Search

Reactive oxygen species (ROS) are produced as by-products of cellular processes. In eukaryotes high levels of ROS cause oxidative stress, leading to intracellular damage and age-related diseases, whereas low levels of ROS are vital for many cellular functions including signal transduction and cell proliferation. It is therefore essential that cells can sense the different types and levels of ROS to ensure that they respond in an appropriate manner. Ubiquitination is a posttranslational modification which plays vital roles in fundamental cellular processes, including protein degradation and many signalling pathways. Ubiquitination of substrates occurs in a cycle involving the coordinated activity of conjugation and deconjugation enzymes and, interestingly, most ubiquitin pathway enzymes utilise catalytic cysteines for function. Recent work from our lab and others has begun to reveal that the relative sensitivity of these cysteines to oxidation is important for sensing the different types and levels of ROS. In the present study we hypothesised that deubiquitinases (dUbs), which remove ubiquitin from substrates through the activity of an active site catalytic cysteine, may also be regulated by ROS. Hence to test this hypothesis, the tractability and powerful genetic tools available in the model eukaryote Saccharomyces cerevisiae were utilised to investigate the potential regulation and function of all dUbs in this organism in response to different oxidising agents. Excitingly, an initial screen of available S. cerevisiae dUb gene deletion mutants and strains expressing epitope-tagged dUbs identified wide and varied responses of dUbs to different oxidising agents. For example, several dUbs were found to be important for cell survival under different oxidative stress conditions. Furthermore, specific dUbs were also found to be modified in response to specific oxidising agents. In particular, further investigations into specific dUbs observed that Ubp12 was reversibly oxidised into a HMW intramolecular disulphide complex in response to H2O2 but no other oxidising agents tested. Significantly, the catalytic cysteine of Ubp12 was shown to be essential for this complex, suggesting that oxidation regulates Ubp12 activity. Another dUb, Ubp2, was also found to be oxidised in response to H2O2. Other work has shown that Ubp2 and Ubp12 regulate mitochondrial dynamics, and therefore the present study led to a model [iv] suggesting that the regulation of Ubp12 and Ubp2 by H2O2 may regulate mitochondrial dynamics in response to different levels of H2O2. In contrast to Ubp2 and Ubp12, further investigations into another dUb, Ubp15, was found to be oxidised into HMW complexes in response to both H2O2 and diamide. Interestingly, these Ubp15 HMW complexes have different mobilities suggesting differences in the way Ubp15 responds to different oxidising agents. Significantly, similar to Ubp12, the catalytic cysteine of Ubp15 was essential for the formation of the H2O2-induced HMW complex. Ubp15 has previously been implicated in the regulation of cell cycle progression, and consistent with this link the present results suggest that Ubp15 may be important for regulating G1 phase arrest/delay, and for regulating the release into S phase, following H2O2 treatment. Collectively, the work described here has begun to provide insight into how cells sense and respond to different types and levels of ROS by regulating specific dUbs. Furthermore, the yeast dUbs are conserved in higher eukaryotes and hence these studies have potential implications for the regulation of fundamental processes such as the cell cycle and mitochondrial dynamics.

One of the hallmarks of cellular life is the protective membrane. The membrane not only encapsulates the nutrient rich cytoplasm, but also facilitates many important processes essential for most biological life such as the ATP producing oxidative phosphorylation, meaning that tight regulation of membrane homeostasis must be sustained. In Bacteria the membrane is also home to the final stages of peptidoglycan synthesis, thus coupling cell wall production and the membrane. The protein governing where the insertion of new peptidoglycan occurs in rod-shaped Bacteria is the actin homologue MreB (Errington, 2015). Previous studies have suggested that MreB and its homologues are involved in cell envelope homeostasis, not only through fatty acid adaptations, but through regulation of peptidoglycan degrading autolysins as well (Dominguez-Cuevas et al., 2013; Strahl et al., 2014). In this thesis I aimed to describe which homeostatic changes occur within the membrane when deleting the MreB cytoskeleton, as well as describe if these changes alter the physical parameters associated with fluidity and temperature dependent growth phenotypes. Additionally, I wanted to elucidate if the deletion of mreB and its homologues changes the sequence of actions following disruptions to membrane homeostasis such as dissipation of membrane potential, which under normal circumstances rigidifies the membrane, delocalizes multiple membrane associated proteins, and eventually induces lysis (Strahl and Hamoen, 2010). Through fatty acid analysis in multiple MreB cytoskeletal mutants I found that the lack of MreB and its homologues lead to altered fatty acid composition, which in turn causes membrane fluidity to be altered in a complex manner, indicating that MreB is important in maintaining a well-coordinated membrane homeostasis. Additionally, I examined membrane rigidification caused by membrane depolarization, and discovered that MreB was not involved, but rather electrostriction was the most likely candidate. Finally, through lysis assays and microscopy I examined membrane depolarization induced lysis. I found that the lysis is caused by MreB being delocalized, thus redirecting the cell wall synthesis machinery and mis-regulating autolysins essential for efficient membrane elongation, finally causing loss of communication between anabolism and catabolism of the cell wall. Together, these findings elucidate the important role MreB occupies in the membrane, not only guiding the cell elongasome, but in maintaining well-coordinated membrane homeostasis.

The spindle assembly checkpoint (SAC) monitors the chromosomes and kinetochore- microtubule attachment to prevent premature anaphase onset (Lara-Gonzalez et al., 2012), and this ensures the fidelity of cell division. The mitotic checkpoint complex (MCC), the core SAC effector, contains two sub-complexes, CDC20-MAD2 and BUBR1-BUB3 (Sudakin et al., 2001). However, the exact mechanism underlying the assembly of the MCC regarding when, where and how still is not fully addressed. It is believed that the formation of the CDC20-MAD2 sub-complex is an initial and essential step in MCC assembly (Sudakin et al., 2001), thus the assembly of the MCC can be depicted by the observation of the formation of the CDC20-MAD2 complex (Fraschini et al., 2001, Meraldi et al., 2004, Poddar et al., 2005). Using the Duolink based in situ proximity ligation assay (PLA), the lab has previously used individual cell analysis to show the temporal and spatial in vivo formation of the CDC20-MAD2 complex throughout the cell cycle in HeLa cells and existence of a specific prophase form of the CDC20-MAD2 complex (Li et al., 2017). In this study, we provide evidences showing that the profile of the assembly of the CDC20-MAD2 complex revealed by using the PLA can genuinely reflect the dynamic in vivo interaction of these two proteins in individual cells. We also provide evidences to support the idea that the prophase specific CDC20-MAD2 complex is functional in preventing the premature degradation of cyclin B1 in prophase which ensures the proper G2/M transition. p31comet (MAD2L1BP), as an important SAC silencer (Habu et al., 2002, Yang et al., 2007), which, in conjunction with TRIP13 (Thyroid hormone receptor interacting protein 13, or PCH-2 in C. elegans) promotes the disassembly of the MCC and the conversion of the ii C-MAD2 back to O-MAD2 for recycling (Xia et al., 2004). It has been suggested that expression of p31comet is cell cycle regulated, that there are two putative destruction motifs of D-box and KEN-box (Habu et al., 2002), and that it is an ubiquitin substrate (Udeshi et al., 2013, Wagner et al., 2011)，which implying that it is a substrate of the APC/C. However, the potential turnover property of p31comet is yet to be characterized. In this study, we have examined the expression profile of p31comet throughout the cell cycle using a specific anti-p31comet antibody fluorescent staining and have confirmed that it is cell cycle regulated. It starts to accumulate from prophase and peaks at prometaphase and is sustained until telophase and then it declined. This is supported by the observation that the level of p31comet is reduced at three hours after cells released from a mitotic block caused by nocodazole treatment and re-enters the cell cycle. The accumulation of p31comet before the prometaphase and its accumulation in response to the activation of the SAC by nocodazole treatment are the results of protein synthesis. However, the potential accumulation of p31comet levels in cell samples after siRNA with the components of the APC/C, especially the APC3, examined by western blot produced ambiguous inconclusive results. The preliminary data produced by depletion of Cullin-1, another E3 ligase responsible for the degradation of cell cycle substrates in G1/S, using siRNA resulted in an increase in p31comet, which implies that Cullin-1 is potentially response to p31comet degradation though this would require further confirmation. The various kinds of cell death occurred during mitosis are collectively termed 'Mitotic Catastrophe'(Vitale et al., 2011). Targeting the mitotic catastrophe signaling pathway in cancer offers a significant therapeutic advantage, and the anti-mitotic drugs such as taxanes, epothilones, and vinca alkaloids are widely used in the cancer clinic. M2I-1 (MAD2 inhibitor-1), a small molecule, has been shown to disrupt the CDC20-MAD2 interaction in vitro and weaken the SAC in vivo(Kastl et al., 2015). We report here that M2I-1 can disrupt the in vivo interaction of CDC20 and MAD2, and significantly increases the sensitivity of several cancer cell lines to anti-mitotic drugs, with cell death occurring after a prolonged mitotic arrest, when cell death is triggered by the premature degradation of cyclin B1 and the perturbation of the microtubule network by nocodazole or Taxol. Interestingly, the level of iii MCL-1, a pro-survival protein of the Bcl-2 family is significantly elevated, but higher levels of the pro-apoptotic proteins MCL-1s, a short form of the MCL-1, act antagonistically. Taken together, our results demonstrate that M2I-1 exhibits antitumor activity in the presence of current anti-mitotic drugs such as Taxol and nocodazole and has the potential to be developed as an anticancer agent.

The aim of this thesis was to assess the safety profile of different novel nanoparticles when interacting with lung bronchial epithelial cells in vitro, especially their role in inflammation and cytotoxicity. In addition, the permeation behaviour of self-nanoemulsifying drug delivery systems (SNEDDS) was measured by a novel mucus diffusion model utilizing standardized transwell diffusion plates. Methods: Primary bronchial epithelial and Calu-3 cells were cultured in 24 and 96 well plates. Different formulations of nanoparticles (NPs) were used to measure IL-8 production by cells as a marker of inflammation (measured by sandwich ELISA). The Celltitre blue (CTB) and MTT assays were used to test cell viability. The effect of cell density on the CTB assay was studied (500k, 400k, 300k, 200k, 100k, 50k, 42k, 33k, 25k, 17k, and 8.3k) in 24 well plates. Permeation studies were examined by using a mucus diffusion model. The rheology of the mucus was investigated to evaluate the gel structure of the mucus. Results: The viability of cell exposed to 52nm polystyrene NPs showed no statistical difference between cells with NPs and cells without NPs after 24 hour of exposure. The nanoparticles were shown to interfere with both the CTB and MTT assays. The standard curves in the CTB assay vary as the cell density decreased. Although IL-8 was shown to be slightly increased in human primary cells after exposure to different concentrations of NPs, IL-8 increased significantly after exposures to the same proportion of sterile water as these nanoparticles were delivered in (a control for osmolarity). The composition of SNEDDS might play a major role in their capacity to permeate the mucus such as lauroglycol. The small intestine mucus gels have some degree of frequency dependency; between 0.1 and 1 Hz, there was a decline in both iv G` and Gʺ followed by a rise over 1-10Hz. Gʺ increased relative to G` suggesting a tendency to flow. This change at high frequency is verified by an increase in the phase angle, demonstrating a weaker gel. This biphasic behaviour may be explained by the gel forming interactions having time to break and make at a low frequency but at a high frequency, they do not have time to reform. Conclusion: I have found that both CTB and MTT assays may not be sensitive enough to test viability of cells exposed to nanoparticles. Therefore, other assay systems are required to test these NPs with respect to cytotoxicity. The composition of the SNEDDS played an important role in determining the permeation of the mucus gel layer.

Mitochondrial DNA (mtDNA) mutations accumulate somatically with age in normal individuals causing defects in oxidative phosphorylation (OXPHOS), with no selective pressures promoting their accumulation or removal. mtDNA mutations are also inherited through the germline causing mtDNA disease. In contrast to somatic mutations, they are lost in blood over time, suggesting selective pressure against them. This project aims to further investigate the selective loss of inherited mtDNA point mutations in mitotic tissues, using gastrointestinal epithelium as a model. I have determined the mtDNA mutation load and OXPHOS protein levels in various gastrointestinal segments of patients with the inherited m.3243A > G and m.8344A > G mtDNA mutations. The data shows selective loss of these mutations in the mitotic epithelium, but not the post-mitotic smooth muscle and suggests that the selection occurs in the gastrointestinal stem cells. I have further characterised a mouse model of mitochondrial disease carrying the inherited m.5024C > T mutation and have used them to study potential mechanisms of selective mtDNA mutation loss. These mice show: (1) loss of the m.5024C > T in the mitotic spleen and gastrointestinal epithelium with age but not in the post-mitotic tissues; (2) the rate of mutation loss is related to the rate of epithelial cell turnover; (3) the mutation loss in the intestinal epithelium is consistent with maintained levels of OXPHOS proteins with age, in contrast to the accumulation of cells with OXPHOS defects in aged wild-type mice; (4) significantly upregulated mitochondrial proliferation in the intestinal epithelium. These key findings highlight differences in the selective pressures on inherited and somatic mtDNA mutations in mitotic tissue, and the contrasting tissue-specific dynamics of inherited mtDNA mutations in mitotic and post-mitotic tissues. These findings are important to further the understanding of phenotype development, progression and potential treatment of mtDNA disease, and the effect of mtDNA mutations on ageing stem cell biology.

Low back pain affects most people at some point in their lives, and intervertebral disc degeneration is a common cause of this pain. The nucleus pulposus is the first structure to fail in intervertebral disc degeneration. Gaining a greater understanding of the nucleus pulposus cell and its response to mechanical stimuli may be useful in tailoring the next cell-based regenerative therapy and preventative therapies. The nucleus pulposus is a highly hydrated, porous tissue and changes in spinal loading create fluid movement throughout the tissue, which causes fluid shear stress on the tissue and cells. This research has investigated shear stress rates of 0.1, 1.0 and 4.0 dyne/cm2 on bovine nucleus pulposus cells. Steady and pulsed flow were studied over 1, 4, 8 and 24 hours using Ibidi VI0.4 chambers. Results showed shear stress affects the morphology and gene expression of bovine nucleus pulposus cells. Morphological changes associated with cell detachment and reduced proliferation were seen in cells exposed to 1.0 dyne/cm for 4 and 8 hours and 4.0 dyne/cm2 for 1, 4 and 8 hours. After 24 hours in 1.0 dyne/cm2 and 4.0 dyne/cm2 all cells had detached from the Ibidi VI0.4 slides. There was an increase in cell proliferation in cells exposed to 0.1 dyne/cm2 for 4 hours in steady flow and no change in cell number compared to control at 1, 8 and 24 hours in steady and pulsed flow. Changes in gene expression in 4.0 dyne/cm2 flow were consistent with changes associated with catabolic metabolism; downregulation of collagen 2 and aggrecan and upregulation of aggrecanase 1 and 2. Results comparing steady and pulsed flow and gene expression changes at 0.1 and 1.0 dyne/cm2 were inconclusive, further studies with a larger sample size could help to clarify these effects. These findings are important in guiding researchers in the development of optimal perfusion bioreactor conditions for cell-seeded nucleus pulposus regeneration therapies. The implications for research to increase our understanding of fluid flow through the intervertebral disc during exercise as a preventative measure for disc degeneration are also discussed.

The human mitochondrial genome is approximately 16.6kbp and encodes 22 tRNAs, 2 rRNAs and 13 polypeptides. These polypeptides are synthesised by the mitoribosome, composed of a large (39S) and small (28S) subunit. Although recently published cryo-EM structures of the human, porcine and yeast mitoribosomes have identified a number of significant differences including new components with uncharacterised functions. One such component was mS38 (AURKAIP1) as a part of the mt-SSU of the mitoribosome. This was consistent with previous work in my host laboratory using a siRNA screen of candidate mitochondrial proteins, which identified AURKAIP1 as having a role in mitochondrial gene expression. I continued this work and confirmed levels of AURKAIP1 are tightly controlled within the cell. Overexpression of both AURKAIP1 and AURKAIP1-FLAG caused disruption to the mitoribosome and mitochondrial translation, and depletion of AURKAIP1 impaired cell growth and resulted in cell death. Although the exact function of AURKAIP1 is still elusive, AURKAIP1 may have a role as an RNA chaperone. When overexpressed, AURKAIP1 was cleaved by the mitochondrial matrix protease LONP1, from a 25kDa to a 15kDa species by two cleavage events, although this could not be determined in the endogenous species. This work led to the characterisation of LONP1 variants that were believed to be pathogenic. Whole exome sequencing identified two novel mutations (p.Tyr565His; p.Glu733Lys) in LONP1 in a paediatric patient with a severe mitochondrial disease. By cloning and sequencing LONP1 from patient samples, I showed compound heterozygosity in the patient for 2 independent mutations, confirming the disorder to be autosomal recessive. Skeletal muscle showed severe defects in OXPHOS complexes I, III and IV, reduced mtDNA copy number and a decrease in the steady state levels of TFAM. Steady state levels of LONP1 remained unaffected. The p.Tyr565His mutation is localised to a highly conserved aromatic-hydrophobic (Ar-φ) motif present in hexameric ATP-dependent proteases and in vitro assays indicated a retained ability to hydrolyse ATP but a loss of proteolytic activity. This suggests that the mutation is likely to prevent the substrate from reaching the proteolytic site.

Ribosome biogenesis is a tightly regulated process that is essential for cell survival. Genetic mutations in genes encoding ribosomal proteins or ribosome biogenesis factors lead to the development of rare genetic disorders, known as ribosomopathies. The 5S RNP is a ribosome assembly intermediate composed of the 5S rRNA, ribosomal protein RPL5 and RPL11. During the 5S RNP assembly, the newly transcribed 5S rRNA is stabilized by RPL5 first, before RPL11 binds. Once it is assembled, the 5S RNP is incorporated into the large ribosomal subunit precursor. Ribosome biogenesis defects, which induce ribosomal stress, cause the accumulation of the 5S RNP. The accumulated non-ribosomal 5S RNP induces tumour suppressor p53, by binding and inhibiting its main negative regulator, MDM2. It has been suggested that any factor involved in the 5S RNP assembly and its integration into the ribosome will affect p53 signalling. I therefore set out to evaluate factors that have been proposed to be connected to the 5S RNP. Different proteins, including HEXIM1, Mybbp1a, B23, NML and SRSF1, have been suggested to be involved in the 5S RNP-p53 signalling, therefore I aimed to investigate their roles in this signalling pathway. Furthermore, I also aimed to investigate the role of RPL11 phosphorylation in ribosome biogenesis and the 5S RNP-p53 signalling. In this work, I present evidence supporting the proposed role of splicing factor SRSF1 in 5S RNP assembly. I also demonstrate that SRSF1 could be a part of the 5S RNP complex that is essential for p53 activation during ribosomal stress, while protein HEXIM1, Mybbp1a and NML are not needed for this pathway. I further provide evidence suggesting the threonine 73 phosphorylation in RPL11 could control its binding to the 5S RNP, whereas serine 51 phosphorylation in RPL11 could be important in the 5S RNP incorporation into the large ribosomal subunit. Last but not the least, my results also suggest that serine 140 in RPL11 could be transiently phosphorylated during the ribosomal maturation process. In conclusion, this work addresses the important role of the 5S RNP, in both ribosome biogenesis and the 5S RNP-p53 signalling.

Within ciliates (protozoa of the phylum Ciliophora), anaerobic species are widespread and typically possess organelles which produce H2 and ATP, called hydrogenosomes. Hydrogenosomes are mitochondrial homologues and are a product of evolutionary convergence, having been found in wide-ranging and diverse anaerobic eukaryotes. Ciliates seem to have evolved hydrogenosomes on multiple occasions from the mitochondria of their aerobic ancestors. The hydrogenosomes of the ciliate Nyctotherus ovalis were studied in detail previously but little is known about the hydrogenosomes from other ciliate species. In the present study seven species of ciliate, Cyclidium porcatum, Metopus contortus, Metopus es, Metopus striatus, Nyctotherus ovalis, Plagiopyla frontata and Trimyema sp. were cultured and their hydrogenosomes were investigated using genomic and transcriptomic sequencing from whole genome amplifications from single and small numbers of isolated cells. The data were then used to reconstruct putative hydrogenosome metabolic pathways. Components of these pathways are typically encoded by the ciliate nuclear genomes but Nyctotherus ovalis, Metopus contortus, Metopus es, Metopus striatus and Cyclidium porcatum have also retained mitochondrial (now hydrogenosomal) genomes which were sequenced for the first time. The most complete of these genomes were from Nyctotherus ovalis and Metopus contortus. These have both retained genes for proton-pumping subunits of the electron transport chain Complex I and ribosomal subunits needed for their synthesis. The ciliates Plagiopyla frontata and Trimyema sp. appear to have completely lost the organelle genome during the conversion of mitochondria into hydrogenosomes. The ciliate hydrogenosomes for which the most data was obtained appear to have retained some of the enzymes needed to produce energy by substrate-level phosphorylation but some species have also retained a partial electron transport chain and Cyclidium porcatum has retained nuclear encoded subunits of the mitochondrial F1F0 ATP synthase complex. Nuclear genes encoding enzymes that play a key role in H2 production, FeFe-hydrogenase, pyruvate: ferredoxin oxidoreductase and pyruvate: NADPH+ oxidoreductase, were also sequenced from the sampled ciliates and their evolutionary origins were investigated using phylogenies. These suggest that ciliate FeFe-hydrogenases are monophyletic and iii have a separate bacterial origin from FeFe-hydrogenases in other eukaryotes. No evidence was found to support an alpha-proteobacteria or mitochondrial ancestry of these enzymes as predicted by the Hydrogen Hypothesis (Martin and Müller, 1998). Each of the ciliates investigated contained methanogenic Archaea endosymbionts, which can consume the H2 produced by the hydrogenosomes. Some of these endosymbionts were identified to the species-level. The associations they have formed with their hosts appear to be stable over short time-scales but not over longer evolutionary periods, as closely related ciliates like Nyctotherus ovalis and Metopus contortus do not have closely related endosymbionts, providing no evidence for long-term co-speciation.

Mitochondria are dynamic organelles whose principal role is the generation of cellular energy (ATP) through oxidative phosphorylation (OXPHOS). 13 OXPHOS subunits are encoded by the mitochondrion’s polyploid circular genome (mtDNA), and the nuclear genome (nDNA) encodes the remaining subunits as well as proteins required for mtDNA maintenance. In addition to mitochondrial number, mtDNA copy number (mtDNA CN) varies between cell and tissue type, depending on metabolic demand and baseline mtDNA quality, and ranges from hundreds to thousands of copies per cell. mtDNA CN is often linked to mitochondrial dysfunction and the ubiquity of mitochondria results in a broad spectrum of dysfunction and clinical phenotypes; ranging from primary mitochondrial disorders to complex diseases such as cancer, type 2 diabetes, and Parkinson’s disease. Given the variability in mtDNA between individuals, it is possible that mtDNA CN is influenced by secondary factors. I hypothesise that nDNA diversity is a major component of mtDNA variability between individuals and will test this hypothesis by conducting a genome wide association study (GWAS) in a large, European, asymptomatic cohort (>8000 individuals), comparing nDNA genotype to mtDNA copy-number as a QTL. Peripheral blood mtDNA CN was correlated to array-based and imputed nDNA genotype in a two-stage QTL analysis, utilising three independent replicative cohorts: UKBS, Newcastle, and ALPAC. In addition the effect of potential confounding biological variables such as age, gender, blood count, and potential methodological confounders such as assay variation, technical and biological replicate numbers, and differences in genotype platform were all assessed and used to improve the GWAS analysis. Individual cohort analysis identified nuclear gene UNC13C (Unc-13 Homolog C), two intergenic, and one intronic SNP, which is in close proximity to PSMD3 (Proteasome 26S Subunit, Non-ATPase 3), to be genome wide significant (GWS) (p < 1.00E-07) in individual cohort analysis. However these hits could not be replicated in meta-analysis. mtDNA variant analysis in all three cohorts revealed that mtDNA SNPs G5460A and G5046A, which identify as mitochondrial haplogroup W, were significantly associated to a significant reduction in mtDNA CN. Furthermore, our work identified gender-specific genetic differences, which was supported by a Preliminary iv significant decrease in mtDNA CN in males with age, but not females, and significant changes in mtDNA CN relative to blood cell type and proportions highlighted the importance of regulating for cellular heterogeneity. Additionally, no difference in mtDNA CN was observed between pre- and post-menopausal women. This work indicates that there are likely genetic variants present at the population level modulating mtDNA CN, but that this process is complex and multifaceted.

The concentration of intracellular free Ca2+ ([Ca2+]i) plays an essential role in cell cycle progression. Understanding the Ca2+ signalling mechanisms that drive and maintain cell cycle arrest in the quiescent state, or stimulate quiescent cells to re-enter the cell cycle and proliferate, will be crucial for developing therapeutic potentials for highly detrimental diseases such as cancer and aging disorders. The work in this thesis investigates remodelling of the Store-Operated Ca2+ Entry (SOCE) signalling pathway that is associated with cell cycle arrest in the quiescent G0/G1 phase. Serum starvation was used to induce cell cycle arrest in quiescent G0/G1 phase to allow investigation of remodelling of SOCE when cells exit the cell cycle. Cancer HeLa, precancerous NIH 3T3 and immortal non-cancerous hTERT RPE-1 cell lines were used in this comparative study. Serum starvation induced cell cycle arrest in G0/G1 phase in HeLa and NIH 3T3 cells that was accompanied by a marked down regulation in SOCE. In hTERT RPE-1 cells, serum starvation induced a modest down regulation in SOCE that was not accompanied by cell cycle arrest. Further experiments revealed that SOCE downregulation was attributed to changes in expression and localisation of the Ca2+ sensing protein STIM1 and the Ca2+ release-activated channel (CRAC) protein Orai1. Serum was added-back in order to induce arrested cells to re-enter the cell cycle. Cell cycle re-entry was associated with restoration of SOCE, STIM1 and Orai1 expression in NIH 3T3 cells, and the restoration of only Orai1 expression in HeLa cells. By comparing these changes to that of hTERT RPE-1 cells, which were not arrested in G0/G1 cells, it appears that there may be a significant role for cell cycle arrest in quiescent G0/G1 phase in SOCE remodelling. In addition, these results suggest an additional role for Orai1 as a positive regulator of cell cycle progression in HeLa cells. In NIH 3T3 cells, SOCE and its proteins, STIM1 and Orai1, were remodelled with cell cycle arrest in quiescent G0/G1 phase and restored with cell cycle re-entry, indicating that each of STIM1 and Orai1 appears to have a role in SOCE and cell cycle progression. SOCE seems to be coupled to cell proliferation in NIH 3T3 cells but not in HeLa cells. This study provides evidence that there are significant differences in SOCE remodelling with cell cycle arrest in quiescent G0/G1 phase in cancerous and pre-cancerous cells, and identifies potential targets for drug therapy aimed at regulating the cell cycle.

The disposable soma theory predicts that levels of maintenance and repair are influenced by resource availability. The interplay of cell signalling response to stress and damage with nutrient sensing networks presents an opportunity to further our understanding of how this influence is achieved. A systems modelling approach is applied to address three related problems, allowing an in-depth examination of underlying biochemical dynamics of these complex networks. p53 is a well-known regulator of G1 phase cell-cycle checkpoints. As part of a collaborative effort I have examined its role in regulating p21 senescence signalling in response to DNA damage. In support of a computational model, I generated experimental data showing that lower doses of radiation, applied over a longer time period, were able to produce a similar population probability of senescence as a single high dose. The activation of mTORCl by amino acids is well established, but the effect of amino acids on the rest of the Insulin/IGF signalling pathway is less clear. Again in a collaborative effort, we generated models representing multiple possible regulatory inputs of amino acids into Insulin/IGF signalling. These calibrated models underwent a combinatorial selection and merging process, based on comparative goodness of fit, determining the most likely candidate inputs to be IRS1, AMPK and mTORC2 which were then confirmed experimentally. FOXO3a transcription factor sits at a junction, regulated by ROS, IGF (via AKT) and DNA damage (via p53 and ATM) and regulating energy availability (via AMPK) and senescence (via p53). Using a systems modelling approach I explore the dynamics of IGF/FOXO signalling under conditions of chronic oxidative stress and find a novel activation of mTORCl by ROS via TSC1/2. Together this thesis represents an in-depth study of the interplay between nutrient, stress and damage signalling, revealing potential dynamics for the effect of nutrients and chronic stress on the early stages of cellular senescence.

The eukaryotic cell cycle is driven by a set of kinases called the cyclin-dependent kinases (CDKs). During G1, CDK2/cyclin E phosphorylates the CDK inhibitor, p27, leading to its ubiquitination by the SCFSkp2 complex and subsequent degradation. Skp2 is the substrate recognition component of the SCFSkp2 complex and has been found to be up-regulated in many cancers. High levels of Skp2 indicate poor prognosis. Its overexpression often correlates with low p27 levels. Skp2 was first identified in a complex with CDK2, cyclin A, Cks1 and Skp1. CDK2/cyclin A binds to SCFSkp2 and is required for the recruitment of p27 to the SCFSkp2 complex. The aims of this project were to structurally characterize the CDK2/cyclin A/Skp1/Skp2 complex and to identify the functional significance of the Skp2/cyclin A interaction. In the absence of a crystal structure, site-directed mutagenesis was used to identify the regions on cyclin A and Skp2 which mediate this interaction. Mutation of residues 244-247 of cyclin A blocked its interaction with Skp2. Mutation of this region did not interfere with binding to CDK2, or p27, and CDK2/cyclin A kinase activity did not appear to be affected. CDK2 was found not to have a role in binding to Skp2. Cyclin A binds to an extended region within the N-terminus of Skp2. Residues Leu32, Leu33, Ser39 and Leu41 of Skp2 have been reported to comprise the cyclin A binding region (Ji et al., 2006). This site was confirmed and a further two residues were identified as contributing to the interaction. The Skp2 N-terminal region is involved in binding many regulatory proteins and is heavily post-translationally modified. The binding of cyclin A to this sequence may therefore affect the structure and/or accessibility to modifying enzymes. The characterisation of these mutants has shown that they are potentially useful tool mutants for studying Skp2 function and regulation in human cell-based assays.

Cell wall is a unique and essential component of bacterial cell. It defines cell shape and protects cell from bursting through its own internal osmotic pressure. It also represents a significant drain on the cells resources, particularly in Gram positives, where the wall accounts for more than 20 % of the dry weight of the cell, and approximately 50 % of ‘‘old’’ cell wall is degraded and new material made to permit cell growth. After the discovery of penicillin, there has been active study of bacterial cell wall structure and metabolism, as it represents the major target for antibacterial compounds. The biosynthetic pathways for cell wall precursors has been well investigated in bacteria generally, but the coordination of cell wall metabolic processes and the fate of turnover cell wall materials have only been well characterised in Gram-negative bacteria (e.g Escherichia coli). In Gram-positive bacteria, it has generally been accepted that the old wall is released from the surface and lost to the environment during growth, with apparent recycling of this material during stationary phase for Bacillus subtilis. It is also known that the Gram-positive wall is subject to significant post-synthetic processing, involving the linkage of wall teichoic acids and the cleavage of molecules from the structure, e.g. D-alanine, although the function of these is unclear. Understanding the importance of these processes has relevance for both the pathogenicity and biotechnological use of bacteria, as well as for understanding bacterial cell biology. As it is known that the peptidoglycan fragments (e.g muropeptides) induce the innate immune response in higher organisms and so act as a signal for infection, particularly for Gram-positive bacteria. Thus, understanding how they are generated and recycled by the bacteria may offer potential insights into novel therapeutics, also the accumulation of cell wall muropeptides should be avoided in biotechnological products. In this thesis, the D-alanine metabolism was manipulated to understand the mechanistic details of cell wall metabolism and D-alanine recycling in B. subtilis, using genetic, biochemical, bioinformatics and fluorescent microscopy approaches. Through these analyses, a D-alanine transporter (DatA, formerly YtnA) was identified by genetic screening. The roles of DatA and the carboxypeptidases, LdcB and DacA, in recycling of cell wall derived D-alanine have experimentally been confirmed. We also found that D-alanine aminotransferase (Dat) can act to synthesis D-alanine under certain conditions. From the data obtained a model for peptidoglycan assembly (coordinated synthesis and turnover) during growth of B. subtilis has been developed to take into account the various aspects of cell wall metabolism.

Maintenance of redox balance is essential for fundamental physiological processes, including growth, differentiation, migration, and circadian rhythms. However, the mechanisms by which cells maintain this balance and sense acute, dynamic redox changes under signalling and/or stress conditions are poorly defined. 2-cys peroxiredoxins (Prxs) are abundant, ubiquitous thioredoxin peroxidases with important roles in ageing and cancer. Counter-intuitively, Prxs are sensitive to hydrogen peroxide (H2O2)-induced inhibition of this peroxidase activity by hyperoxidation to a thioredoxin-resistant form. Prx hyperoxidation has been proposed to have various functions in signalling and protein homeostasis. However, the circumstances which cause Prx to become hyperoxidized in vivo remain unclear. The fission yeast Schizosaccharomyces pombe contains a single, well-studied 2-Cys Prx with important roles in responses to H2O2. Using quantitative in vivo and in vitro kinetic data, multiple mathematical models have been developed to investigate when the S. pombe peroxiredoxin Tpx1 becomes hyperoxidised in vivo. This approach suggested that Prx hyperoxidation occurs when the peroxide-buffering capacity of a cell becomes saturated. This was confirmed experimentally using both S. pombe and human cell lines, which suggested that the thiol-proteome is responsible for this peroxide buffering capacity. Accordingly, we propose that Prx hyperoxidation signals that the cells’ antioxidant defences are overcome and that repair mechanisms need to be deployed to limit damage and restore homeostasis. We have also used an integrated modelling and experimental approach to investigate the roles of Tpx1 in promoting the H2O2-induced oxidation (activation) of the AP-1-like transcription factor Pap1. This has revealed that Tpx1 has 2 roles, participating as a direct H2O2-transducer to initiate Pap1 oxidation and also by competitively inhibiting the thioredoxin-like protein Txl1 from reducing Pap1. Finally, we use a simple computer model representing multiple cellular peroxidase processes to demonstrate that changes in gene expression in response to H2O2 limit H2O2-induced damage by improving the cells’ ability to inhibit increases in intracellular H2O2 concentration. Together this work has provided new insights into the cellular mechanisms responsible for the regulation of cellular responses to H2O2.

There are considerable differences in the timing, type and extent of age-related decline between individuals who share the same chronological age, which may be driven by a combination of genetic, stochastic and environmental factors. Biomarkers of ageing (BoA) that can discriminate between individuals who differ in their biological age will therefore be useful to understand biological mechanisms, develop and test interventions and allow the prediction of age-related events so interventions can be implemented. In recent years, a variety of mechanistic candidate BoA have been discovered on the basis of a greatly improved understanding of the cellular and molecular biology of ageing. These include various measures of oxidative stress, which is thought to contribute causally to the ageing of organisms via its acceleration of cellular senescence. However, their reliability and validity as BoA, especially within population based cohorts are scarce. This study therefore focused on various oxidative stress-related measures as candidate BoA including: reactive oxygen species (ROS) production from dysfunctional mitochondria, by measuring superoxide levels, mitochondrial mass and mitochondrial membrane potential in blood mononuclear cells by flow cytometry; and also markers of lipid peroxidation, F2-isoprostanes, by measuring 8-iso Prostaglandin F2 by Automated Dissociation Enhanced Lanthanide Fluorescence Immunoassay (AutoDELFIA). Despite providing evidence of experimental reliability for all measures and also some evidence of construct validity for ROS production from dysfunctional mitochondria in terms of: associations with chronological age, associations with some markers of oxidative stressinduced cellular senescence, validation in a dietary restricted animal model of ageing and a role in an immunosenescent phenotype; there was no evidence of predictive validity in terms of longevity or age-related health outcomes in a population based cohort of the very old, the Newcastle 85+ study. This questions the predictive validity of these parameters as candidate BoA in the very old population.

Cellular senescence, the irreversible loss of proliferating capacity of somatic cells, is an important tumour suppressor mechanism but also driver of ageing. These somehow contradictory functions are dependent on the development of the so-called senescent phenotype, which involves over-production of pro-inflammatory and pro-oxidant signals, however the exact mechanisms underlying its induction remain incompletely understood. In this thesis we aimed to understand how mitochondria and pro-inflammatory factors interact during senescence and how they contribute to the senescent phenotype. Firstly, we show that mitochondria are critical for the establishment and maintenance of cell senescence. Elimination of mitochondria rejuvenated senescent human fibroblasts, abrogating the pro-inflammatory phenotype, heterochromatin foci and expression of cyclin-dependent kinase inhibitors p21 and p16. Importantly, a considerable percentage of these cells were able to resume proliferation. Mechanistically, we show that mTORC1 integrates signals from the DNA damage response towards PGC-1β-dependent mitochondrial biogenesis, playing a causal role in the development of senescence. Secondly we show that inhibition of IL-8, a prominent proinflammatory cytokine of the SASP, partially abrogated the senescent phenotype by reducing mTOR-dependent mitochondrial mass and ROS production during senescence. Finally, we demonstrate that inhibition of mitochondrial content in vivo by either rapamycin or PGC-1β deletion prevents age-dependent increase in senescence in mouse liver. Our results suggest mitochondria as an important target for interventions aiming to reduce the load of senescent cells in ageing tissues.

Mitochondria have their own protein synthesis machinery that synthesises the oxidative phosphorylation components encoded by their mtDNA. This translation process consists of four main phases: initiation, elongation, termination and ribosome recycling. Termination and its control have been the least investigated. Recently, however, the termination factor, mtRF1a, has been characterised as sufficient to release all the nascent proteins from the mitoribosome. Furthermore, bioinformatics has identified three additional members of this mitochondrial release factor family namely, mtRF1, C12orf65 and ICT1. The latter is now known to be incorporated into the mitoribosome but its exact function remains unclear. My project has therefore focussed on characterising the remaining two factors; mtRF1 and C12orf65, and investigating their possible involvement in mitochondrial protein synthesis. It has been demonstrated that protein synthesis is not perfect and bacterial ribosomes not infrequently stall during translation. This can result from limiting amounts of charged tRNAs, stable secondary structures, or truncated/degraded transcripts. Ribosome stalling has been shown to cause growth arrest. In order to prevent that and maintain high efficiency of mitochondrial protein synthesis such stalled complexes need to be rapidly recycled. Bacteria have developed at least three distinct mechanisms by which ribosomes can be rescued. Contrastingly, despite the presence of truncated mRNAs in mitochondria, no such quality control mechanisms have been identified in these organelles. This study investigates the potential role of mtRF1 and C12orf65 in quality control of protein synthesis in mitochondria. Both mtRF1 and C12orf65 demonstrate conservative motifs which would suggest their potential role in ribosome rescue. My findings indicate that the conserved motifs in mtRF1 are crucial to maintain normal cell metabolism and that its mutated forms negatively affect cell growth. Since these motifs are required for ribosome dependent peptidyl-tRNA hydrolysis, the data presented strongly imply that mtRF1 plays a crucial role in intra-organellar protein synthesis.

The genome of the fission yeast Schizosaccharomyces pombe is host to a family of long terminal repeat (LTR) retrotransposons called Tf2, which is comprised of 13 full length elements and around 250 solo LTRs. The expression of these elements is subjected to chromatin based silencing during normal growth conditions but is induced in response to hypoxia via the action of Sre1 , a homologue of human SREBPtranscription factor. The aim of this study was to dissect the mechanisms that regulate Tf2 expression and to determine how they contribute to the control of element mobilisation. Characterisation of the Tf2 LTR element identified sequences that mediate its transcriptional control, while further studies identified roles for the Spt6 and Asf1 histone chaperones in Tf2 element silencing. Furthermore, similar transcription controls were shown operate at the related Tf1 LTR retrotransposons that are present in the genomes of other wild type s. pombe strains. To analyse the effects of transcriptional regulation upon Tf2 element propagation, an assay was established to measure the mobilisation frequency of a marked endogenous Tf2 element (Tf2-12natAI). Using this system the affect of key transcriptional regulators upon Tf2 mobilisation frequency was determined. The HIRA nucleosome assembly complex has previously been shown to be required for Tf2 silencing and accordingly element expression was found to be increased >10-fold in the absence of HIRA. Despite this finding only a marginal (1.8-fold) increase in the frequency of Tf2 mobilisation was observed. In contrast, expression of a constitutively active form of the transcription factor Sre1 (Sre1-N) resulted in a large increase in mobilisation (>10-fold) despite inducing Tf2 expression to less than 50% of that observed in the absence of HIRA. Furthermore, loss of the CENP-B homologue, Abp1 resulted in only a small increase in Tf2 expression (3-fold) but a significant increase in mobilisation (4-fold). Therefore, mobilisation frequency is not necessarily correlated with expression levels, suggesting that Tf2 mobilisation is subjected to additional controls. These studies also revealed that the Asf1 histone chaperone restricts Tf2 element propagation, while components of the RNAi pathway promote Tf2 mobilisation. In an effort to identify novel regulators of Tf2 expression, the homologues of the s. cerevisae bromodomain containing AAA-ATPase Yta7 were characterised. s. pombe contains two Yta7 homologues (SPAC31G5.19and SPBP22H7.05c)which were named Yta71 and Yta72 respectively. Yta71, but not Yta72, was found to restrict the expression of Tf2 and Tfi elements and their solo LTRs. Deletion of yta71+ was also found to result in sensitivity to the spindle poison thiabendazole and an increased level of chromosome loss. Consistent with these findings Yta71 is required for the integrity of centromeric heterochromatin and also the heterochromatin associated with the mating type (mat) locus. Thus Yta71 is required for transcriptional silencing at multiple loci in fission yeast.

Cyclin-Dependent Kinases (CDKs) play a fundamental role in eukaryotic cell cycle progression, particularly at cell cycle checkpoints, and are therefore important targets for anticancer drug discovery. Activation of CDK2 in complex with Cyclin A regulates entry into S phase of the eukaryotic cell cycle. CDK7, a dual-function enzyme, acts both as a CDK-Activating Kinase (CAK) and as a component of the general transcription factor TFIIH. However, experiments with MAT1-knockdown mice have shown that cell cycle arrest by CAK inhibition would not be detrimental for transcriptional activity in non-dividing cells, as CDK9 in complex with Cyclin T can perform transcriptional duties in the absence of TFIIH. Previous studies have resulted in the identification of NU6102 (1, IC50 mM = 0.005 (CDK2), 4.4 (CDK7)) as a potent and selective CDK2 inhibitor, and NU6247 (2, IC50 mM = 0.12 (CDK2), 0.23 (CDK7)) as an equipotent CDK2/7 inhibitor. It was shown that the sulfonamide group of 1 confers potency and selectivity for CDK2, whereas the pendant piperazinyl substituent of 2 diminishes CDK2 actvity whilst improving activity versus CDK7. S NH N N NH N O O O 2 N N AccordinglyAs part of the work described in the present thesis, sulfonamide 3 (IC50 mM = 0.012 (CDK2), 0.67 (CDK7)) was synthesised and found to be is a potent CDK2 inhibitor, but with some CDK7-inhibitory activity (IC50 mM = 0.012 (CDK2), 0.67 (CDK7)). Further elaboration of the side-chain function has enabled the development of structure-activity relationships (SARs), and the identification of purines (e.g. 4, IC50 mM = 2.6 (CDK2), 0.56 (CDK7)) exhibiting some selectivity for CDK7, albeit with a loss of potency. 4 Subsequent SAR studies conducted on 2 have enabled the following observations to be made: firstly, the purine 6-cyclohexylmethoxy substituent is necessary for activity, with the corresponding 6-unsubstituted purine (5, IC50 mM = 46.9 (CDK2), 20.8 (CDK7)) exhibiting a 100-fold loss of potency against both CDK2 and CDK7. A terminal basic group (e.g. piperazinyl in 2) is required for activity, as replacement by a cyclohexyl substituent results in loss of activity against both kinases (6, 11% inhibition at 10 mM (CDK2), 13% inhibition at 100 mM (CDK7)). The sulfone linker is not a prerequisite for CDK7 activity, with the simple alkylpiperazine derivative (7, IC50 mM = 0.48 (CDK2), 0.51 (CDK7)) exhibiting comparable potency and selectivity. Finally, there appears to be some opportunity for expansion into the gatekeeper pocket of CDK7 by introducing small substituents at the purine C-8 position, with the potential for selectivity over CDK2 (8, 47% inhibition at 100 mM (CDK2), IC50 = 5 mM (CDK7)). Isolation and biological evaluation of the vinyl sulfone 9, an intermediate in the synthesis of 2, indicated a time-dependent inhibition of CDK2, suggesting that 9 is an irreversible inhibitor of CDK2. This would be the first reported irreversible inhibitor of a cyclin-dependent kinase, and therefore the activity of the compound against CDK2 was investigated using protein crystallography and site-directed mutagenesis 5 techniques. From these studies, encouraging evidence has emerged that 9 acts as an irreversible inhibitor of CDK2, covalently binding to a lysine residue within the ATP-binding pocket.

The defensins have been shown to be an important component of the innate immune system across many species from plants to man. To date a total of 14 beta defensins have been identified within the chicken genome and anti-microbial activity of an array of these peptides against pathogens has been demonstrated. The innate immune system has been shown to be important in chickens during the first week of life when exposure to pathogens in the environment occurs and the adaptive immune system is not fully developed. The research presented in this thesis attempts to investigate the effects of bird age (aged 0, 7, 14 and 35 days), genetics and rearing environment on components of the innate immune system. A farm trial was performed using three lines of Aviagen birds (lines X, Y and Z) and two distinct rearing environments (low and high hygiene farms). To determine which of the 14 avian beta defensins (AvBD) to investigate, a panel of potential single nucleotide polymorphisms (SNPs) within the AvBD locus was submitted as part of an Aviagen Ltd genomics initiative. Frequencies of the polymorphisms across the three lines of birds were determined. From these data, gene expression of AvBDs1, 4 and 10 were fully investigated using end-point PCR and then quantitative real-time PCR (qRT-PCR). A pronounced finding from the qRT-PCR was the marked intra and inter-group variation in gene expression levels, which lead to few statistical significant differences. All three genes were shown to be expressed across a panel of ten tissues analysed, but distinct patterns were also seen. Significantly AvBD1 gene expression in the duodenum indicated that of the 7 day old birds the line X birds reared in the low hygiene environment had the highest level of gene expression. In relation to AvBD4 gene expression, the highest level was observed in the spleen of the 0 day old birds, but overall environment did not appear to affect AvBD4 gene expression of the tissues examined. High levels of AvBD10 gene expression were observed in bird kidney and testicle tissue, but again environment in the case of the former tissue did not appear to statistically affect gene expression levels, the YH 7 day old bird testicle samples did have statistically significant higher expression levels compared to the other groups. The SNP analysis revealed three non-synonymous polymorphisms within the AvBD1 mature peptide locus. The three lines of birds had quite different patterns of these polymorphisms and so three different forms of the peptide along with a single form of the AvBD10 peptide were synthesised using a bacterial hyper-expression system. Peptide levels were quantified using an ELISA and subsequently tested in a bacterial time-kill assay using Salmonella enterica serovar Typhimurium phoP, Staphylococcus aureus and two strains of Enterococcus faecalis. All recombinant peptides showed anti-microbial activity against the bacteria tested. The exception was a clinical isolate of E. faecalis which showed resistance to the killing activities of recombinant AvBD10 peptide. Duodenal gut protein extracts were also tested using the same bacterial assay and marked differences in the anti-microbial activities of these samples were seen. The samples taken from the day 0 birds were found to have significant anti-microbial activity compared to those of the older birds. LC/MS identified differences in the proteomes of the respective gut extracts. These data support that bird genetics, age and the environment have an effect on AvBD gene expression and gut anti-microbial activity. These differences are not uniform for all genes and groups of birds, but clear patterns were observed.

VDAC (voltage-dependent anion channel) is a eukaryotic pore-forming β-barrel protein. It is found in abundance in the outer membrane of mitochondria, and is involved in mitochondrial homeostasis and function. A plethora of evidence also suggests the involvement of VDAC in programmed cell death, often referred to as apoptosis. However, despite its importance, the exact molecular structure of VDAC remains unsolved, with only 2D crystals of low resolution being available. Published topology models based upon experimental data suggest a β-barrel conformation comprising an N-terminal a-helix and between 12 and 19 membrane spanning p-sheets. This thesis describes the study of the topology of VDAC 1 from Saccharomyces cerevisiae with an aim to resolve the many contradictions between published topology models.

This project was initiated in order to investigate the basic electrochemistry of Pb/PbSO 4 in sulphuric acid. Very little has appeared in the literature on this subjectt workers have been concerned with the PbSO4 /PbO 2 region, as can be seen from the literature survey in Chapter 5. However, in the last few years interest has increased in the lead acid battery and I hope this thesis will help further technical advance. The first four chapters review basic background material against which the measurements were made. The following chapters then describe the experimental work and finally some attempt is made to link the basic electrochemistry to the working of a battery.