The ‘endless forms most beautiful’ that populate our planet rely on the process of cell division to ensure equal segregation of the cellular content including the DNA to the two daughter cells. The accurate segregation of chromosomes in eukaryotes relies on connection between replicated sister chromatids, a phenomenon known as sister chromatid cohesion. Sister chromatid cohesion is mediated by a conserved ring-like protein complex known as cohesin. Defects in this process can promote aneuploidy and contribute to meiotic segregation errors with adverse consequences for developing embryos. Despite numerous advances into understanding cell division at the molecular level, we still lack a comprehensive list of the participating proteins and complexes. The aim of this thesis was to use available functional genomic and proteomic data to identify novel regulators of mitosis in human cells. Using an RNAi approach, we identified a set of factors involved in pre-mRNA splicing whose depletion prevents successful cell division. Loss of these splicing factors leads to a failure in chromosome alignment and to a protracted mitotic arrest that is dependent on the spindle assembly checkpoint. This mitotic phenotype was accompanied by a dramatic loss of sister chromatid cohesion that we could show happens as soon as DNA replication. While depletion of pre-mRNA splicing mediators had no effect on cohesin loading onto chromatin, it prevented the stable association of cohesin with chromatin. Immunoblotting revealed that the depletion of splicing factors caused a 5-fold reduction in the protein levels of Sororin, a protein required for stable association of cohesin with chromatin in post-replicative cells. Further analysis suggests erroneous splicing of Sororin pre-mRNA upon depletion of splicing factors. Importantly, the sister chromatid cohesion loss caused by depletion of splicing factors could be suppressed by a Sororin transgene that lacks introns. Our results suggest that that pre-mRNA splicing of Sororin is a rate-limiting step in the maintenance of sister chromatid cohesion in human cells. Our work reveals that a primary cellular pathology of compromised pre-mRNA splicing is a mitotic arrest accompanied by split sister chromatids. Our work linking splicing and sister chromatid cohesion has implications for the pathology of Chronic Lymphocytic Leukemia (CLL). One of the splicing factors that we implicate in sister chromatid cohesion is SF3B1, whose gene is one of the most frequently mutated genetic drivers found in CLL patients.

Abasic sites are thought to be one of the most frequently formed lesions within cells. They are particularly dangerous during DNA replication as they can block the progression of replication forks. Such stalled forks have the potential to collapse, which can impact genome stability and therefore cell survival. To complete replication in the presence of abasic sites, cells use DNA damage tolerance pathways that can bypass abasic sites without their repair. The in vivo study of DNA damage tolerance is complicated by the temporal and spatial nature of naturally occurring damage. To understand the molecular details of these pathways, it is useful to have systems that can recapitulate events ideally at single replication fork resolution. Plasmids harbouring damage site-specifically have been introduced into cells, allowing the study of the genetic control and mutagenic bypass at these sites. However it is uncertain if events observed fully reflect those occurring in the context of chromatin. I have developed a system where abasic sites can be formed site-specifically at a known location in the Saccharomyces cerevisiae genome. By creating these lesions during the G1-phase of the cell cycle, upon release into S-phase, the response of the DNA replication fork can be analysed. Their formation has been characterised with respect to their site-specific targeting, distribution, effects on replication and activation of the DNA damage checkpoint. This system can now be used to analyse the damage bypass response by looking at the recruitment of proteins to the fork by chromatin immuno-precipitation and live-cell fluorescence microscopy. Additionally, replication forks undergoing abasic site bypass can sbe cross-linked to identify novel factors involved. These techniques will provide further insight into DNA damage tolerance events occurring at abasic sites.

Although the complexity of animal body plans increased greatly following the evolution of the mesoderm, interspecies genome comparisons show that the Cnidaria, which lack mesoderm, share the majority of their genes with the triploblastic Bilateria. Indeed, the genome of the diploblastic sea anemone Nematostella vectensis includes many genes implicated in mesoderm formation in the triploblasts. The aim of my thesis was to investigate the role of one such gene, the Nkx homeobox transcription factor ladybird (lbx1), in Xenopus and Nematostella using high throughput methods such as ChIP-Seq and RNA-Seq. Gain- and loss-of-function experiments in Xenopus revealed that lbx1 has a role in the development of primary neurons as well as in the formation of body wall and head muscles. During neurulation, lbx1 is expressed in the presumptive hindbrain, where it appears to regulate the size of the neural progenitor pool by activating sox2. Lbx1 is also capable of activating FGF signalling, as revealed by in vitro studies using mesoderm induction assays in animal caps. Downregulation of Nvlbx in Nematostella causes no effect on the ectoderm of Nvlbx morphants, but there are severe changes in the endoderm, including a loss of mesenteric muscle cells. RNA-Seq analysis of such embryos suggests that Nvlbx interferes with Notch or FGF signalling pathways, both of which have been implicated in Nematostella mesentery formation. Together, my functional studies and high throughput data suggest that lbx1 regulates FGF signalling in both Nematostella and Xenopus, and that this mechanism is likely to be conserved throughout the metazoan lineage.

Cockayne’s Syndrome is a rare hereditary disorder that is characterised by severe neurological abnormalities, cutaneous photosensitivity, severe growth failure and premature aging. The aetiology of this disorder is attributed to mutations and deletions in two proteins, Cockayne’s Syndrome A (CSA) and Cockayne’s Syndrome B (CSB). Both proteins are essential in Transcription-Coupled Nucleotide Excision Repair (TC-NER), a highly specialised pathway that rapidly removes transcription-blocking DNA lesions from the transcribed strand of active genes. Besides a role in TC-NER, CSB has an additional function in general transcription, and both CSA and CSB have been implicated in the repair of oxidative DNA lesions. CSB is known to be ubiquitylated by the CSA ubiquitin ligase complex and at least during TC-NER this process may be coupled to CSB proteosomal degradation. In addition, a recent study from our laboratory identified an ubiquitin-binding domain in CSB, which is essential for its function in TC-NER. The initial aim of the study described in this thesis was to characterise the functional role of protein ubiquitylation in TC-NER through the identification and characterisation of the protein that interacts with CSB’s ubiquitin-binding domain. Realising that this protein might be ubiquitylated CSB itself, an alternate approach was the identification and mutational analysis of ubiquitylation sites in CSB. The latter approach was more successful. The expression and subsequent purification of CSB from human cells facilitated the identification of six ubiquitylation sites. Unexpectedly, mutation of the ubiquitylation site at CSB’s lysine 991 gives rise to genome instability even in the absence of DNA damage. Surprisingly, the CSB K991R cells are proficient in TC-NER, but defective in transcription, and hypersensitive to oxidative damage. To the best of my knowledge, this is the first separation of function mutation described for CSB, and provides new evidence for the possible involvement of de-regulated transcription and persistent oxidative DNA damage in the aetiology of Cockayne’s Syndrome.

Immunoglobulin G (IgG) is the most abundant antibody class in serum, and in humans is present as four subclasses, IgG1- 4, which have distinct properties. IgG binds to antigens via its two Fab regions, and to its effector ligands via its Fc region. IgG is also widely used as a pharmaceutical due to its specific binding to targets. Solution scattering and constrained modelling, along with analytical ultracentrifugation, were used to determine IgG structures in solution. Rabbit IgG was shown to have an asymmetric solution structure, and these structures explained the ability of its two ligands, the Fc receptor and complement C1q, to bind to the top of its Fc region as it is fully accessible and unhindered by the Fab regions. Rabbit IgG also displayed buffer-dependent dimerization. Human IgG4 is distinct from other subclasses as it can undergo Fab-arm exchange and is unreactive compared to the other subclasses. The IgG4 structures were also asymmetric, and blocking of effector binding sites by the Fab regions offered an explanation of the unreactivity of IgG4. Self-association of IgG4 was also observed under certain buffer conditions. In the immune system, human IgG1 is able to bind to all its ligands, and was shown to have an asymmetric solution structure. However, docking studies of the C1q ligand demonstrated that the Fab regions are sufficiently far apart to allow access to effector binding sites. In distinction to rabbit IgG and IgG4, IgG1 did not display significant dimerization under any buffer condition tested. Study of IgG4 under low pH conditions, to mimic chromatography purification steps, revealed that the IgG4 monomer has a more compact structure at pH 3, and gradually forms dimers and higher oligomers. Neutralization of IgG4 to pH 7.4 leads to large amounts of aggregation, with time held at pH 3 and protein concentration important factors in the amount of aggregation observed.

Wnt signalling is required for neural crest induction; however the direct targets of the Wnt pathway during neural crest induction remain unknown. I show here that the homeobox gene Gbx2 is essential in this process and is directly activated by Wnt/β-catenin signalling. Gbx2 has previously been implicated in posteriorization of the neural plate. Here I unveil a new role for this gene in neural fold patterning. Loss of function experiments using antisense morpholinos against Gbx2 inhibit neural crest and expand the preplacodal domain, while Gbx2 over expression leads to transformation of the preplacodal domain into neural crest cells. I show that the neural crest specifier activity of Gbx2 is dependent on the interaction with Zic1 and the inhibition of preplacodal genes such as Six1. In addition, I demonstrate that Gbx2 is upstream of the neural fold specifiers Pax3 and Msx1. My results place Gbx2 upstream of the neural crest genetic cascade being directly regulated by the inductive molecules, and support the notion that posteriorization of the neural folds is an essential step in neural crest specification. I propose a new genetic cascade that operates in the distinction between preplacodal and neural crest territories.

Activation of the transmembrane death receptor Fas (CD95/APO-1) by a membrane bound ligand (FasL/CD95L) activates the extrinsic pathway of apoptosis. Intracellular Fas death domains (DDs) are induced to oligomerise enabling binding to the adaptor protein FADD, thereby leading to the recruitment of procaspase 8 and other proteins to form the death inducing signalling complex (DISC).This thesis describes an investigation of the structure and function of the cytoplasmic Fas-DD. A model for the solution structure of the Fas-DD was published in 1996, it has since been reported that the death domain can form at least one other conformation when in complex with FADD. As a foundation to the work in this thesis, modern multidimensional NMR techniques have been used to solve the structure of the FasDD, to further probe the potential for alternative conformations. It has previously been reported that Fas can be phosphorylated at Tyr291, providing a platform for the recruitment of binding partners that can affect non-apoptotic signalling. The second part of this thesis details the development of an expressed protein ligation methodology to prepare a Tyr291 phosphorylated Fas DD to provide a basis for in vitro studies of the structural, dynamic and functional effects of phosphorylation. It is widely accepted that Fas is palmitoylated at Cys199 and recognised by the membrane cytoskeletal protein, ezrin. Fas palmitoylation is important for clathrinmediated internalisation of the DISC, and amplification of the caspase cascade. There are multiple reports detailing the binding of ezrin to Fas, but it is not clear whether this interaction occurs in a palmitoylation-dependent manner. Efforts to characterise an interaction between bacterially expressed intracellular Fas and ezrin proteins were carried out using a number of biophysical assays, described in the third part of this thesis. Building upon this, the fourth section explores the preparation of a palmitoylated Fas construct suitable for biophysical analysis by incubating recombinant Fas with palmitoyl-CoA.

ATP and ADP mediate gut smooth muscle relaxation through two receptors, P2Y1 and P2Y11. This project aims to investigate the interaction between these two receptors by developing a recombinant model of the P2Y receptors expressed in gut smooth muscle cells (SMCs) by transfecting the human P2Y11 receptor cDNA into CHO-K1 cells, which express an endogenous P2Y1 receptor. Individual clonal cell lines expressing different densities of hP2Y11 were isolated from this stably-transfected CHO-K1:P2Y11 pool and characterized. A clone expressing a “high” density of hP2Y11 (13) and a clone expressing a “low” density of hP2Y11 (6) were selected for further study. Control 1321N1 cell lines expressing each receptor in isolation (1321N1-hP2Y1 and 1321N1-hP2Y11) were used for comparison purposes. The potency (EC50) of eight different nucleotide agonists was determined in calcium assays in the co-expressing cell lines. ADP and 2meSATP responses were biphasic in clone 13 but monophasic in clone 6. To investigate the nature of the two sites of the biphasic curves in clone 13, the effect of MRS 2179, NF 340 and Reactive Red on agonist responses was determined. MRS 2179 antagonized the high affinity site of the biphasic ADP and 2meSATP responses in clone 13 without affecting the low affinity site. NF 340 had no effect on agonist responses in clone 13. Reactive Red antagonized both sites of the biphasic curves in clone 13. These data suggest that the high-affinity site of the biphasic ADP and 2meSATP responses in clone 13 corresponds to P2Y1. The low-affinity site of the 2meSATP curve is most likely P2Y11. The low-affinity site the ADP response displays both P2Y1 and P2Y11-like. The novel ADP site, therefore, is elicited by differences in the expression level of P2Y11 and may correspond to a P2Y1:hP2Y11 receptor heteromer or a macromolecular complex containing both P2Y1 and P2Y11.

No description available.

Echinococcosis is a serious and often fatal zoonotic disease caused by parasites in the genus Echinococcus. Echinococcus spp. cycle between intermediate and final hosts, and it is the accidental ingestion of eggs in faeces of final hosts (usually canids) that causes the disease in humans. There is evidence that echinococcosis is re-emerging in Kyrgyzstan, with increasing numbers of human cases reported from the south of the country. However, little is known about canine echinococcosis in the local domestic dog population, despite the fact that dogs are the main source of human infection. As such, this thesis focuses on canine echinococcosis in the Alay Valley, southern Kyrgyzstan. In order to study canine echinococcosis, reliable tools for diagnosing infection in dogs are needed. Previous studies have found that coproELISAs measuring Echinococcus spp. antigens in faecal samples can accurately detect canine echinococcosis. As part of this study, polyclonal antibodies were extracted from hyperimmune rabbit sera and optimized in a hybrid sandwich coproELISA for the detection of Echinococcus spp. in faecal samples with high diagnostic sensitivity and specificity. However, coproELISAs are genus specific, and identifying species/strains of Echinococcus spp. requires coproPCR. Although previously published coproPCR protocols were available for detection of E. granulosus and E. multilocularis, such a protocol was not available for E. canadensis, which was found to occur in the Alay Valley as part of this study. As such, a new analytically specific and sensitive coproPCR protocol for the detection of E. canadensis was developed. The prevalence of canine echinococcosis in four communities in the Alay Valley was estimated by sampling 333 dogs in May 2012. The coproELISA prevalence was found to be high, with an average of 26.4%. All faecal samples collected in May 2012 were DNA extracted and tested by coproPCR. CoproPCR testing of coproELISA positives found that 33.3% tested positive for E. canadensis, 8.2% tested positive for E. granulosus, and 11.0% tested positive for E. multilocularis. Establishing pre-intervention canine coproELISA prevalences is crucial for evaluating the impact of any future control programs. As the ecology of dogs is important when studying diseases spread by them, dog demography, dog roles, dog husbandry and dog roaming was studied in four communities in the Alay Valley, as well as environmental faecal contamination being assessed. The local dog population was large, with 1 dog/9.36 people. Most dogs were male and below five years of age. Dogs played various roles in the communities, including as sheep dogs, guard dogs, and pets. Most dogs were free-roaming and could move up to 2km away from their homes. The large population of free-roaming dogs was reflected in high levels of environmental contamination, with between 0.11 and 1.20 faecal samples/100m2 recorded. Following the implementation of a World Bank control scheme which aimed to dose all owned dogs with praziquantel four times a year, the effects of this programme on canine echinococcosis were evaluated. In order to do this, Lot Quality Assurance Sampling (LQAS) was applied to ten communities in the Alay Valley, with communities sampled 9 and 21 months after the start of dosing. Results suggested that after 21 months of dosing, at least 75% of dogs were being dosed in 8/10 communities, and coproELISA prevalences were reduced in 5/4 communities respectively after 9 and 21 months of dosing. As control programmes require large commitments of time and resources, it is important to be able to evaluate how well these are meeting their targets. Here, reliable tools were developed to study canine echinococcosis, the pre-intervention canine echinococcosis coproELISA prevalence was established, dog ecology and demographics were studied, and LQAS was used to assess the first two years of an echinococcosis control programme. It is hoped that these studies contribute to a better understanding of the re-emergence of echinococcosis in Kyrgyzstan and the impacts of control schemes on canine echinococcosis.