The aim of this investigation was to define the in vitro conditions necessary to support cytochrome P450-mediated metabolism in rat cultured hepatocytes, such that this system could then be used as an in vitro model in the study of cytochrome P450-mediated cytotoxicity. Maintenance of P450-dependent enzyme activities in culture was not affected by supplementation of culture medium with haem, but was markedly influenced by the age and sex of the hepatocyte donor animal. Induction in primary culture by phenobarbitone and beta-naphthoflavone was investigated, and found to be quantitatively and qualitatively different to the induction observed in vivo, hepatocytes in culture being particularly refractive to induction by phenobarbitone. The maintenance in primary culture of a range of enzyme activities was determined following treatment of rats in vivo with isoniazid and dexamethasone, in addition to phenobarbitone and beta-naphthoflavone, and in general, there was good maintenance of the induced activities. The activities were chosen as possible selective substrates for the different induced isozymes, with a view to using the activity profiles to characterise different classes of inducer; however, although selective induction was observed with isoniazid, beta-naphthoflavone and dexamethasone, all the chosen activities were induced by phenobarbitone. The final part of this work involved determining cytotoxicity in vitro, following induction of P450 in vivo with phenobarbitone and beta-naphthoflavone. Seven known hepatotoxins were investigated, and the results obtained agreed well with available in vitro and in vivo literature data. In summary, a range of constitutive and induced enzyme activities were maintained at high levels in hepatocytes cultured for twenty-four hours from adult male rats, and an induction in vivo/hepatocyte culture protocol shown to be a viable in vitro model for the study of metabolism-mediated toxicity, as an alternative to induction and detection of toxicity in vitro. NB. This ethesis has been created by scanning the typescript original and may contain inaccuracies. In case of difficulty, please refer to the original text.

The FRAME KB cytotoxicity assay is an ill vitro test for basal cytotoxicity which measures the sub-lethal inhibition of cell growth by toxic substances. Exponentially growing 3T3-Ll mouse fibroblasts are exposed to a range of concentrations of a test substance for 72 hours, then relative cell number is estimated by the protein/kenacid blue dye-binding method. The assay was evaluated for its ability to predict parameters of in vivo acute lethal potency. In vivo/in vitro comparisons were performed for a set of miscellaneous chemicals and for a set of metal compounds. The degree of correlation was closer for the metal compounds than for the unrelated set, in the in vitro/mouse i.p. LDso comparison. The cytotoxicity assay was more useful than metal "softness" (a physico-chemical parameter) for predicting metal compound toxicity ill vivo. An investigation of the ill vitro toxicities of a group of commercial chemicals and formulations revealed very poor ill vivo/in vitro correlations. Some were toxic to the 3T3-Ll cells, yet of very low toxicity to rats. This was partly due to the poor solubility of some of the substances, which probably caused their virtual non-toxicity to rats by oral dosage. Chemical volatility is another methodological problem for ill vitro assays. A simple modification of the FRAME KB cytotoxicity assay was successfully developed in order to prevent the underestimation of the cytotoxicities of volatile liquids. The assay also demonstrated potential use for providing data for the safety assessment of surfactants and toiletry formulations. It is emphasised that the FRAME KB cytotoxicity assay should never be used in isolation, but as part of a battery of tests chosen for a particular type of toxicity and/or type of chemical or formulation. The F9 embryonal carCInoma cell line was evaluated for its potential usefulness in in vitro toxicity testing. F9 cells were induced to differentiate morphologically and biochemically, and it was found that cells in different stages of differentiation did not respond in the same way to toxic chemicals.

Most of the biological effects of cannabis are due to the activation of specific cannabinoid receptors. To date, two such receptors have been discovered and are found predominantly in the central nervous system (the CB1 receptor) or the immune system (the CB2 receptor). Endogenous cannabinoid receptor ligands, the endocannabinoids, have also been isolated and the mechanisms of their synthesis and degradation postulated. By modulating the activation of cannabinoid receptors and endocannabinoid metabolism, synthetic cannabimimetic compounds have enormous therapeutic potential for the treatment of such diverse symptoms and diseases as pain, inflammation, cancer, hypertension, schizophrenia and multiple sclerosis. This thesis describes the design, synthesis and subsequent biological evaluation of three classes of novel, potentially cannabimimetic drugs, namely aryl ethanolamides, phenylphosphinic acids and alkylphosphinic acids. In order to assess cannabimimetic activity, the ability of these compounds to bind to the cannabinoid receptors and to inhibit endocannabinoid uptake and enzymatic hydrolysis was examined. Affinity for the CB1 receptor was assessed using radioligand binding assays in rat brain membranes. Although none of the compounds proved to be high-affinity CB1 receptor ligands, two aryl ethanolamide compounds exhibited some affinity for this receptor, suggesting that this general class of compound may have cannabimimetic potential. In order to ascertain whether the test compounds had affinity for the CB2 receptor, a radioligand binding assay was developed using porcine spleen membranes. To date, only the human, murine and rat CB2 receptors have been cloned and there has been no detailed examination of the cannabinoid binding profile of the porcine CB2 receptor. The Kd of the radiolabelled cannabinoid [3H]-CP-55,940 was determined in porcine spleen membranes and the Bmax subsequently calculated. The Ki values of a number of cannabinoid receptor ligands were then determined. These values were shown to be similar to the corresponding values obtained using cloned CB2 receptors. However, when the test compounds were assessed in this assay system, no affinity for the CB2 receptor was observed. To determine the effect, if any, of the test compounds on the endocannabinoid uptake system, accumulation of the radiolabelled endocannabinoid [3H]-anandamide into N18TG2 mouse neuroblastoma cells was examined. [3H]-Anandamide accumulation had previously been reported in this cell line but, until now, this mechanism had not been characterized. This accumulation was shown to be time-, temperature- and concentration-dependent and was inhibited by AM404 and bromocresol green, known inhibitors of the endocannabinoid carrier system. [3H]-Anandamide accumulation exhibited a Km value similar to those previously described for rat astrocytes and neurones and the time taken to achieve half maximal rate was shown to be considerably greater than in these rat cells. None of the test compounds significantly inhibited [3H]-anandamide uptake by N18TG2 cells although one phenylphosphinic acid compound, with structural similarities to AM404, appeared to be inhibitory at high concentrations. The final biological target examined was fatty acid amide hydrolase (FAAH), the enzyme that catalyses the hydrolysis of endocannabinoids. For FAAH studies, a novel, inexpensive and rapid spectrophotometric assay was developed as an alternative to the traditional radiochemical- and chromatography-based assays. Using this novel assay system, the Km and Vmax values of rat liver FAAH were determined and shown to be similar to those published in the literature. Known FAAH inhibitors were shown to inhibit FAAH in a concentration-dependent manner with IC50 values comparable to previously published data. In addition, this assay was used to demonstrate differences in FAAH activity between soluble and insoluble membrane preparations from rat liver and brain, possibly indicating the presence of, as yet, unknown FAAH enzymes. Attempts were also made to adapt this assay for use on a microtiter plate, where it was possible to detect FAAH inhibitors. Therefore, this spectrophotometric assay may prove to be of use in the high-throughput screening of chemical libraries for drugs that cause cannabimimetic effects via FAAH inhibition. None of the test compounds synthesized inhibited FAAH activity and this, combined with their lack of biological activity at the other targets tested, showed that they exerted no cannabimimetic effects.

In vertebrates, haematopoiesis occurs in two waves. The primitive wave gives rise to transient myeloid and erythroid cells whereas the definitive wave generates haematopoietic stem cells (HSCs), which maintain the blood system throughout life. These HSCs are able to self-renew and to give rise to progenitors that differentiate into mature cells of all blood lineages. Little is known about the cellular origin and molecular programming of HSCs. This knowledge is useful to generate HSCs in vitro from embryonic stem cells or induced pluripotent cells. In zebrafish, HSCs form in the intermediate cell mass (ICM), in the trunk of the embryo. Here, they develop dorsal to the primitive red blood cells and in close association with the ventral wall of the dorsal aorta (DA). Like their mammalian counterparts, they express the transcription factors runx-1 and c-myb. As in other vertebrates, zebrafish HSCs are thought to arise from the haemogenic endothelium in the ventral wall of the DA. A signalling cascade that involves the Hedgehog, Vascular endothelial growth factor (Vegf) and Notch signalling pathways is needed for arterial specification of the DA and for HSC formation. Short-term lineage tracing experiments showed that cells in the ventral wall of the DA first seed the tail mesenchyme (caudal haematopoietic tissue, CHT) through blood circulation before they seed the final sites of haematopoiesis, the thymus and kidney in the adult fish. Here, we conducted a tol2-transposon based gene trap vector screen with eGFP as the reporter gene with the aim to label nascent HSCs in vivo and to identify novel genes involved in haematopoiesis. We obtained 174 transgenic lines with tissue-specific eGFP expression in non-haematopoietic and haematopoietic tissues. We identified two lines with marker gene expression in haematopoietic cells. One of the transgenic lines, I-551:eGFP, showed reporter gene expression in primitive red blood cells and in endothelial cells in the ventral wall of the DA at 25 hours post fertilization (hpf). Using inverse PCR we identified the trapped gene in I-551:eGFP as gfi1.1, the homolog of the mouse Growth independence factor 1 (Gfi1), a transcriptional repressor expressed in HSCs. Here, we present results that the transgenic line Gfi1.1:eGFP enables us to follow emerging haematopoietic progenitors from the ventral wall of the dorsal aorta in their subsequent migration to the CHT, before they seed the final haematopoietic sites, the kidney and thymus. We show that Gfi1.1:eGFP expression is restricted to the ventral wall of the dorsal aorta by combining the transgenic line with endothelial and aorta-specific transgenic lines. We further demonstrate that the endothelial expression of the eGFP in the aorta is dependent on the vegf and notch signalling pathway and co-localizes to cells which also express the transcription factors runx1 and c-myb. When Gfi1.1:eGFP embryos are injected with the runx1 morpholino (MO), gfi1.1 expression in haemogenic endothelial cells initially occurs, which indicates that initial gfi1.1 expression is independent of runx1. But a reduction in the number of eGFP positive cells is observed at 50 hpf in the CHT. Using time-lapse imaging, we were able to visualize gfi1.1 positive cells detaching from the haemogenic endothelium. This observation indicates that gfi1.1 positive cells in the CHT are derived from the haemogenic endothelium and therefore nascent HSCs. We therefore strongly suggest that this transgenic line labels haemogenic endothelial cells at 26hpf. The transgenic line Gfi1.1:eGFP therefore provides a tool for studying HSC development since its expression labels the emergence of nascent HSCs from haemogenic endothelial cells and continues to be expressed in larval and adult haematopoietic sites.

Embryogenesis in vertebrates is regulated by highly complicated signaling networks, which involve various signaling pathways and factors. Sox3 is known to have critical roles during the whole of embryonic development in vertebrates. In zebrafish, it has been shown that Sox3 restricts organizer formation as well as inhibiting Fgf signaling, which is required for the expression of organizer genes. On the other hand, SUMOylation has been suggested to be a regulator of the transcriptional activity of Sox3. The SUMOylaton has previously been demonstrated on mouse Sox3 and chick Sox3. In this study, I examined the role and regulation of Sox3 in early embryogenesis in zebrafish. In the first part of the study, I inspected the detailed expression of gsc and chd, in comparison to foxd3 expression. It was demonstrated that Sox3 and Fgf signaling could repress both gsc and chd independently from each other. Sox3 was also found to be able to directly act on the promoter region of gsc. In the second part of the study, it was shown that the SUMOylation of Sox3 appeared to occur in zebrafish embryos. The SUMOylation of Sox3 was also shown to correlate with the chromatin fraction, although there was a very small fraction of Sox3 SUMOylated. The biological effect of SUMOylation of Sox3 has also been analysed. It was shown that SUMOylation of Sox3 could enhance the transcriptional repressor activity of Sox3. The result of luciferase reporter assay on the promoter region of boz also suggested that SUMOylation eliminated the transcriptional activator activity of Sox3. These data raise the possibility that SUMOylation of Sox3 might act as the switch from a transcriptional activator to a repressor.

Chapter 1 provides an introduction to metalloenzymes and discusses Ni-containing superoxide dismutase (NiSOD), which features Ni-thiolate ligation at the active site, and Cu containing galactose oxidase, which utilises phenoxyl radicals to perform its catalytic function. Studies concerning low molecular weight analogue complexes of each active site are reviewed and their relevance with respect to enzyme structure and function is discussed. Chapter 1 concludes with a description of the aims of this thesis project. Chapter 2 discusses of the preparation of a series of Zn(II), Ni(II) and Cu(II) Schiff-base diphenolate complexes incorporating the four novel pentadentate pro-ligands, [H2tBuLOH], [H2tBuLOCl], [H2tBuLOtBu] and [H2tBuLOOMe]. The different pKa values of the para-substituted arylamines (H, Cl, tBu and OMe) appear to modulate the electrochemical properties and redox behaviour of the Zn(II), Ni(II) and Cu(II) diphenolate complexes. Cyclic voltammetric, spectroelectrochemical and EPR spectroscopic studies reveal the first oxidation processes of [Zn(tBuLOOMe)]and[Cu(tBuLOOMe)]are associated with ligand-based oxidation processes, yielding kinetically inert species possessing phenoxyl radical character. Conversely, the first oxidation process of [Ni(tBuLOOMe)] appears to be metal-based. Chapter 3 discusses of the syntheses of Schiff-base diphenolate and dithiolate complexes with a diamine bridge that incorporate pendant heterocyclic groups, with the potential to coordinate in an axial position to a metal centre that possess equatorial N2O2 or N2S2 coordination environments. We targeted the syntheses and characterisations of these complexes as analogues of the active site of NiSOD.Initial attempts to synthesise a diamine incorporating a pendant imidazole donor followed previously established synthetic approaches. The use of protecting groups to protect the NH group of the imidazole ring was unsuccessful. The syntheses of Ni(II) Schiff-base diphenolate and dithiolate complexes was attempted by using the pro-ligands, [H2tBuLOOH]and [H2tBuLSOH]and a post complexation step that involved a reaction of the OHpendant of the Schiff-base ligand with an aromatic or heterocyclic functionalised acid chloride. Functionalisations of the [Ni(tBuLOOH)]and [Ni(tBuLSOH)]to incorporatependant imidazole (Imid), pyridyl (PyN), furan (Fu) and pyrrole (Pyr) yielded crude products and attempts to purify these products proved unsuccessful.[Ni(tBuLOPh)]could be isolated and X-ray crystallographic studies of [Ni(tBuLOR)] (R = OH and Ph) and [Ni(tBuLSOH)] demonstrate that each centre adopts an approximate square planar geometry. Electrochemical investigations of [Ni(tBuLOR)] (R = OH, Ph)and[Ni(tBuLSOH)] showed that each exhibit irreversible redox processes and demonstrate that the OH and phenyl ester functionalities pendant to the ligand backbone do not stabilize theproducts formed following the oxidation of [Ni(tBuLOR)] (R = OH, Ph)and[Ni(tBuLSOH)]. Chapter 4 extends the research described in Chapter 3 through the synthesis of [Zn(tBuLOOH)] and [Cu(tBuLOOH)] and the attempted functionalisation of the OH group in these complexes to incorporate additionalpendant heterocyclic groups including imidazole(Imid), pyridyl (PyN), furan (Fu) and pyrrole (Pyr) together with phenyl (Ph). X-ray crystallographic studies show that [Cu(tBuLOOH)] possesses an approximately square planar coordination geometry with the metal centre bound by diimine and diphenolate donors. The frozen solution X-band EPR spectra of [Cu(tBuLOR)] (R =Ph, PyN, Fu, Pyr and Imid) are similar to one another and confirm each complex possesses a paramagnetic Cu(II) S = ½ d9 metal centre. The similarity between each EPR spectrum as R is varied suggests that the functional group pendant to the ligand backbonedoes not interact greatly with Cu (II) centre in [Cu(tBuLOR)]. [Zn(tBuLOOH)] was isolated and mass spectrometric data suggested that [Zn(tBuLOR)] (R =Ph, PyN, Fu, Pyr and Imid) had formed, however pure products could not be isolated. Electrochemical studies of [Cu(tBuLOR)] (R = OH, Ph, PyN, Fu, Pyr and Imid) revealtwo irreversible redox processes. [Cu(tBuLOPyN)] possesses first and second redox processes that are ca.0.14 V and 0.23V more positive than their [Cu(tBuLOR)] (R = OH, Ph, Fu, Pyr and Imid) counterparts. This suggests that the pyridyl functionalization of the pendant arm in [Cu(tBuLOPyN)]may stabilize its oxidation product. Chapter 5 provides a conclusion to the thesis and draws together the principal themes of each chapter.

There are more than 60 Rab proteins in mammals constituting the largest family among the Ras superfamily of GTPases. Rab proteins are essential components of the intracellular secretory machinery, regulating transport and exocytosis of various intracellular organelles including secretory granules. Rab-regulated secretory functions are exerted by interaction with their downstream effector proteins, preferentially when they are activated into their GTP-bound state. Accordingly, Rab proteins can switch from active to inactive states through cycling from GTP- and GDP- bound forms, respectively. The Rab27 subfamily consists of Rab27a/b isoforms that have similar but not identical functions. Those functions include the regulation of trafficking, docking and fusion of various lysosome-related organelles and secretory granules; such as melanosomes in melanocytes and lytic granules in cytotoxic T lymphocytes. Rab27a/b exert their specific and versatile functions by interacting with eleven effector proteins, preferentially in their GTP-bound state. In recent years, a number of studies have identified roles for Rab27 proteins and their effectors in cancer cell invasion and metastasis, immune response, inflammation and allergic responses. These findings suggest that Rab27-effector protein interaction inhibitors could contribute to the development of effective strategies to treat these diseases. However, there has been very little research reported on the development and identification of such inhibitors. The foremost aim of the present study was to identify and develop candidate inhibitors of Rab27-effector protein interactions. To this end, fluorescent GST-pull down and FRET-based protein-protein interaction assays were developed to be used as in vitro read-outs for Rab27-effector interactions. In addition, a melanosome clustering assay was used as an indirect read-out for Rab27-effector protein interactions in living cells. The binding affinity of Slp1 and Slp2 effectors was determined using those in vitro assays. Furthermore, essential determinants of Rab27-Slp2 effector interaction were characterised and used to develop Slp2 super-effectors. The findings of this study suggest that the development of Rab27 super-effectors is an effective strategy to inhibit Rab27-effector interactions.

In mammalian cells vesicle transport is an important process, many diseases are caused by defects of trafficking. Key elements for organelle trafficking are Rab proteins which regulate this system. Rabs are GTPases and belong to the Ras superfamily proteins. Rab27a is one of the main actors of this process, it promotes the recruitment of secretory vesicles at the release site, the plasma membrane, by its role of interconnection between the vesicle and the motor protein myosin V. Dysfunction leads to disease such as Griscelli syndrome. Rab27a is a molecular switch, in the active state is able to bind effectors. The switching from the GDP-bound to the GTP-bound is catalysed by Rab3GEP. Furthermore, to function properly Rab27a has to be targeted to the organelle membrane, this function has been attributed to the guanine exchange factor as well. Rab3GEP contains a DENN domain (differentially expressed in normal versus neoplastic) at the N-terminus and a death domain at the C-terminus. Rab3GEP belongs to the family of DENN domain proteins, in general DENN domains are considered as domains with a guanine exchange factor activity. Down-regulation of this Rab3GEP in melan-a melanocytes induces to melanosome clustering at the perinuclear area. There is no knowledge about the mechanism of Rab3GEP in the activation and targeting of Rab27a to the vesicle membranes. To better characterise Rab3GEP protein, truncations and point mutations of this GEF were performed and they were tested in cell based assays (using melanocytes with Rab3GEP knock out) and in pull down assays (using the capacity to activate Rab27a as a read out of the GEF activity). Moreover, Rab3GEP wild type, and some mutants were mis-targeted from the cytosol (normal localisation) to mitochondria, in order to investigate their capacity to rescue melanosome distribution in melan-Rab3GEP KO cells and to examine their ability to re-target Rab27a to a different membrane organelle. Results reported in this thesis indicate that point mutations within the DENN domain impair the ability of Rab3GEP to rescue the melanosome disperse distribution and to activate Rab27a in vitro assay, suggesting the crucial role of this domain for the Rab3GEP activity. However, the DENN domain alone is not sufficient to rescue melanosome distribution and to activate Rab27a. This suggests an essential role of the other parts of the protein. Moreover, evidence in this thesis indicate that the cytosolic localisation of Rab3GEP is essential for the ability of Rab3GEP to rescue melanosome distribution, and that the targeting activity of Rab3GEP towards Rab27a is dependent on its GEF activity. However, evidence obtained by studying melan-Rab3GEP KO cells indicate that despite the important role of Rab3GEP in Rab27a activation/targeting, it is not essential.

The adult mammalian kidney forms from the reciprocal interaction between two tissues; the ureteric bud which will form the collecting duct system and the metanephric mesenchyme which will form all the cells in the nephrons. This thesis used ex-vivo embryonic kidney culture to show that during mouse kidney development metanephric mesenchymal cells migrated towards the periphery of the kidney. When this migration is pharmacologically inhibited the Six2 expressing metanephric mesenchyme cells no longer expanded their population and kidney development ceases. LIM kinase, which had been shown to regulate cell migration, when inhibited not only prevented cell migration in both embryonic mouse kidneys and in HK2 cells but also prevented embryonic kidney cells from completing mitosis and caused them to undergo apoptosis. This thesis showed that inhibition of LIM kinase in HK2 cells resulted in the formation of multiple alpha-tubulin foci, multiple centrosomes, the premature dispersal of the cohesin complex protein SMC3 in the absence of a fully formed spindle, and cell death. Thus concluding that active LIM kinase is required for the generation of the mitotic spindle and the appropriate dispersal of SMC3. This thesis furthered the understanding of how the metanephric mesenchyme develops and showed that these cells migrated away from the ureteric bud and this migration is required for further growth of the kidney. This thesis also showed that active LIM kinase is required for the completion of mitosis in both embryonic kidneys and HK2 cells.

The root system of a plant serves to anchor a plant to its terrain, to take up water and nutrients and to provide resistance to various biotic and abiotic stresses. Understanding the mechanisms dictating root, and lateral root (LR) development is crucial for work towards global food security. In the last two decades, microRNAs have been found to play crucial roles in many plant developmental processes. Up to now, no global expression profiling has focused specifically on Arabidopsis thaliana LR development. This study utilised mRNA and small RNA sequencing in order to build two large datasets. RNA was taken from developing LR primordia at selected stages and sequenced. The two datasets were validated and found to be of great depth and quality. They were complimented by an Affymetrix dataset (Voβ et al., 2015). A large number of differentially expressed, known microRNAs were identified as were a large number of targets. In many cases, expression of miRNAs and their targets during LR development revealed seemingly reciprocal expression patterns including a number which have previously been linked to LR development. In some cases, miRNA/target couples were highlighted as having a potential, unreported, role in LR development. This study focused particularly on the potential role of the miR159 family and one if its targets, the transcription factor MYB33. This study reports a dramatic LR defect in a miR159ab double mutant and a miR159 target mimic line and discusses a potential regulatory role of this miRNA and this target during LR initiation. As well as this, ten novel microRNAs were identified and validated. Putative targets were identified for a number of these microRNAs. This study produced overexpression lines for two of these microRNAs though no phenotype was identified. No significant developmental defects were identified in the roots of mutants for a number of the putative targets. This study presents the first comprehensive analysis of global expression profiling during LR development and has resulted in the production of two high quality datasets.