The principal aim of the research described in this thesis was to develop a technique of non-invasively measuring the oxygen saturation within the retinal vasculature of human subjects (retinal oximetry). The evaluation of a hyperspectral fundus camera used to acquire retinal images in different wavelengths of visible light, and the image analysis techniques used to perform retinal oximetry are described. Validation of the oximetry techniques was performed using an artificial eye containing human blood of known oxygen saturation: the calculated oxygen saturation was compared to the gold standard measurement. The mean differences between the calculated and measured oxygen saturations were small. Hyperspectral imaging/oximetry of normal subjects was performed to characterize the oximetric features of the retinal vasculature. The mean oxygen saturation (± SD) of the temporal retinal arterioles and venules were 110.8% (± 11.8%) and 27.7% (± 3.2%) respectively. The application of the retinal oximetry technique was explored in patients with retinal arterial and venous occlusion to determine whether oximetric changes in the retinal vasculature could be detected. Variation in measured oxygen saturation of the retinal arterioles and venules respectively were apparent, and corresponded with angiographic features of retinal capillary loss. The techniques were applied to patients with asymmetrical primary open angle glaucoma to determine whether oximetric changes could be detected. The mean oxygen saturation of the temporal retinal venules were significantly higher [44.8% (± 24.2%)] in the more advanced glaucomatous eyes compared to normal subjects. Hyperoxia of the retinal venules suggests reduced oxygen consumption as a consequence of inner retinal dysfunction in glaucoma. However, because of the small sample size, further research on a larger population of subjects is required to support this finding. Hyperspectral imaging could be used to detect oximetric abnormalities in the retinal vasculature in patients with retinovascular occlusion and glaucoma.

The key to understanding how the brain works is to understand the computations it performs. The structure of anatomical and functional connectivity determines what the brain can compute and how it does so. Correlations have served as a tool for analysing connectivity for over five decades. The mammalian visual cortex has become the most intensively researched cortical area and is unmatched for our knowledge of its anatomical layout and, most importantly, stimulus selectivity. Furthermore, recent perspectives on correlations have arisen from information theory and network models of the brain. The purpose of this dissertation is to determine the precise structure of functional connectivity in cat primary visual cortex. We aim to contribute to and extend previous work by analysing the structure of neural responses and correlations during spontaneous activity, the presentation of artificial stimuli and the presentation of natural stimuli. We report on a comprehensive set of twenty functional and neurophysiological factors, and reveal how previously unexplored factors govern correlations in visual cortex in vivo. Furthermore we find novel functional relationships between factors governing the responses of neurons, and report on a set of properties which allow to distinguish narrow from broad spiking cells. Much attention is devoted to the precise functional dependency of correlations upon firing rate, with the development of methods to remove the firing rate modulation. We show that timescale is an important determinant of correlations, and that natural stimuli generate different correlations than artificial stimuli. We also show that during spontaneous activity, neurons are more likely to fire together if they are tuned to a similar orientation. These results emphasize that both spontaneous and stimulus driven cortical activity contain rich structure that is far from a decorrelated state.

No description available.

Molecular genetics has transformed our understanding of disease and is gradually changing the way medicine is practiced. Genetic mapping provides a powerful approach to discover genes and biological processes underlying human disorders. Recent advances in DNA microarray and sequencing technology have significantly increased the power of genetic mapping studies and have ushered in a new era for biomedicine. In this thesis, linkage analysis (including homozygosity mapping), exome sequencing and candidate gene sequencing have been utilised to genetically dissect autosomal recessive retinal disease. Subsequently, clinical findings from patients found to be similar in terms of molecular pathology have been pooled. DNA and basic phenotypic data from over 500 unrelated individuals were available for the project. Disease-causing variants in three genes that have not been previously associated with human recessive disorders are reported: (a) biallelic mutations in TRPM1 abrogate ON bipolar cell function and cause complete congenital stationary night blindness; (b) biallelic mutations in KCNJ13, a gene encoding an inwardly rectifying potassium channel subunit cause Leber congenital amaurosis; (c) biallelic mutations in PLA2G5, a gene encoding group V phospholipase A2, cause benign fleck retina. The consequences of mutations in these and other disease-related genes (RDH5, GRM6, KCNV2, OAT and SAG) on retinal structure (spectral domain optical coherence tomography, fundus autofluorescence imaging) and visual function (electrophysiology, perimetry testing) have been studied; features that may have mechanistic relevance have been identified. Additionally, DNA sequence variation of a highly polymorphic gene (C2ORF71), recently associated with photoreceptor degeneration, has been studied and quantified in patient and control samples. Basic bioinformatics tools to analyse genomic data have been developed (bash, perl, python and R programming languages). Overall, results presented in this thesis contribute to an understanding of Mendelian retinal disease that is not only observational but also mechanistic.

Oxygen sensing is a fundamental biological process which is critical for appropriate development of the eye and implicated in neovascular eye disease including age-related macular degeneration, diabetic retinopathy and retinopathy of prematurity. This thesis describes a programme of work designed to investigate the role of hypoxia-inducible transcription factors (Hif’s), its downstream effector proteins, and its upstream regulator, the von Hippel Lindau protein (Vhl), in the development of the eye and neovascular eye disease. The first part of this work investigates the consequences of Hif activation in the developing retinal pigment epithelium (RPE) using a tissue specific knockout technology in mice. It demonstrates that appropriate regulation of Hif’s by Vhl is essential for normal RPE and iris development, ocular growth and vascular development and indicates that ocular hypoxia may be a previously unrecognised mechanism in the development of microphthalmia. The second part of this work studies the role of Hif1a in myeloid cells in the development of pathological neovascularisation using tissue-specific knockout technology and murine models for retinal and choroidal neovascularisation. It demonstrates that Hif1a signalling in myeloid cells contributes substantially to the development of retinal and choroidal neovascularisation and provides a rationale for developing antiangiogenic treatments that target Hif1a signalling in myeloid cells in neovascular eye disease. The third part of this work investigates the oxygen distribution in the vitreous and its relation to HIF1a and its downstream molecules in proliferative diabetic retinopathy (PDR) in man. It identifies significant intraocular oxygen gradients in PDR with areas of hyperoxia and hypoxia and demonstrates increased levels of HIF1a in the vitreous in PDR which correlate with increased levels of inflammatory and angiogenic cytokines in PDR. These findings suggest that HIF1a activation by inflammation and/or hypoxia is a central feature in the progression of PDR and that its inhibition may potentially serve as a target for therapeutic intervention.

Fenestrae are transcellular membrane pores that mediate blood-tissue exchange in highly specialized vascular endothelia such as in choroidal capillaries. Substances that traverse the pore never encounter the contents of the cytoplasm and are transported in a rapid and presumably energy-efficient manner. Fenestrae arise in attenuated regions of the endothelial cell periphery and are highly organized in clusters termed sieve plates. My PhD project was based on the identification of novel components of fenestrae and how these components contribute to mechanisms regulating fenestra formation. Using an in vitro biogenesis model coupled with proteomic analysis, we identified several proteins enriched in fenestrated plasma membranes. Localisation of candidate proteins was accomplished by immunolabelling, confocal microscopy, and transmission electron microscopy. Functional roles for the candidate proteins in fenestra biogenesis were probed through gain- and loss of function techniques. Coimmunoprecipitation was used to uncover protein-protein interactions, and biochemical reagents were applied to probe the signalling pathways involved in fenestra formation. Through extensive investigation, we identified the ERM (ezrin/radixin/moesin) protein moesin as a component of fenestral sieve plates. Inhibition of moesin function by expression of a dominant negative mutant or siRNA resulted in inhibition of fenestra formation, whereas knockdown of another regulator of the actin cytoskeleton, annexin II, led to a robust increase in fenestra formation. Biochemical and structural analyses showed that these modulators control the formation of an actin-fodrin submembrane cytoskeleton that is essential for sieve plate and fenestra formation, and that this cytoskeleton is directly linked to the fenestra pore protein PV-1. The transmembrane protein Na,K-ATPase is also a structural component of the submembrane complex, and functions as a regulator of fenestra formation in vitro and in vivo. These findings provide a conceptual framework linking the actin cytoskeleton to membrane remodeling during fenestra biogenesis and new molecular tools for probing fenestra structure and function.

The highest concentration of zinc in human tissues is found in the retinal pigment epithelium (RPE)-choroid complex. Despite the association of zinc deficiency with age-related macular degeneration (AMD) and the widespread use of zinc supplementation to slow the progression of AMD, very little is known about how zinc affects the RPE and the choroid. Molecular and cell biology techniques were used to uncover how changes in zinc levels could play a role in regulating the RPE-choroid complex. First, QRT-PCR was used to assess the expressions of all 24 known zinc transporters in cadaveric human RPE, cultured RPE cells and cells isolated from other parts of the retina, ZIP12 was identified as a potentially important transporter to regulate zinc levels at the RPE-choroid interface. As there is very little published about ZIP12, bioinformatics and data mining were used to understand how this protein might function. Confirmation of these predictions was achieved through the cloning and expression of V5-tagged ZIP12 protein in different cell lines. Based on these experiments, we concluded that ZIP12 is a plasma membrane transporter that mediates zinc influx. In parallel, we tested the hypothesis that extracellular zinc levels in Bruch’s membrane might be involved in regulating both the RPE as well as the fenestrated choroidal capillaries using cultures of ARPE19 and bEND5 cells respectively. The presence of extracellular zinc in the growth media affected the characteristics of ARPE19 cells as well as fenestrae formation in bEND5 cells. In summary, the range of zinc transporter at the RPE-choroid interface was defined and properties of one particular transporter, ZIP12 which may have a specific role at this site, were elucidated. Using cellular systems some of the effects of zinc on the RPE-choroid complex were investigated. Future studies are required to elucidate the role of zinc in the AMD pathogenesis.

The vascular system delivers oxygen and nutrients to tissues throughout the developing organism, including the nervous system. Vice versa, the nervous system innervates resistance arteries to modulate vascular function. The two systems share several guidance cues and cell-­‐surface receptors. One receptor is neuropilin 1 (NRP1), which is present on blood vessels and neurons. The sympathetic nervous system is a neural crest cell (NCC)-­‐derived structure that innervates the heart and blood vessels to modulate heart rate and vasoconstriction. Semaphorin3A (SEMA3A) signals through NRP1 to pattern the axonal projections of sympathetic nerves. I show here that this signalling pathway also controls the earliest stage of sympathetic nervous system development -­‐ sympathetic NCC migration through the somites. Accordingly, sympathetic NCCs stray into ectopic territories and differentiate into sympathetic neurons in mice with disrupted NRP1/SEMA3A signalling. I also show that NRP2/SEMA3F signalling provides a backup pathway for NRP1/SEMA3A signalling in sympathetic NCC guidance. I further show defective sympathetic innervation of the heart and dorsal aorta in postnatal mice with disrupted NRP1/SEMA3A signalling and describe a previously unidentified role for NRP2 in sympathetic axon guidance. I found that the recently discovered SEMA3G does not play a part in sympathetic axon guidance to target arteries, despite its unique arterial expression. The alternative NRP1 ligand, a vascular endothelial growth factor isoform termed VEGF164, is essential for the sprouting of new blood vessels from existing ones in a process called angiogenesis. A large body of in vitro evidence suggests that heparan sulphate proteoglycans (HSPGs) are required for VEGF164 -­‐driven angiogenesis by promoting its interaction with its receptors VEGFR1, VEGFR2 and NRP1. In vivo data supporting the idea that HSPGs are essential for angiogenesis, however, are sparse. I here found that mouse embryos lacking enzymes required for the sulphation of HSPGs, or lacking enzymes essential for HSPG production in specific cells, had no obvious vascular branching defects in the hindbrain and do not phenocopy mutants lacking VEGF164. These observations suggest that HSPGs are not essential for VEGF164-­‐driven angiogenesis. In contrast, I found that the VEGF164/NRP1 guided migration of facial branchiomotor neurons was dependent on the presence of HSPGs. Taken together, these results provide evidence for the differential requirement of HSPGs in VEGF164-­‐driven neural, but not endothelial cell patterning in the hindbrain.

Mutations in the phosphoinositide 5-phosphatase OCRL1 cause Lowe Syndrome and Dents-2 disease, both involving selective renal proximal tubulopathy. Epithelial cells lining renal proximal tubules are highly polarised with distinct apical and basolateral membranes separated by intercellular junctions. Using Madin Darby canine kidney (MDCK) cells as a model of the renal tubular epithelium, we found a pool of OCRL1 targeted intercellular junctions and was required for the correct organisation of apical and basolateral membranes. One of the first events when cells were depleted of OCRL1 was a block in apical recycling. Apical cargo accumulated in exaggerated Rab11 positive recycling compartments with ectopic accumulation of the OCRL1 substrate PI(4,5)P2. Among the apical cargo were Gp135 and key regulators of apical membrane formation, including Cdc42 and the Par6-aPKC polarity complex. Rescue of recycling required the 5-phosphatase domain of OCRL1, suggesting down-regulation of 5-phosphoinositides is necessary for cargo exit from apical recycling endosomes. Eventually, the Rab11 positive vesicles containing apical cargo were targeted to the plasma membrane to re-form the apical domain. However, instead of being targeted to the cell apex, the apical vesicles were trafficked towards the lateral membrane where lumens formed, resembling the bile canaliculi in hepatocytes. This process was co-ordinated with cell division. Lateral lumens formed close to the midbody formed during cytokinesis. During cytokinesis Rab11 normally regulates a trafficking pathway to the midbody. Therefore, in cells lacking OCRL1 apical cargo held up in Rab11 positive compartments may be aberrantly trafficked along this pathway when cells divide. We also found that the orientation and subsequent resolution of mitoses was altered in cells lacking OCRL1, which may contribute to improper positioning of the midbody and subsequent failure to polarise correctly. In summary, these results implicate OCRL1 in multiple steps of the process that co-ordinates apical recycling and cell division in polarised cells.

The inherited corneal diseases form a clinically and genetically heterogeneous group of disorders. They include the various types of progressive corneal dystrophies as well as some corneal structural abnormalities for which there is thought to be a genetic basis. These conditions are distinct from the corneal degenerations that result solely from aging or environmental effects. In this thesis I have concentrated on some selected inherited disorders. To try to improve our understanding of the disease mechanisms I have phenotyped affected families, performed candidate gene screening, and made genotype-phenotype correlations. I have collected the largest cohort of families with keratoconus reported to date and probands were screened for mutations in the VSX1 gene previously reported to be associated with this disorder. No disease-causing mutations were identified confirming that this gene only plays a very minor role in the pathogenesis of keratoconus. In a white British family with cornea plana the c.740A>G mutation within the KERA gene was identified and evidence was sought for a common founder with previously reported Finnish patients with cornea plana. One novel mutation was found and common founder in some of the cases was suggested. Disease-causing changes were found in seven Czech families with anterior and stromal corneal dystrophies known to be associated with the TGFBI gene and, of great interest, was a novel phenotype in a family with a p.H626P change. A set of Czech families with macular corneal dystrophy was screened for mutations in the CHST6 gene. In one British family with early-onset Fuchs endothelial corneal dystrophy we demonstrated a previously reported p.L450W mutation in the COL8A2 gene. Finally, by screening all three known genes implicated in posterior polymorphous corneal dystrophy four novel mutations were identified in the ZEB1 gene which provides additional evidence for the genetic heterogeneity of this disorder.