No description available.

No description available.

Organisers are specialised groups of cells that non-autonomously pattern adjacent cell populations. The dorsal midline, or roof plate, of the developing CNS is one such organiser and is required for the specification of specific subtypes of dorsal neurons. Organisers often comprise boundaries between molecularly distinguishable compartments, however the roof plate does not fit with this model; for the most part it constitutes a narrow strip of cells that separate two molecularly indistinguishable compartments (the two halves of the neural tube), but at certain anteroposterior locations, such as the hindbrain, it is expanded to form a thin epithelium that tents over a ventricle. Using chick embryos, I have investigated a hypothesis that reconciles the roof plate with this emergent model, in which the organiser properties of the roof plate are invested in its boundaries. Using in vitro co-culture, I show that the gdf7- positive roof plate boundary and its signalling properties can be regenerated in roof platederived tissue at the interface between hindbrain roof plate epithelium and neuroepithelium. Further, this gdf7-positive boundary is required for the expression of cath1, which marks the dorsal-most pool of neural progenitors in the hindbrain. Many organisers require Notch signalling and downstream Hairy/ Enhancer of split (Hes) transcription factors for their formation or maintenance. Using electroporation of the hindbrain roof plate epithelium – neuroepithelium boundary, I find that Delta-Notch signalling is sufficient to convert cells from a roof plate epithelial to a roof plate boundary fate. Further, correct levels of expression of chairy2 (a hes1 homologue) are required for the maintenance of the roof plate boundary. Finally, I show that the roof plate boundary is a bidirectional signalling centre that not only patterns adjacent neuroepithelium, but is also required for the differentiation of choroid plexus epithelium from roof plate epithelium.

Development of retinotopy is an area of longstanding investigation, and it is clear that this development is influenced by both molecular and activity-based cues. The interaction between these different cues, however, is unclear: conventionally molecular guidance cues set the basic topography, which is then refined by activity-based mechanisms. A number of computational models have investigated this interaction. However, due to the limited availability of quantitative data it has not been possible to test whether the models can account for the precise developmental sequence of retinotopic map formation. This thesis provides such quantitative anatomical data about the spatiotemporal sequence of development of precision in the retinocollicular projection in the mouse. This characterisation has been done by utilising retrograde transport of fluorescent microspheres administered by discrete injections into the SC and recovered from retinal flat-mounts. To enable quantification of topographic precision, retinae are standardised using an algorithm that permits retinal outlines to be stitched back together into a sphere and cell locations plotted in spherical coordinates, which enables mapping across multiple animals and facilitates comparison between different age-points, genotypes and potentially different species. The anatomical description reveals the precise dynamics of retinocollicular projection refinement, as projection refines from having almost no precision along the SC AP axis at birth to having topographically accurate, albeit imprecise, but symmetrical precision by the end of the first postnatal week. The data, moreover, reveals that the period of refinement extends significant beyond the first postnatal week and refinement is even seen after the third postnatal week. It is also found that the early refinement is crucially dependent on specific patterns of spontaneous retinal activity, since altering the activity-patterns by removing the nAChR-|32 subunit completely disrupts the refinement normally seen in the first postnatal week. Furthermore, it is found that refinement does occur in these animals after the first postnatal week, but it does not rescue the phenotype.

The BBB is a biological firewall that carefully regulates the cerebral microenvironment by acting as a physical, metabolic and transport barrier to molecules bound for the brain. This selectively permeable interface was modelled in this thesis using the recently established immortalised human cerebral microvascular endothelial cell line (hCMEC/D3) to investigate interactions with endogenously and exogenously derived molecules of clinical significance. The endogenous molecules in question are the cationic amino acids (CAA) L-arginine, the precursor for nitric oxide (NO), and asymmetric dimethylarginine (ADMA), an endogenously derived analogue of L-arginine that acts as a potent inhibitor of NO production. As well as being an important vasodilator, NO has regulatory roles in the brain and on the BBB itself. Transport mechanisms utilised by L-arginine are known, but are not fully understood for ADMA – particularly so at the BBB. This is of clinical significance giving the emerging role of ADMA in many brain and cerebrovascular diseases. Understanding these transport mechanisms and other interactions of ADMA with the BBB is therefore very important for the study of disease emergence, detection and progression. We discovered in the hCMEC/D3s that high concentrations of ADMA could induce endothelial dysfunction in a BBB permeability model, leading to an increase in paracellular permeability to the paracellular marker FITC-dextran (40kDa). We also illustrated interactions of ADMA with a variety transport proteins, basing the study design on our observed L-arginine interactions. The CAA transport system y+ was heavily implicated for both molecules through the use of established inhibitors. Furthermore, the expression of CAT-1, the best known protein from this group was confirmed in the hCMEC/D3s. We also found evidence of an efflux transport system for ADMA (but not Larginine), implicating the neutral and CAA transporter ATB0,+. The protein expression of ATB0,+ was also confirmed in the cells. Intracellular ADMA was even shown to induce transstimulation of extracellular L-arginine, providing evidence for a role of ADMA in the 'L-arginine paradox', a phenomenon observed in vivo that administered L-arginine can alleviate the effects of NO reduction (such as vasoconstriction), despite there being 20-30 times the amount of L-arginine present to saturate the NO producing enzyme. In summary, our endogenous molecule findings from this thesis identify the likely transport mechanisms used by ADMA and implicate ADMA in endothelial dysfunction as well as the ‘Larginine paradox’. These data are not only important with regards to the brain, but apply to other microvascular endothelia such as those found in peripheral cardiovascular system, where ADMA remains a major area of investigation. The exogenous molecules studied during this PhD are drugs currently used to treat the second-stage of human African Trypanosomiasis (HAT). HAT is a neglected parasitic disease that continues to persist in sub-Saharan Africa. It is fatal if untreated. Recently, it has been described that eflornithine and nifurtimox combination therapy (NECT), improves the efficacy of both drugs compared to their monotherapy, although why this happens remains unclear. We hypothesised that it may be due to improved CNS delivery, although we failed to show improvements in accumulation of either eflornithine or nifurtimox with NECT or when the individual drugs were in combinations with the other anti-HAT drugs. Interestingly, the combination of eflornithine and pentamidine caused a decrease in eflornithine accumulation, implicating an unidentified pentamidine-sensitive transport system – possibly an adenosinesensitive influx transporter. The cellular influx transport mechanisms used by eflornithine has been suggested to be those used by CAA due to the structural similarity of eflornithine with the CAA ornithine and so this was studied. We revealed in the hCMEC/D3s that eflornithine had degrees of sensitivity to a variety of transport mechanisms, in which system y+ appears to be the principal influx mechanism. Similar anti-HAT drug combination therapy studies with nifurtimox were performed and also illustrated a significant interaction with pentamidine; although conversely to eflornithine we demonstrated an increase in nifurtimox accumulation as a result of nifurtimox-pentamidine combination. Previous in situ observations by our group suggested nifurtimox was a substrate for efflux transport systems at the BBB that are separate from P-gp and this too was investigated, identifying the well known drug efflux pump BCRP as the principal nifurtimox efflux transporter. With regards to exogenous molecules, we provide evidence of CAA influx mechanisms for the anti-HAT drug eflornithine and an efflux system for nifurtimox, principally involving BCRP. We also found that NECT and combination therapy of eflornithine or nifurtimox with the other anti-HAT drugs did not improve eflornithine or nifurtimox accumulation when compared to controls – except when pentamidine was combined with nifurtimox. This finding suggests that nifurtimox-pentamidine combination may improve efficacy of nifurtimox in the field. Collectively, these data further demonstrate of the suitability of the hCMEC/D3 cell line as a powerful tool for human in vitro BBB investigation across a range of study areas.

The diacylglycerol lipases (DAGL α and β) synthesise 2-AG, the major endocannabinoid (eCB) in the developing and adult brain. 2-AG acts most prominently on the CB1 and CB2 receptors and growing evidence points to a role of eCB signaling in adult neurogenesis as well as a wide range of different physiological roles in the brain. The focus of this thesis is on DAGL driven eCB signaling in neural stem cells (NSC). A role of the eCB system has previously been established in migration, proliferation, and neurogenesis. Here we investigated DAGLα and eCB signaling in the context of differentiation. The Cor-1 cells were adapted as a NSC model system and can be differentiated into glia and neurons. This differentiation was unaffected by CB signaling. While DAGLα expression is unaffected by glial differentiation, a rapid down regulation of DAGLα during neuronal differentiation was detected and we investigated the possibility of a specific degradation pathway being involved. DAGLα has a consensus motif for a putative destruction (d-) box that might target DAGL for ubiquitin-proteasome mediated degradation. No evidence was found to support that the d-box is involved in DAGLα degradation, however first indications towards degradation through the ubiquitin-proteasome pathway were identified. What is upstream of eCB signaling in NSCs remains an open question. The most likely candidates are EGF, FGF-2 and insulin and their effect on eCB signaling was investigated. Evidence was found indicating that neither EGF nor FGFR signaling is upstream of eCB signaling in Cor-1 cells. Microarray analysis indicates a potential common signaling node between EGFR and eCB signaling. mTOR was investigated in this context, but while we show that mTOR appears to be downstream the EGFR, no indication was found that it is downstream of the eCB receptors. Further experiments are needed to determine the relationship between insulin and eCB signaling. It is essential to be able to measure DAGL activity, and a surrogate substrate DAGL activity assay was adapted to Cor-1 cells in order to address what is driving eCB signaling.

The domain of the developing cerebellum in dorsal rhombomere 1 is specified by signals from two adjacent organising centres, the midbrain-hindbrain boundary (isthmus) and the roof plate of the fourth ventricle. Two major precursor pools give rise to all the cells of the cerebellum: the ventricular zone and the cerebellar rhombic lip. The rhombic lip comprises a dorsal progenitor population, defined by the expression of the transcription factor Atoh1, which gives rise to different cell types contributing to both extra-cerebellar and cerebellar populations. In chick embryos, I have investigated the timing, morphology and gene expression of successive cohorts of rhombic lip derivatives using a combination of in ovo electroporation and in situ hybridisation. Using this approach, I first derived an accurate spatiotemporal map of cell production within rhombomere 1, which helps illuminate the evolutionary mechanisms generating diversity in the structure and function of the cerebellum. I then explored how signals from the roof plate and isthmus specify distinct neuronal derivatives of the cerebellar rhombic lip at different time points. In these experiments, I show that the isthmus abuts an expanded region of Atoh1 expression from which early, extra-cerebellar, Lhx9-expressing rhombic lip derivatives are born. Through surgical manipulation of the isthmus in cultured explants and through induction of an ectopic isthmic organiser through in ovo electroporation of early genetic regional determinants (Otx2), I demonstrate that this expanded region of Atoh1 is dependent upon the isthmus for its induction and maintenance. I also demonstrate the role of the isthmic gene, Fgf8, in the maintenance of the expanded domain through in ovo electroporation of plasmids encoding Fgf8 and a dominant-negative FGF receptor. This FGF-dependent domain of Atoh1 is regulated independently from the specification of rhombic lip derivatives, suggesting a novel role for both Atoh1 and FGF signalling in rhombomere1. Analysis of triple knockout FGF receptor mutants and FGF hypomorph mice confirmed the results of dominant negative overexpression in chick. To investigate temporal factors influencing the specification of temporal cell fate, I examined the role of thyroid hormone, which is synthesised in the fourth ventricle choroid plexus adjacent to the rhombic lip. I show that pharmacological manipulations of thyroid signalling can specifically disrupt the expression of markers of late-born rhombic lip derivatives, pointing to an important role for steroid hormone receptor signalling in specifying the fate of specific temporal cohorts in the rhombic lip lineage.

Although a coupling between cerebral energy consumption and neuronal activity was initially suggested over a hundred years ago, the exact relationship remains unclear. This thesis addresses four aspects of the energy supply to the brain. First, blood flow is increased to areas where neurons are more active, and it is this blood flow increase which forms the basis of functional imaging techniques. In contrast to the common assumption that blood flow regulation occurs solely at the arteriole level, I have demonstrated experimentally that blood flow control, mediated by pericytes, also occurs in capillaries. Second, it has been proposed that glutamate transporters coordinate CNS glucose and oxygen usage: glutamate released from neurons is taken up into astrocytes, evoking glycolysis which is suggested to export lactate to power neuronal oxidative phosphorylation. I tested this hypothesis by measuring oxygen use when glial glutamate uptake was blocked, and found that this had little effect on the oxygen use evoked by neural activity, arguing against the idea that glutamate uptake and astrocyte lactate export regulate neuronal oxidative phosphorylation. Third, although it is known that most of the brain's energy is used on reversing the ion fluxes which generate action potentials and synaptic currents, in general it is unknown how the energetic resources to the brain are allotted to carry out different aspects of neuronal information processing. I constructed an energy budget for the cerebellum using the measured properties of cells in this area. This provides insight into how evolution has allocated energy at the cellular and subcellular levels to different parts of a neural computation. Finally, I estimated the amount of brain energy use expended on conscious perception of stimuli, as opposed to being used on unconscious information processing. Surprisingly, only a small percentage of brain energy is used on conscious perception.

A review of the literature indicated that the blood supply of the midbrain had been studied but little in detail. An investigation was therefore carried out on human and monkey material. Of the techniques tried, coloured gelatin, radio-opaque, and especially Indian ink injections proved most informative. 1. The generally accepted sources of blood supply were confirmed. Accepted data on the origin, variations, number and size of vessels have been presented in schematic form, restricting description of the various named vessels involved to relevant detail. 2. Superficial vascular anastomoses on the midbrain have been confirmed. Midline surface anastomoses over the tectum have been shown to exist. The extent of midbrain supply by different arteries as observed by surface inspection and in microscopic sections has been described and illustrated. 4. Differences from and agreements with previous descriptions, and the originality of certain observations,have been pointed out. 5.An absence of arteriovenous anastomoses in the midbrain pia mater has been confirmed. 6. Entirely original observations on the intramesencephalic veins have been presented. Collecting veins of the midbrain have been described. Incidence of variations affecting the basal and lateral mesencephalic veins have been presented in schematic form. 8. Anastomoses of the dorsal mesencephalic and mesencephalic-cerebellar veins have been described. Through these anastomoses the possibility of a short circuiting of the internal or great cerebral veins to the superior petrosal sinus has been suggested. 9. Capillary density studies on representative tracts and nuclei of the midbrain (and some neighbouring structures) were carried out by a dot graticule method. Data on their relative vascularities have been tabulated. 10. The results have been discussed and conclusions drawn.

The inferior colliculi (ICs) are the principal nuclei of the auditory midbrain. Each IC processes converging inputs from numerous brainstem nuclei as well as from thalamus and cortex. The ICs are interconnected in mirror image by one of the largest afferent inputs to each IC, the commissure of the inferior colliculus. There is exiguous knowledge about how each IC influences the processing of auditory information in its contralateral counterpart. This thesis investigates how one IC modulates the neural representation of sounds in the contralateral IC. To this end, I established and validated an experimental model in anaesthetised guinea pig whereby neuronal activity in one IC was selectively and reversibly deactivated. Cryoloop cooling produced temperature changes sufficient to deactivate spiking activity in the dorsal half of one IC, whilst leaving other centres in the auditory pathway unaffected. Single units were recorded in one IC before, during and after deactivation of the other IC. The characteristic frequency (CF) of IC neurons was unaffected during cooling, but the threshold of the population was raised. The area of non-V-shaped frequency response areas (FRAs) changed more than V-shaped FRAs. Differential changes were also observed in the firing rate of units with different temporal response patterns. Onset responders increased their firing rate whilst the firing of Chopper units was reduced. The temporal firing patterns of all neurons were unchanged by cooling. Changes in first spike latency (FSL) were negatively correlated with changes in firing rate. These data indicate that each IC differentially modulates the frequency selectivity, sensitivity, firing rate and FSL, but not the temporal firing pattern or CF of neurons in the contralateral IC. These findings demonstrate that the analysis of auditory stimuli in each IC is dependent on intercollicular processing. The ICs should therefore be viewed as working cooperatively rather than independently.