Small intestinal (SI) stem cells differentiate into short-lived progeny, except lysozyme-expressing Paneth cells. In vivo, the Tet-Op histone 2 B (H2B) - green fluorescent protein (GFP) fusion protein transgenic mouse has been employed to analyse the slow-cycling putative small intestinal epithelial stem cells, at cell position +4 (cp4 cells), through doxycycline-inducible transient expression of H2B-GFP. The aim of the study was to employ the same genetic mouse model in order to develop a culture system in which was possible to detect and investigate H2B-GFP-retaining putative SI stem cells. SI crypts isolated from 6-12 weeks old Tet-Op-H2B-GFP transgenic mice were established in culture (designated organoids) using growth factors and Matrigel. For in vitro transgene expression, doxycycline was added to the complete culture medium for 24/48 hours (pulse). H2B-GFP and lysozyme expression was studied by confocal and fluorescence microscopy. Percentages of H2B-GFP-retaining putative SI stem cells and of H2B-GFP-retaining Paneth cells persisting in organoids were determined by scoring GFP-immunoreactive cells. Ulex europaeus-I lectin (UEA-I) cell labelling, combined with flow cytometry, was employed in a pilot study aimed at establishing a protocol for the separation of H2B-GFP-retaining putative SI stem cells from H2B-GFP-retaining Paneth cells persisting in the small intestine of chased mice. After both 24 hours and 48 hours exposure (pulse) to the optimal concentration of doxycycline, the nuclei of all cells in SI organoids expressed H2B-GFP. During subsequent culture, in absence of doxycycline (chase period), there was gradual loss (due to cell division) of H2B-GFP. After 6 day chase, small numbers of H2B-GFP-retaining putative slow-cycling stem cells and Paneth cells were seen. Flow cytometric analyses indicated that to label SI crypt cells with UEA-I lectin is a promising approach for isolating putative SI stem cells for functional characterization. A method to induce ubiquitous transient expression of H2B-GFP in cultured SI organoids, compatible with regular organoid development, has been optimized. It allows detection, after a short chase period, of the slow-cycling H2B-GFP-retaining putative SI epithelial stem cells. The developed in vitro culture system will enable further characterisation of these cells and Paneth cells.

The SOX family of transcription factors are well-known regulators of diverse cellular events during development. SOX1, which belongs to the SOXB1 sub-family, is a key regulator of neural stem cell fate and a known specific marker of the neuroectoderm lineage. SOX1 plays an important role in early embryonic and postnatal CNS development. Recently, several studies have implicated SOX1 as a tumour suppressor gene in different cancer types. Conversely, SOX1 has also been reported to act as an oncogene in a prostate cancer model. In order to better understand SOX1 gene regulation, this project set out to gain a deeper insight into the regulation of SOX1 in the context of stem cells and cancer, and to identify potential regulatory mechanisms that can significantly regulate its function. Initially, SOX1 gene expression and its promoter DNA methylation pattern was analysed in a range of cancer cell lines to establish whether SOX1 epigenetic silencing was consistently found in cancer lines. Differential SOX1 expression across the analysed cancer cell lines suggests differential regulation of SOX1 in cancer, accompanied by cancer type dependent epigenetic silencing of SOX1 by DNA methylation. The second part of the study focused on the characterisation of the structure and expression of a newly identified SOX1 overlapping transcript (SOX1-OT), using RT-PCR and 5’5’RACE techniques. The SOX1-OT genomic locus was found evolutionary conserved across different species. SOX1-OT expression was further analysed in a human neuroprogenitor cell line across different time points of neural differentiation, highlighting its possible role in neural differentiation. Furthermore, the SOX1-OT gene expression profile was matched with SOX1 gene expression in a panel of different stem cell and cancerous cell lines. The co-expression profiles of SOX1-OT and SOX1 in stem cells and carcinogenesis indicated towards a potential role of SOX1-OT regulating SOX1 gene expression. Finally, a comprehensive bioinformatics analysis was performed to investigate evidence of SOX1 post translational modifications (PTMs). In silico prediction of phosphorylation, acetylation and sumoylation sites support SOX1 PTMs. The predicted PTMs within different SOX1 protein domains may affect its function through altering its DNA binding activities, cellular localisation and interaction with partner proteins. In conclusion, SOX1 expression in different stem and cancer cell lines is likely to be regulated by promoter DNA methylation, a long non coding RNA (SOX1-OT) and its function by different types of PTMs. These regulatory features may in the future advance the understanding of the SOX1 transcription regulatory network in stem cell developmental processes and its role in cancer development.

Cancers are amongst the leading cause of morbidity and mortality today. Besides the tremendous amount of research, it still appears to be a long way till we can fully understand the pathology and find its cure. Scientists are still striving to find out the precise pathogenesis, factors leading to progression and the mechanisms of spread of cancers. The ultimate objective is to find out how these can be prevented and treated. Many molecules are a focus of attention in this regard and CD24 is amongst them. CD24 is normally present on haematopoietic cells and embryonal epithelial cells but the expression is generally lost with cellular maturity and differentiation. Upregulation of CD24 has been documented in a large variety of cancers, besides non-malignant pathologies. In the recent literature, CD24 has been linked to significant cancer associated properies such as proliferation, metastasis and cancer stem cells. It is an interesting molecule with a very small protein core decorated with heavy glycosylation. The pattern and composition of CD24 glycosylation also varies within different tissues. This study attempted to explore the molecule a bit further regarding its structure, functions, molecular interactions and possible downstream signalling partners. The first part of the project was to evaluate the possible effects of CD24 in mediating cellular response to DAMPs (Damage Associated Molecular Patterns) in colorectal cancer cell lines. Malignancies have increased CD24 levels and there is increased tissue damage and necrosis thereby release of DAMPs (Damage Associated Molecular Patterns) in their microenvironment. The effects of CD24 on modulation of DAMPs response were analysed by exposing cells to autologous DAMPs and performing functional assays (migration and proliferation assays). The results of wound healing assay showed significant inhibition of colorectal cell migration by DAMPs but this was independent of CD24 status. Furthermore, transwell assays also showed significantly reduced directional motility, independent of CD24 status, in cells exposed to autologous DAMPs. The study also tested the effects of DAMPs on cellular proliferation in three colorectal cancer cell lines using Presto Blue assay. The results indicated a significant increase in cellular proliferation when exposed to DAMPs, irrespective of the CD24 status of the cell line. We proposed thereby that in colorectal cancer cells, CD24 does not appear to modulate cellular response to DAMPs unlike seen in immune cells. Nevertheless, DAMPs did show effects on autologous cell migration and proliferation. However it is acknowledged that the arguments can be strengthened by the use of purified DAMPs to demonstrate similar results and also by showing abrogation of effects after addition of anti-DAMPs antibodies to our functional assays. Another interesting aspect of CD24 is its localisation in the cell membrane. It is attached to the cell membrane via GPI anchor and resides in lipid rafts. Interestingly, it has no cytoplasmic or transmembrane domains that are used by most signalling molecules. It is unclear how CD24 molecule mediates diverse cellular properties and molecular responses in the absence of traditional signalling domains. The second part of this study was aimed at exploring some of the potential signalling partners and their functional relevance if any to CD24. The study explored Cten and CD24 interactions, as both molecules are proposed to be mediators of increased cellular motility. CD24 and Cten were also observed to have some common downstream signalling targets. In addition, the expression levels of both molecules also presented similar trends amid different cancer cell lines. The above observations led us to contemplate if the two molecules have mutual signalling and functional relationship. We observed that CD24 up regulation led to increased levels of Cten protein, whereas, knock down of CD24 resulted in down regulation of Cten protein. Also the results of our functional studies showed that the knockdown of Cten in co-transfection experiments abrogated the increased cell motility by CD24. Based on these observations we proposed that CD24 appears to modulate Cten levels and this regulation has significant functional relevance as well. The Co-immunoprecipitation experiments indicated that these two molecules did not seem to have physical interaction with each other, suggesting the possibility that the regulation of Cten by CD24 may be arbitrated by intermediate molecules. Similarly, we also investigated the molecular relationship between CD24, ILK and FAK using co-transfection technique. The results revealed enhanced cell migration through the membrane by CD24 but it was reduced after the knock down of both ILK and FAK. These findings provide further insight that not only these molecules are being regulated by CD24 as proposed by many recent studies but are also functionally relevant. The current study proposed that the increased cancer cell motility demonstrated by CD24 may well be mediated through FAK and ILK in addition to Cten. The argument needs further validation as the confirmation of successful transfection and expression levels by western blots are missing in this study. This is recognised as a limitation. Being heavily glycosylated CD24 is considered a mucin- like molecule and many studies point towards the possibility of its functional diversity to be related to these sugars. But to date not much is known about this aspect of CD24. Hence, the third part of our study was aimed at finding the functional and signalling significance of these sugar binding sites. CD24 has both O and N-sugar binding sites. While the molecule has multiple O-glycosylation sites, there are only two potential sites for N-glycosylation. Employing the fact that for CD24 molecule to be N-glycosylated, asparagine is a must and should be present in a specific sequence. We aimed at replacing the asparagine by glutamine, hence disabling the site to be decorated by N-sugars in CD24. We designed to mutate these N-glycosylation sites using Phusion-site directed mutagenesis kit. The procedure uses mutant primers in a PCR-based methodology to create the desired mutation. Once successful the next step of the study was to use this mutant CD24 in further experiments to explore the signalling and functional significance of these sugars in CD24 molecule. However, the results of our gene sequencing experiments showed that the attempt at generating the required clones was not successful and needed retrial. This step could not be carried out due to time limitation. In summary, it can be said the study aimed at exploring novel aspects of CD24 biology from three different perspectives with some exciting new facts and findings. Though these are undermined by a number of limitations that can be rectified as discussed in the thesis and would generate stronger and publishable contribution to the current understanding of CD24 biology.

Real time monitoring of stem cells has been a growing area of interest over the past decade because of new regenerative medicine approaches. Also, the effect of culture composition on stem cell metabolic pathways and their regulation of cellular fate are of increasing importance. In our project, non-destructive metabolomic analysis of stem cells and their physiological status during proliferation and differentiation stages will be investigated. This will be achieved through mass spectrometric analysis of released metabolites by LCMS and GC/MS; therefore stem cells should ideally be cultured in a specialized format to maintain their physiological properties during non-invasive live analysis while being compatible with real time GC/MS. To address this issue, mesenchymal stem cell attachment, proliferation and differentiation were tested in 20ml glass vials required for GC/MS analysis. Environmental conditions in which MSCs can proliferate under sealed condition and at room temperature were also optimized. Our data show the potential of MSCs to attach, proliferate and differentiate in 20ml glass vessels. Furthermore, cells could maintain their metabolic activity for approximately 6hrs not only under hypoxic but also hypothermic conditions. Therefore, live time analysis of their chemical volatiles could be performed in a format compatible for LC-MS and GC/MS which will be performed for undifferentiated MSCs and their differentiated populations (osteogenic and adipogenic lineages) in the next phase of the project. A specific treatment approach has emerged from hMSCs osteogenic study and identified a particular pathway suggested to be involved in chondrogenesis, and linked to the collagen II biosynthesis. The treatment is based on on natural products (L-lysine and ascorbic acid) and when tested on human cell cultures, was observed to significantly increase the morphological and functional markers of chondrogenesis, suggesting its potential to promote de novo cartilage formation in vitro.

The heterochromatin-associated proteins are subject to several different posttranslational modifications; hence, their level must be tightly controlled; otherwise as transcription factor co-repressor(s) complexes with these proteins, it may lead to stable silencing. An obvious mechanism to limit the expression time of a protein is to destroy it via the ubiquitin-proteasome system. FBXW7 (F-box and WD repeat domain–containing 7) is an E3-ligase targets transcriptional modulators and proto-oncogenes for degradation with crucial functions in cell-fate determination and tumorigenesis. In addition, most of current studies focused on epigenetic modifications that influence on the core histones within the euchromatin-heterochromatin transition, whereas the heterochromatin proteins and their partners’ identity remained largely unclear. Dr Nateri’s lab have recently identified several proteins which are targeted by the FBXW7 E3 ligase for the ubiquitin-mediated degradation. Among others, my study was focused on the role(s) of heterochromatin protein– binding protein 3 (HP1BP3) protein in epigenetic-reprogramming and its underlining mechanisms, including EMT and cell cycle progression, in normal and cancer cells. It's worth mentioning that apart from the single publication (Hayashihara et al., 2010a), the role of HP1BP3 was unknown when I began my project. HP1BP3 modulates the entry/exit of nucleosomal-DNA through binding to HP1α protein. HP1α is enriched in the pericentromeric heterochromatin, and it has been reported that HP1α recruitment in this region depends on SUV39H1/2-mediated H3K9 trimethylation. Widespread epigenomic alterations, occurs during cell differentiation, cell cycle progression and malignant transformation, but how epigenetic mechanisms contribute to the transcriptional reprogramming that accompanies EMT is still poorly understood. Furthermore, chromatin modulation events are important to control the cell-cycle-dependant gene expression during development and differentiation. Dysregulated expression of upstream cell-cycle regulators can affect DNA replication, repair, and/or division, leading to carcinogenic. Herein, our data show that the loss of FBXW7 mediated HP1BP3 induction alters heterochromatin states, through rescuing HP1α from its repressive function, impairing SUV39H1-mediated the methylation of histone H3 lysine 9 (H3K9me3), and stimulating the acetylation of H3K9 (H3K9ac) that lead to activation of epithelial-mesenchymal transition (EMT) pathway in Tiger skin fibroblast and HCT116 human colorectal cell lines. This induction of HP1BP3 upregulates the level of mesenchymal markers/regulators (Ncadherin, ZEB-1, Vimentin, and Snail1) in Tiger fibroblast cells while downregulating the epithelial marker (E-cadherin) and upregulating mesenchymal markers (ZEB-1, Vimentin, and Snail1) in HCT116 cells. In addition, upregulated HP1BP3 is an inducer of both G2/M cell cycle arrest and G1 to S phase transition via downregulating Cyclin B1 and SUV39H1/H3K9me3 while upregulating H3K9ac mark, in human Tiger fibroblasts and HCT116 CRC cells. Taken together, these findings point towards the important biological functions of HP1BP3 and its contribution in regulation of chromatin/EMT associated genes expression which consequently can be implicated in the pathogenesis of different types of FBXW7-mutated cancer.

A state of chronic metabolic inflammation and activation of the cell stress response in organs such as liver and adipose tissue are important pathogenic adaptations with the onset of obesity and the metabolic syndrome. The extent to which these processes are modulated by the early life nutritional experience is not well established, especially in large animal models. The overall aim of this thesis was to identify whether nutritional programming during prenatal and postnatal development enhances metabolic inflammation and cell stress response of obesity. A nutritional model of fetal growth restriction achieved by maternal nutrient restriction (NR) to 60% of requirements during late gestation (110 days to term at 147 days) in twin bearing sheep was used. Combination of prenatal and postnatal nutritional interventions were studied with the following three study protocols: 1. Offspring of twin bearing sheep born to mothers nutrient restricted or fed to appetite were separated after weaning at 3 months of age and then exposed to either restricted physical activity leading to obesity or to unrestricted activity and remained lean. 2. Following maternal NR, both twins or only one twin were reared on their mother’s milk during suckling period in order to achieve a relatively faster growth rate in the latter. 3. Twin offspring of sheep randomised to NR or feeding to requirement during late gestation were separated after birth and randomised to either formula feeding or being fed by the mother until weaning followed by obesogenic rearing. Total body weight of sheep in the obese group was raised by ~30% and was unaffected by any intervention. Obesity led to an increased insulin response to the glucose tolerance test, together with hepatic triglyceride deposition, and adipocyte hypertrophy with macrophage infiltration in omental adipose tissue. NR exacerbated obesity associated hepatic triglyceride deposition and upregulated gene expression of hepatic autophagy and omental unfolded protein response. Formula feeding of sheep offspring following NR was associated with slower weight gain and decreased gene expression for MTOR. Sheep offspring fed by mother as singleton gained weight at faster rate during suckling period as compared to offspring fed by their mothers as twins. Neither postnatal interventions exacerbated the state of obesity associated metabolic inflammation and cell stress response. It is possible that the increased hepatic autophagic gene expression is a reflection of defective autophagy and future work should include study of markers of autophagic function. Possible mechanisms of upregulated omental adipose UPR in offspring of sheep undergoing NR could include a programmed decrease in adipocyte number or selective survival of preadipocytes with effective ER stress response. Such adaptations followed by obesity would predispose the adipocytes to initiate inflammation and cell death pathways.

Stem cells may offer solutions for many health issues facing the world’s population. Early Biotech-led approaches are supporting novel mesenchymal stem cell (MSC) therapies through biomedical trials. However, their potential benefits are currently curtailed by challenges linked to high cell dose requirements which pose availability and manufacturing challenges, combined to suboptimal delivery methodologies. Whilst systemic delivery may be suitable for many pharmaceuticals, more complex and selective treatments such as emerging cell therapies require smarter targeting strategies on safety and cost/benefit grounds. Several groups are developing targeting strategies to guide stem cells to specific locations and hold them in situ whilst performing a repair. The targeting approach presented here uses superparamagnetic iron oxide microparticles (MPs) loaded within stem cells to facilitate control of the cells using magnets. Magnetic resonance imaging (MRI) can be used to monitor the loaded cells’ contribution to the repair process. Questions remain around MP safety and effects on both delivered cell therapies and the receiving patient. Presented data demonstrates labelling of MSC populations with a commercially available MP called SiMAG in two sizes (500 nm and 1000 nm). Particles were assessed for characteristics which influence their suitability for labelling and were found to have a non-uniform variable structure and size. Labelling was found to be both rapid and effective with low 10 µg/Fe/mL labelling doses able to distinguish cell populations by flow cytometry. Super-resolution microscopy, fluorescent microscopy and transmission electron microscopy were used to determine the location of particles within the cell and were noted to accumulate around the nucleus in large vesicles. Uptake into the cell was found to be influenced by serum with 10% serum resulting in a 75% drop in relative uptake over a 24 hour period. Potential sharing of MP between MSCs was investigated both qualitatively with fluorescent microscopy and quantitatively in a MSC co-culture experiment. No statistically significant sharing of MPs between MSCs could be seen to be taking place between populations. The fate of MPs within MSCs was investigated using pH nanosensors to interrogate the internal cell pH. A novel flow cytometry assay using pH nanosensors and MPs was used to examine the internal pH of large populations of cells. This yielded results which suggest a pH decrease over 4 days from pH 5.5 to 4.7 followed an increase to 5.4 by day 6. This effect is suggested to be caused by a complex pH mediated degradation of MPs followed by increase in pH due to the degraded iron overloading the cell. This degradation was carried out in simulated lysosomal conditions and found to act in a similar way. Macrophages were also used to test degradation and again they were found to reduce the fluorecense of the MPs rapidly over 7 days. The ability of MSCs to tolerate MPs without impacting cell health was probed with a range of assays. The metabolic assay Presto blue demonstrated doses of 10 µg/Fe/mL did not impact the metabolic status of the cell. This was tested with other potential surface chemistries of the same particle design and these were also found to be well tolerated. Membrane intergrity was assessed with flow cytometry for both 500 and 1000 nm SiMAG and was found to have no damaging effects present at 10 µg/Fe/mL. SiMAG 1000 nm was found to have no membrane compromising effect all the way up to 100 µg/Fe/mL. Cell identity was assessed with common MSC markers established by the Dominici position paper and no change to expression was found to occur even with repeated, high dose long-term (14 day) labelling strategies. As particles accumulate round the nucleus, deleterious effects of MP on DNA were tested using the comet assay and visual inspection with no statistically significant increase up to the maximum tested of 100 µg/Fe/mL. Similarly, no effect on cell cycle status was noticed for populations of MSCs. The retention of cell “function” was tested not only following labelling, but following hypothermic storage of cells to simulate shipping to a clinical setting. This was carried out for a range of clinically relevant cell types including mesenchymal stem cells (MSCs), cardiomyocytes (CaM) and ReNeuron neural stem cells (ReN). MSCs were found to freely differentiate to tri-lineage osteogenic, chondrogenic and adipogenic lineages. CXCR4 expression was measured as a marker of MSCs ability to home in on damage and was found to be raised in response to MP presence. CaM were found to resume beating both after hypothermic storage, as well as at high (1000 µg/Fe/mL SiMAG doses). ReN cells were found to be more sensitive to SiMAG with only 10 µg/Fe/mL doses tolerated although successful neural differentiation was still possible. The ability to culture MSCs and label them in a scalable manufacturing scenario was also examined and found to be possible. SiMAG was demonstrated to be a suitable labelling agent both for imaging as well as magnetic manipulation. Precise magnetic manipulation of labelled cells was demonstrated both as an individual cell and as a cell population moving through a simulated tissue gel. Entrapment of labelled cells from a simulated circulatory system was also shown to be possible with close to 100% of cells recruited in the first pass. The fluorophores on SiMAG were not strong enough to be visualised on their own and quantum dots were used to demonstrate successful retention of labelled MSCs in an ex vivo rat model. MRI was however shown to be a suitable method for assessing the location of labelled cell populations at even low cell concentrations ~1x106 and low SiMAG doses of 5 µf/Fe/mL. In conclusion, this study indicates that MPs represent a cell-safe and effective potential tool to better target cells. The key benefit of this cell labelling technology is in the high degree of understanding over the entire labelling process from entry through to degradation. In addition, this labelling technology has been shown to be cell-safe in a large number of cells both for physical health and basic function. Whilst this should be further explored for further, more specific applications, it makes a compelling case for SiMAG as a multi-functional tool for cell manipulation and tracking.

LIM domain containing protein 1 (LIMD1) is a tumour suppressor located at 3p21.3, a region that harbours multiple tumour suppressor genes and is commonly subject to homozygous deletions and loss of heterozygosity in many cancers. The mechanism of LIMD1 tumour suppressive activities are not fully elucidated, however to date it has been shown to bind to the retinoblastoma protein (pRb) and repress E2F driven transcription as well as being a critical component of miRNA mediated gene silencing. Recent work has also identified LIMD1 as a possible negative regulator of hypoxia inducible factor α (HIF1α) and the hypoxic response. In lung cancer, LIMD1 protein expression is down regulated in up to 79% of tumours when compared to normal tissue with gene deletion and loss of heterozygosity accounting for 32 and 12% respectively, leaving 30% of tumours with unexplained mechanism of LIMD1 protein loss. In an aim to identify other possible mechanisms of LIMD1 loss, scrutinisation of the LIMD1 promoter identified a CpG Island in the 5’ promoter region, within which a small region was found to be critical for transcriptional activation. This region was methylated in the non-LIMD1 expressing MDA-MB435 cell line, but became hypomethylated and LIMD1 expressed following treatment with the DNA methylation inhibitor 5-Aza-2’-deoxycytidine. In primary lung tumours, analysis of genomic DNA also identified increased methylation of this region as well as a reduction in LIMD1 mRNA levels when compared to matched normal lung tissue. Furthermore, in silico analysis identified a conserved binding motif for the Ets transcription factor PU.1. Experimentally PU.1 was verified as binding to the LIMD1 promoter with siRNA mediated depletion of PU.1 significantly reducing endogenous LIMD1 protein levels, thus identifying two possible novel mechanisms of LIMD1 silencing. Transcription of LIMD1, like that of other HIF1α regulatory proteins, was enhanced when cells were exposed to hypoxia (1% O2), facilitated by HIF1α binding a hypoxic responsive element (HRE) within the promoter. At the molecular level, in vivo LIMD1 forms an endogenous complex with proline hydroxylase 2 (PHD2) and the von Hippel-Lindau (VHL) protein, with LIMD1 loss decreasing the efficiency of HIF1α degradation and impeding the resultant cellular adaptation to chronic hypoxia. In summary these studies identified epigenetic silencing of LIMD1 as a possible explanation for LIMD1 protein loss in transformed cells. Furthermore, LIMD1 transcription was identified as being activated by PU.1 and enhanced by HIF1α, and a revised, LIMD1 integrated, model of HIF1α regulation is proposed.

The fatty acid amides (FAAs) family includes endocannabinoids, such as anandamide, as well as endocannabinoid-like molecules, such as N-palmitoylethanolamine (PEA) and N-oleoylethanolamine (OEA). Members of the FAA family show agonist activity at transmitter-gated channels (TRPV1), as well as peroxisome poliferator-activated receptors (PPARs). Given that FAAs appear to be hydrolysed principally through the action of the enzyme fatty acid amide hydrolase, inhibition of FAAH should lead to accumulation of a variety of FAAs. Therefore, in this study it was investigated whether pharmacological inhibition of FAAH could influence PPAR activity in SH-SY5Y human neuroblastoma cells or HeLa human cervical carcinoma cells. FAAH activity was assessed by monitoring liberation of [3H]-ethanolamine from labelled anandamide. FAAH protein and RNA expression were measured by immunoblotting and qRT-PCR respectively. Endocannabinoid levels were measured by LC-MS/MS. In order to evaluate PPAR activation, a PPRE-linked luciferase construct was co-transfected with expression plasmids for either PPAR α, β or γ. Binding to PPAR receptors was assessed with a competitor displacement assay (Invitrogen). In intact SH-SY5Y cells, sustained FAAH inhibition by URB597 (~75 %) led to accumulation of AEA, 2AG and PEA, but not OEA. Treatment with URB597, OL135 or PF750, three structurally and functionally distinct FAAH inhibitors, induced activation of endogenously expressed PPARs, while no activation was observed in FAAH-1 negative HeLa cells. Furthermore, exposure to URB597, OL135 or PF750 led to activation of over-expressed PPARs in SH-SY5Y cells. To rule out direct activation of PPARs by the FAAH inhibitors, cell-free binding assays showed that URB597, OL135 and PF750 could not bind to PPARα, PPARβ or PPARγ. Surprisingly, treatment with URB597 in HeLa cells led to intracellular accumulation of PEA but not AEA, OEA or 2AG. This might be due to inhibition of either FAAH-2 or NAAA, both of which are expressed in HeLa cells. Moreover, the presence of either URB597 or OL135 led to activation of PPARγ receptors over-expressed in HeLa cells. In conclusion, data in this study showed activation of PPAR nuclear receptors in vitro by inhibition of FAAH activity and subsequent augmentation of endocannabinoid tone. These data suggest that, at least in a model setup, it is possible to modulate the endocannabinoid tone without any previous external stimulus of their synthesis and trigger a functional effect.

A DNA double-strand break is an exceptionally toxic lesion that threatens the structural and functional integrity of the genome. In this thesis the repair of DSBs was investigated using the bacterium Escherichia coli, which repairs DNA breaks almost exclusively by homologous recombination. The studies described focus on the repair of damage induced by reactive oxygen species, but especially on the RecN protein, which is associated specifically with the repair of double-strand breaks. The RecN protein is highly conserved across bacterial species and in E. coli has been identified as a key factor in the repair of DNA breaks. In this thesis three RecN homologs were analysed. RecN from Haemophilus influenzae is shown to be capable of replacing the functions of E. coli RecN in vivo. However, homologs from Aquifex aeolicus and Bacteriodes fragilis cannot do so. Biochemical analysis of all three RecN homologs was undertaken. The H. influenzae RecN and A. aeolicus RecN were shown to have weak ATPase activity and an ability to interact with single-stranded DNA. ATPase deficient mutants of the RecN proteins were created and used to demonstrate the functional importance of the ATP hydrolysis. In the case of E. coli and H. influenzae, the ATPase defective mutants failed to function in vivo. In vitro, the ATPase deficient H. influenzae RecN mutant and a similar mutant of A. aeolicus RecN failed to interact with single-stranded DNA. These data are discussed in terms of a relationship between RecN and the structural maintenance of chromosome family of proteins. Finally, a model for RecN activity is presented based on those developed to explain the function of structural maintenance of chromosome proteins and the new data presented here. In this model, RecN is suggested to trap DNA molecules holding a break site and repair template in close proximity, facilitating the repair of DNA breaks by homologous recombination. The possibility of RecN acting as a global, damage induced cohesin is also discussed.