographic data (M/EEG). In the final part, the MFM is used as a generative model in a DCM for M/EEG and compared to the NMM using Bayesian model selection.

The ventral fronto-parietal network has been considered to play a crucial role in reorienting attention towards significant environmental events, while the dorsal system is thought to be dominant in controlling goal-directed behaviour (Corbetta and Shulman 2002). I begin by reviewing literature which suggests this distinction may not be so clear cut and suggest my own scheme which takes into account this evidence (Singh-Curry and Husain 2009). Specifically, ventral areas, particularly the right inferior parietal lobe (IPL), appear to be activated by tasks involving sustained attention, responding to salient taskrelevant events, detecting novel stimuli and switching between tasks. Accordingly, I hypothesise that the right IPL may play a crucial role in reconfiguring behaviour between a task-engaged state and a more exploratory mode of functioning, which permits the identification of potentially important novel events. The first few chapters of my thesis aimed to test this hypothesis by examining attention deficits in stroke patients with hemispatial neglect, the syndrome which frequently occurs following damage to the right IPL. These patients were shown to have difficulty sustaining attention over time, even when no spatial shifts of attention were required. This deficit in sustained attention was particularly evident for stimuli of lower perceptual salience. More importantly, however, these deficits were found to interact with each other, as well as the direction of spatial attention, suggesting that these functions may be dependent on an interrelated brain network. Consistent with this notion, the results of lesion-symptom analysis indicated that the Right IPL and ventral attention network appears to be crucial in the mediation of all of these processes, including the processing of novel stimuli, supporting my hypothesis. The detection of novel events has also been found to activate the midbrain dopaminergic system (Bunzeck and Duzel 2006), while the principal pathological feature of Parkinson’s disease (PD) is degeneration of these neurons (Hornykiewicz 1998). Although PD is traditionally considered a disorder of movement, more recently it has been recognised that there may be associated cognitive deficits, including disorders of impulse control (Weintraub 2008). At present, however, the factors which predispose some individuals with PD to develop such problems are unclear. Accordingly, in the second part of my thesis, I examined novelty processing and risktaking behaviour in PD in order to identify subgroups which may be particularly vulnerable to developing impulse control problems. In addition to PD patients with impulse control disorders (ICD), those who were classified as akinetic-rigid, as opposed to tremor dominant – without ICD – were found to process novelty more quickly than nonnovel perceptually salient stimuli, unlike tremor dominant PD patients. Novelty seeking was found to be associated with relative preservation of the mesolimbic dopaminergic system in patients without ICD, while increased risk-taking was associated with preservation of the mesolimbic system in ICD patients. Mesolimbic sparing, in addition to the akinetic-rigid motor phenotype of PD may therefore increase susceptibility to impulse control problems in PD.

Convulsive status epilepticus (CSE) is the most common neurological emergency in childhood. Approximately half of new CSE cases occur in children with no perceptible neurological priors (Chin et al., 2006). Prolonged febrile seizures (PFS), a type of CSE that occurs in neurologically normal children, have been retrospectively linked to temporal lobe epilepsy with hippocampal sclerosis. Imaging studies have revealed hippocampal abnormalities soon after PFS, yet, no neuropsychological study to date has investigated these children close to the time of insult. The present thesis investigated the effects of CSE on child development within a month of the incident and, subsequently, a year onwards. The first aim of this thesis was to investigate the effects of CSE on developmental functions using standardized assessments. The second aim was to examine children with PFS for signs of hippocampal dysfunction close to the time of incident. We hypothesized that aetiology would largely influence outcome in our CSE cohort, and, that children with PFS would reveal deficits in a delayed recognition paradigm that is thought to tap onto hippocampal processes. Eighty children were seen a mean of 38 days following CSE (34 PFS) and 50 children (24 PFS) were re-assessed a year onwards. At baseline neuropsychological impairments were evident in children following CSE associated with a PFS, as well as, children following CSE associated with other aetiologies (non-PFS), albeit, these were more pronounced in the non-PFS group. Moreover, in line with our hypothesis, the PFS group revealed deficits in a task of incidental recognition memory alluding to the presence of hippocampal dysfunction in this group. A year onwards deficits were still apparent in the two patient groups, although, the PFS group had shown some improvement on a number of measures. The implications of these findings for our understanding of CSE are discussed in this thesis.

There is increasing evidence for "hypoxia-like" conditions in some inflammatory multiple sclerosis (MS) lesions raising the possibility that the tissue is deficient in energy. Accodingly, mitochondrial abnormalities have been described in some MS tissue, including reduced mitochondrial gene transcripts in MS motor cortex and decreased mitochondrial complex IV expression. To explore the role of mitochondrial defects in neuroinflammatory lesions, experimental inflammatory lesions were induced by the intraspinal injection of lipopolysaccharide in rats under general anaesthesia. The tissue was snap-frozen for histochemical assessment of mitochondrial complex II and/or IV activity at various intervals (1-28 days) post injection: the activity of the complexes was compared with porin expression or complex IV subunit I expression (presence of mitochondria or protein respectively). The oxygen concentration within the dorsal column was determined at 24 hours after lesion induction in vivo using an oxygen-sensitive optical probe. EMG potentials were also recorded at the foot dorsum in response to stimulation of the sciatic nerve. Presence of reactive oxygen and nitrogen species were examined by immunohistochemistry and by the fluorescent marker dihydroethidium (DHE) in vivo. Complex IV activity within the motor neurons was decreased 1 day after injection, further decreased by day 2, and returned to pre-injection baseline by day 5. Decreased complex IV activity exactly coincided with a temporary reduction in motor neuron excitability as measured by H reflex and F wave amplitude, which was maximal at day 2. The oxygen concentration within the dorsal columns at the site of the LPS lesion was significantly elevated when compared with the saline-injected and naïve control animals. DHE fluorescence revealed that superoxide production was increased at the lesion site and immmunohistochemistry revealed oxidative stress to DNA, lipids and protein. LPS-induced neuroinflammation results in a reversible and coincident decrease in both complex IV activity and motor neuron excitability, which was strengthened by the result that intraspinal LPS-injections causes hyperoxia, presumably from mitochondrial dysfunction. Nitric oxide has been implicated in neuronal dysfunction and although a causal relationship has not been established, observations support an interpretation that inflammation-mediated NO causes mitochondrial damage, increased oxygen concentration and formation of reactive oxygen species (ROS). These actions illustrate a vicious circle where ROS induce further damage, which results in energy deficiency displayed by reduced neuronal excitability and neurological deficits. The observations are consistent with energy deficiency and reduced function in neuroinflammatory lesions similar to those found in MS.

The autonomic airway ganglia that comprise the intrinsic lung innervation are derived from vagal neural crest cells (NCC) that migrate tangentially from the foregut into the embryonic lung buds. The aim of this PhD thesis was to investigate the mechanisms that direct NCC from the foregut into the lungs and that subsequently influence their development. A novel combination of cell labelling, using Wnt1Cre:Rosa26YFP double transgenic reporter mice, and Optical Projection Tomography (OPT) imaging was employed to visualize lung innervation. Results showed that NCC migrated into the lungs from the esophagus early in development, accumulated around the epithelial tubules and differentiated into an extensive network of neurons and glial cells. Next, chick intraspecies grafting was used to test the developmental potential of lung and gut NCC. Results showed that when NCC from the gut were back-grafted into the early migration pathway these cells colonised both the lungs and gut, indicating that vagal NCC are not prespecified to colonise either organ and are thus likely to respond to common signalling cues. When potential cues were tested in organotypic lung culture, NCC migrated towards sources of the RET (Rearranged during Transfection) ligand GDNF (Glial-cell-line-derived neurotrophic factor), suggesting that the RET signalling pathway is involved in NCC colonisation of the lung. However, examination of RET mutants indicated that this pathway is not necessary for NCC colonisation of the lung, since lung innervation in Ret-/- mouse embryos was similar to controls. Lung innervation was further examined in several mouse mutants with known NCC defects. Intrinsic ganglia formation was altered in Sox10Dom and Tbx1 mutant mouse lungs, implicating a role for vagal nerve projections in guiding NCC within the lung. Together these studies have described the development of intrinsic lung innervation in the avian and mammal and examined multiple mechanisms underlying NCC development within the lung.

The aim of this Thesis was to characterise the behaviour and interaction of primary muscle derived cells (MDCs) and motoneurons within a collagen-based 3D in vitro culture system. Cells cultured under uniaxial tension within 3D collagen matrices are known to selforientate along the lines of principle strain. In the case of skeletal muscle cells, this leads to the formation of aligned myotubes, thereby generating cultures which more closely recapitulate the architecture of in vivo muscle. Since maturation of muscle in vivo is dependent on functional innervation, integration of this model with a physiologically correct neural input would further improve both the accuracy and complexity of the in vitro construct. Furthermore, reliable neuromuscular junction formation in 3D culture could have substantial benefits for the study of neuromuscular disease and the testing of novel therapeutic agents. The behaviour of primary rat MDCs within an established collagen-based 3D culture system was optimised and subsequently characterised. A comparison of this model to conventional 2D cell culture techniques was carried out using immunohistochemical and PCR analysis. Investigation of myogenin expression levels over a three week culture period in both 2D and 3D found no significant differences between the two systems, indicating a conserved ability for MDC differentiation in both models. Immunohistochemical data illustrated the alignment of uniaxial myotubes in 3D compared with randomly orientated and branched myotubes in conventional culture, demonstrating the improved biomimicity of myotubes developed in 3D and under directional tension. The presence of motoneurons within the 3D co-culture was found to promote maturation of the MDCs as indicated by levels of macroscopic construct contraction and by quantitative PCR analysis. Co-localisation of pre- and post- synaptic markers in culture indicated the presence of putative synaptic contacts within the model. The model presented in this Thesis represents a step forward in the development of physiologically accurate in vitro models of skeletal muscle, which may help in future investigations of skeletal muscle development, physiology and pathology.

The aim of this thesis is to explore the potential of quantitative imaging mark¬ers derived from diffusion-weighted MRI (DW MRI) in the spinal cord to char¬acterise healthy white matter pathways and provide sensitivity to axonal dam¬age, regeneration and collateral sprouting in spinal cord disease. With new innovative treatment strategies emerging for spinal cord patholo¬gies such as spinal cord injury and Multiple Sclerosis, there is a need for new in-vivo biomarkers that can be specific to structural and functional changes and their underlying mechanisms on a microscopic scale. DW MRI has the potential to quantifying those microstructural characteristics beyond the scale of conventional MRI. In the first part of this dissertation I investigate Diffusion Tensor Imaging (DTI), which is the most established DW MRI analysis technique in clinical practice. In two studies we assess DTI in the context of spinal cord imaging. In the first experiment I show that DTI is sensitive to the presence of collateral fi¬bres, e.g., at inter-vertebral level where peripheral nerves enter the spinal tract. In the second experiment I propose a new method for reducing partial volume effects on whole cord DTI measurements, which is specifically tailored for the imaging and analysis challenges in the cord. The second part of this thesis comprises two studies of q-space imaging (QSI) in healthy controls. In theory, QSI offers a more comprehensive descrip¬tion of the diffusion process, but is challenging to set up on a clinical MRI scanner. I present here two QSI protocols, set up for two different scanners with different gradient hardware, receive coils and software limitations. For the first time we perform a systematic study of QSI that assesses the reproducibility and specificity to different white pathways in-vivo in the cervical cord within a group of healthy volunteers. Both studies show superior reproducibility of QSI over conventional analysis, although the results of using QSI parameters to distinguish individual white matter tracts in the cord were inconclusive. The third part of this thesis describes a new imaging method protocol based on the ActiveAx optimisation framework. It uses a complex multi- compartment model, which relates DW MRI data to microstructural parame¬ters like axon diameter and density. I design a new orientation aware method based ActiveAx, which incorporates the known fibre structure of the spinal cord. In a first step I validate the approach in in a post-mortem cervical spinal cord sample of a velveteen monkey. I then demonstrate clinical feasibility and good reproducibility of the new protocol for in-vivo human studies, using the corpus callosum as a preliminary model system for structures with uni¬directional fibre architecture. Finally I present first estimation results of axon diameter and density of the cervical spinal cord in-vivo in a healthy control that agree with the findings in the ex-vivo monkey spinal cord sample.