Precise regulation of cell cycle events by the Cdk-control network is essential for cell proliferation and the perpetuation of life. The unidirectionality of cell cycle progression is governed by several critical irreversible transitions: the G1-to-S transition, the G2-to-M transition, and the M-to-G1 transition. Recent experimental and theoretical evidence has pulled into question the consensus view that irreversible protein degradation causes the irreversibility of those transitions. A new view has started to emerge, which explains the irreversibility of cell cycle transitions as a consequence of systems-level feedback rather than of proteolysis. This thesis applies mathematical modelling approaches to test this proposal for the Mto- G1 transition, which consists of two consecutive irreversible substeps: the metaphase-to-anaphase transition, and mitotic exit. The main objectives of the present work were: (i) to develop deterministic models to identify the essential molecular feedback loops and to examine their roles in the irreversibility of the M-to-G1 transition; (ii) to present a straightforward and reliable workflow to translate deterministic models of reaction networks into stochastic models; (iii) to explore the effects of noise on the cell cycle transitions using stochastic models, and to compare the deterministic and the stochastic approaches. In the first part of this thesis, I constructed a simplified deterministic model of the metaphase-to-anaphase transition, which is mainly regulated by the spindle assembly checkpoint (the SAC). Based on the essential feedback loops causing the bistability of the transition, this deterministic model provides explanations for three open questions regarding the SAC: Why is the SAC not reactivated when the kinetochore tension decreases to zero at anaphase onset? How can a single unattached kinetochore keep the SAC active? How is the synchronized and abrupt destruction of cohesin triggered? This deterministic model was then translated into a stochastic model of the SAC by treating the kinetochore microtubule attachment at prometaphase as a noisy process. The stochastic model was analyzed and simulation results were compared to the experimental data, with the aim of explaining the mitotic timing regulation by the SAC. Our model works remarkably well in qualitatively explaining experimental key findings and also makes testable predictions for different cell lines with very different number of chromosomes. The noise generated from the chemical interactions was found to only perturb the transit timing of the mitotic events, but not their ultimate outcomes: all cells eventually undergo anaphase, however, the time required to satisfy the SAC differs between cells due to stochastic effects. In the second part of the thesis, stochastic models of mitotic exit were created for two model organisms, budding yeast and mammalian cells. I analyzed the role of noise in mitotic exit at both the single-cell and the population level. Stochastic time series simulations of the models are able to explain the phenomenon of reversible mitotic exit, which is observed under specific experimental conditions in both model organisms. In spite of the fact that the detailed molecular networks of mitotic exit are very different in budding yeast and mammalian cells, their dynamic properties are similar. Importantly, bistability of the transitions is successfully captured also in the stochastic models. This work strongly supports the hypothesis that uni-directional cell cycle progression is a consequence of systems-level feedback in the cell cycle control system. Systems-level feedback creates alternative steady states, which allows cells to accomplish irreversible transitions, such as the M-to-G1 transition studied here. We demonstrate that stochastic models can serve as powerful tools to capture and study the heterogeneity of dynamical features among individual cells. In this way, stochastic simulations not only complement the deterministic approach, but also help to obtain a better understanding of mechanistic aspects. We argue that the effects of noise and the potential needs for stochastic simulations should not be overlooked in studying dynamic features of biological systems.

As more patients with large body surface area burns are surviving and requiring reconstructive surgery, there is a necessity for advances in the provision of bioengineered alternatives to autologous skin cover. The aims of this Thesis are to identify feasible source tissues of Endothelial Colony Forming Cells and Mesenchymal Stem/Stromal Cells for microvascular network formation in vitro with three-dimensional dermal substitute scaffolds. The working hypothesis is that pre-vascularised dermal scaffolds will result in better quality scarring when used with split thickness skin grafts. Human umbilical cord blood, peripheral blood and adipose tissue were collected and processed with ethical approval and informed consent. Samples were cultured to form endothelial outgrowth colonies and confluent Mesenchymal Stem/Stromal Cells, which were characterised using flow cytometry and expanded in vitro. Mesenchymal Stem/Stromal Cell multipotency was confirmed with tri-lineage mesenchymal differentiation. Primary cells were tested in a two-dimensional tubule formation co-culture assay and differences assessed using a proangiogenic antibody array. Tubule formation was tested in four different acellular dermal substitute scaffolds; Integra® Dermal Regeneration Template, Matriderm®, Neuskin-F® and De-cellularised Human Cadaveric Dermis. Umbilical cord blood was the most reliable source of Endothelial Colony Forming Cells, the yield of which could be predicted from placental weight. Microvasculature dissected free from adipose tissue was a reliable source of Mesenchymal Stem/Stromal Cells which supported significantly more tubule formation than Mesenchymal Stem/Stromal Cells from whole adipose tissue. Microvasculature Mesenchymal Stem/Stromal Cells secreted significantly higher levels of the proangiogenic hormone leptin, and addition of exogenous leptin to the tubule formation assay resulted in significantly increased tubule formation. Microvasculature was cultured in all four of the scaffolds tested, but depth of penetration was limited to 100µm. The artificial oxygen carrier perfluorocarbon was shown to increase two-dimensional tubule formation and may be useful in further three-dimensional scaffolds studies to improve microvascular penetration.

2-0xoglutarate/Fe(II)-dependent dioxygenases (ZOG Oxygenases) are a relatively poorly characterised enzyme family that hydroxylate biological macromolecules to regulate a variety of essential cellular processes in mammals, including; chromatin remodeling, extra-cellular matrix formation and oxygen sensing. The work in this th esis focuses on a ZOG Oxygenase termed Myc-Induced Nuclear Antigen (MINAS3). This enzyme has been implicated in ribosome biogenesis and cell proliferation, and observed overexpressed in several tumour types, yet the identity afits substrate(s) and their role in cancer is unknown. The aims of the resea rch that has resulted in this thesis were to; (i) develop a cell model of MINAS3 enzyme activity, (ii) apply this model to study the role of MINAS3 activity in cell transformation and cancer, and (iii) discover novel cellular processes regulated by MINA53 activity. As such, I have created an isogenic cell model consisting of K-Ras-transformed MINAS3 knockout mouse embryonic fibroblasts (MEFs) reconstituted with either wildtype or enzyme-inactive MINAS3. Using this model I have shown that MINAS3 activity maintains normal levels of the large ribosomal subunit (60S), and suppresses anchorage-independent growth, autophagy and gene expression. These observations suggest the existence and involvement of one or more substrates. Indeed, proteomic and biochemical analyses in collaboration with the Schofield laboratory (Chemistry, Oxford) confirmed the identity of a MINA53 substrate, the 60S ribosomal protein Rp127a. Together we have shown that Rpl27a is abundantly hydroxylated, and that MINA53 is a histidinyJ hydroxylase; this represents the first discovery of a ribosomal oxygenase. The model developed here did not support a positive role for MINA53 in the transformation of MEFs. Rather it suggested that MINA53 can suppress transformation in some contexts, This prompted a wider investigation that demonstrated underexpression of MINA53 in several tumour types, and the presence of inactivating mutations in breast. ovarian and colon cancer. This thesis provides data supporting further research to understand the role of Rpl27a hydroxylation in the regulation of 60S biogenesis, autophagy and cancer. 2

Cytotoxic T lymphocytes (CTL) kill virally infected and tumourigenic cells via the regulated secretion of specialised secretory lysosomes. These secretory lysosomes contain cytolytic effector molecules, such as perforin and granzymes, which are able to induce apoptosis in target cells. Secretion occurs at the contact point between the CTL and its target, in a highly structured region termed the immunological synapse (IS). Upon formation of the IS, CTL undergo polarisation of their microtubule cytoskeleton and movement of the microtubule organising centre (MTOC) to the IS. Secretory lysosomes are then able to polarise along microtubules, fuse with the plasma membrane and deliver their effector molecules to the IS. The Adaptor Protein 3 complex (AP-3) sorts transmembrane proteins to lysosomes and deficiency in AP-3 results in missorting of proteins from the lysosomal to plasma membrane. CTL from AP-3 deficient patients, who suffer from Hermansky-Pudlak Syndrome Type 2 (HPS2), show reduced killing of target cells. This thesis describes two new patients with HPS2, both with homozygous mutations in the AP3B1 gene, which codes for the β3A subunit of the AP-3 complex. CTL from the new HPS2 patients show reduced cytotoxicity, which is shown here to be due to impaired secretory lysosome polarisation towards the IS. This impairment is common to HPS2 CTL, but varies between patients. In order to determine differences between HPS2 and wild type CTL, the localisation of a range of lysosomal, cytolytic, transmembrane, inhibitory and activation marker proteins is examined. This shows that in HPS2 CTL, LAMP1, CD63 and CD9 are potential AP-3 cargos. In addition, a possible effect on the key lytic effector perforin is identified. Preliminary experiments to allow proteomic comparison of HPS2 and wild type CTL are also presented. Further investigation of these results will help to shed light on the mechanisms involved in secretory lysosome polarisation in CTL.

DNA double-strand breaks (DSBs) are highly genotoxic lesions, which can promote chromosomal rearrangements and tumorigenesis through oncogene activation or loss of heterozygosity (LOH) at tumour suppressor loci. To identify new genes involved in DSB repair and genome stability, an S. pombe deletion library was screened for mutants which exhibited sensitivity to the DNA damaging agents bleomycin and/or MMS. 192 mutants were isolated which exhibited increased sensitivity to one or both of these agents. These mutants were further analysed in a sectoring assay and mutants sought which exhibited elevated levels of break-induced loss and rearrangement of a non-essential minichromosome. Using this approach 57 genes were identified, including all known homologous recombination (HR) and DNA damage checkpoint genes present in the library. Further, quantitative analysis of DSB repair indicated that 25 of these genes functioned to promote efficient HR repair, thus representing a comprehensive HR gene set in fission yeast. Included in this gene set are 10 genes not previously implicated in HR repair; nse5⁺, nse6⁺, ddb1⁺, cdt2⁺, alm1⁺, snz1⁺, kin1⁺, pal1⁺, SPAC31G5.18c⁺ and SPCC613.03⁺. Detailed characterisation of ddb1Δ and cdt2Δ established a role for the Ddb1-Cul4Cdt2 ubiquitin ligase complex in HR. The findings presented here support a model in which break-induced Rad3 and Ddb1-Cul4Cdt2 ubiquitin ligase-dependent Spd1 degradation promotes ribonucleotide reductase activation and nucleotide biosynthesis, which is required for post-synaptic ssDNA gap filling during HR repair. Lastly, the role of HR genes in suppressing chromosome loss and rearrangements was examined. A striking inverse correlation between levels of gene conversion and levels of both chromosome loss and LOH was observed across the HR gene deletion set. These findings support a common and likely evolutionarily conserved role for HR genes in suppressing both chromosome loss and break-induced chromosomal rearrangements resulting from extensive end processing associated with failed HR repair.

Most genome-wide association studies search for genetic variants associated to a single trait of interest, despite the main interest usually being the understanding of a complex genotype-phenotype network. Furthermore, many studies collect data on multiple phenotypes, each measuring a different aspect of the biological system under consideration, therefore it can often make sense to jointly analyze the phenotypes. However this is rarely the case and there is a lack of well developed methods for multiple phenotype analysis. Here we propose novel approaches for genome-wide association analysis, which scan the genome one SNP at a time for association with multivariate traits. The first half of this thesis focuses on an analytic model averaging approach which bi-partitions traits into associated and unassociated, fits all such models and measures evidence of association using a Bayes factor. The discrete nature of the model allows very fine control of prior beliefs about which sets of traits are more likely to be jointly associated. Using simulated data we show that this method can have much greater power than simpler approaches that do not explicitly model residual correlation between traits. On real data of six hematological parameters in 3 population cohorts (KORA, UKNBS and TwinsUK) from the HaemGen consortium, this model allows us to uncover an association at the RCL locus that was not identified in the original analysis but has been validated in a much larger study. In the second half of the thesis we propose and explore the properties of models that use priors encouraging sparse solutions, in the sense that genetic effects of phenotypes are shrunk towards zero when there is little evidence of association. To do this we explore and use spike and slab (SAS) priors. All methods combine both hypothesis testing, via calculation of a Bayes factor, and model selection, which occurs implicitly via the sparsity priors. We have successfully implemented a Variational Bayesian approach to fit this model, which provides a tractable approximation to the posterior distribution, and allows us to approximate the very high-dimensional integral required for the Bayes factor calculation. This approach has a number of desirable properties. It can handle missing phenotype data, which is a real feature of most studies. It allows for both correlation due to relatedness between subjects or population structure and residual phenotype correlation. It can be viewed as a sparse Bayesian multivariate generalization of the mixed model approaches that have become popular recently in the GWAS literature. In addition, the method is computationally fast and can be applied to millions of SNPs for a large number of phenotypes. Furthermore we apply our method to 15 glycans from 3 isolated population cohorts (ORCADES, KORCULA and VIS), where we uncover association at a known locus, not identified in the original study but discovered later in a larger one. We conclude by discussing future directions.

This thesis concerns the history of humanitarian nutrition and its political implications. Drawing on aid agency archives and other historical sources, it examines how food has been delivered in emergencies, from the First World War to the present day. The approach is ethnographic: this is a study of the micro-level practices of relief, examining the objects distributed, the plans made, the techniques used. It is also historical: examining how such practices have changed over time. This thesis makes five interlocking arguments. First, I make a political point: that humanitarian action is always political, and that it is impossible to adhere to ‘classical’ humanitarian principles such as neutrality, impartiality and independence. Second, I make a sociological argument: that the activities of humanitarian nutrition have been shaped by a number of themes, which include militarism, medicine, modernity, and markets. Third, I make a historical argument: that the main features of humanitarian nutrition were solidified between the 1930s and the 1970s, and were largely in place by the time of the Biafran war. Fourth, I make a sociological argument: that these mid-century changes involved a profound redefinition of hunger and food (with hunger conceived as a biochemical deficiency, and food as a collection of nutrients). Finally, I make a normative argument, suggesting that this redefinition has not necessarily benefited the starving: the provision of food in emergencies, I argue, is often concerned with control and efficiency rather than the suffering individuals themselves.

This dissertation concerns the study and prediction of loops in protein structures. Proteins perform crucial functions in living organisms. Despite their importance, we are currently unable to predict their three dimensional structure accurately. Loops are segments that connect regular secondary structures of proteins. They tend to be located on the surface of proteins and often interact with other biological agents. As loops are generally subject to more frequent mutations than the rest of the protein, their sequences and structural conformations can vary significantly even within the same protein family. Although homology modelling is the most accurate computational method for protein structure prediction, difficulties still arise in predicting protein loops. Protein loop structure prediction is therefore a bottleneck in solving the protein structure prediction problem. Reflecting on the success of homology modelling, I implement an improved version of a database search method, FREAD. I show how sequence similarity as quantified by environment specific substitution scores can be used to significantly improve loop prediction. FREAD performs appreciably better for an identifiable subset of loops (two thirds of shorter loops and half of the longer loops tested) than ab initio methods; FREAD's predictive ability is length independent. In general, it produces results within 2Å root mean square deviation (RMSD) from the native conformations, compared to an average of over 10Å for loop length 20 for any of the other tested ab initio methods. I then examine FREAD’s predictive ability on a specific type of loops called complementarity determining regions (CDRs) in antibodies. CDRs consist of six hypervariable loops and form the majority of the antigen binding site. I examine CDR loop structure prediction as a general case of loop structure prediction problem. FREAD achieves accuracy similar to specific CDR predictors. However, it fails to accurately predict CDR-H3, which is known to be the most challenging CDR. Various FREAD versions including FREAD with contact information (ConFREAD) are examined. The FREAD variants improve predictions for CDR-H3 on homology models and docked structures. Lastly, I focus on the local properties of protein loops and demonstrate that the protein loop structure prediction problem is a local protein folding problem. The end-to-end distance of loops (loop span) follows a distinctive frequency distribution, regardless of secondary structure elements connected or the number of residues in the loop. I show that the loop span distribution follows a Maxwell-Boltzmann distribution. Based on my research, I propose future directions in protein loop structure prediction including estimating experimentally undetermined local structures using FREAD, multiple loop structure prediction using contact information and a novel ab initio method which makes use of loop stretch.

Interleukin (IL)-1beta is a powerful pro-inflammatory cytokine with important roles in directing both innate and adaptive immunity. As a result, its production is tightly controlled, with the synthesis of an inactive form (pro-IL-1beta) and the requirement of a second signal. This induces the formation of the inflammasome, a macromolecular complex which mediates the maturation of IL-1beta into the bioactive cytokine. Given its significance, it is important to identify mechanisms of IL-1beta induction and modulation. Firstly, we describe serum amyloid A (SAA), an acute phase protein with immunomodulatory properties, as a novel inducer of IL-1beta. Using cells from genetically modified mice, the molecular mechanisms responsible were dissected, demonstrating the receptors TLR2 and NLRP3 as required for this effect. By instilling SAA into mice, we also show that SAA is able to induce IL-1beta production in vivo. Invariant natural killer T (iNKT) cells have also been shown to be important modulators of immunity, mediating both pro- and anti-inflammatory responses. iNKT cells are non-conventional T lymphocytes which recognise glycolipid in the context of CD1d, with the ability to interact with immature antigen presenting cells in an autoreactive manner. We link the regulatory ability of iNKT cells with IL-1beta production, showing that a low activation signal leads to the induction of an IL-13-dominated cytokine profile, as well as weak engagement of the CD40-CD40L pathway. We show for the first time that through these mechanisms, iNKT cells are able to dampen the secretion of IL-1beta upon subsequent stimulation of dendritic cells. We hypothesise that this effect of iNKT cells is important in controlling inflammatory responses in vivo, and demonstrate exacerbated IL-1beta production and inflammation during influenza virus infection of iNKT cell-deficient animals. This novel anti-inflammatory property of iNKT cells may be harnessed in the therapeutic intervention of inflammatory disorders.

Macrophages are one of the three main human cell types infected by HIV-1. They are highly plastic cells requiring a dynamic actin cytoskeleton for their role in development, homeostasis, tissue repair and immunity. For HIV-1, disrupting actin in macrophages is detrimental in that it leads to a complete block of viral uptake and reduces reverse transcription but, significantly, not fusion. Rho GTPases (Rac1, RhoA and Cdc42) regulate many aspects of actin dynamics including those required for endocytosis. Using a pharmacological approach, it was shown that Rac1 along with Rho GTPase effectors Pak1 and N-WASP are important for productive HIV-1 entry in macrophages. However, pharmacological inhibitors aren’t available for many host factors and may have off-target effects. To overcome this, expression of dominant negative (DN) Rho GTPases was attempted in human stem cell-derived macrophages (esMDMs). While DN Rac1 expressing esMDMs were successfully generated, this was not possible for the other two. DN Rac1 expressing esMDMs, as expected, had less filamentous actin and reduced dextran uptake compared to control esMDMs. In contrast to the pharmacological studies, HIV-1 infection studies in Rac1 DN esMDMs revealed a significant increase in HIV-1 fusion, reverse transcription and nuclear import, which could be due to reduced filamentous actin leading to a slower rate of endocytosis thereby allowing more time for viral fusion within endocytic vesicles. Surprisingly, reduced HIV-1 gene expression was observed in Rac1 DN esMDMs. This was corroborated by transfection studies implicating Rho GTPases in LTR driven gene expression. To overcome the ineffectiveness of RhoA and Cdc42 DN constitutive gene expression in esMDMs, an inducible lentiviral gene expression system based on the use of a constitutive promoter and a FLEx switch mediating irreversible DNA inversions was generated. The novel FLEx vector was the first system shown to induce transgene expression in esMDMs albeit at a very low efficiency.