Search

Cytokinesis is the final stage of cell division when the two daughter cells are physically separated. Late cytokinetic cells are connected by an array of antiparallel microtubules, named midbody. Midbody resolution is mediated by the endosomal sorting complex required for transport (ESCRT) machinery during the final stage of cytokinesis, known as abscission. Abscission is tightly regulated by the Aurora Bmediated abscission checkpoint (NoCut) and, thus, is delayed in response to persisting chromatin bridges, defective nuclear pore complex assembly, replication stress or midbody tension. The ESCRT-III subunit CHMP4C is a major negative regulator of abscission acting downstream of Aurora B. CHMP4C deficiency results in premature resolution of DNA bridges, genetic damage and chromosomal instability. Notably, polymorphisms in CHMP4C gene have been associated with susceptibility to various cancers, including breast and ovarian cancer. In this study we show by live-cell imaging that CHMP4C-depleted cells connected with a DNA bridge display a distinctive mirrored nuclear distribution of 53BP1 DNA damage foci. CHMP4C silencing also leads to death of daughter cell pairs formed after a division with a bridge. Notably, we identify synthetic lethality between CHMP4C depletion and silencing of key breast- and ovarian-cancer associated homologous recombination (HR) genes such as BRCA1 and BRCA2 by an RNAi screen. Therefore, we hypothesise that post-mitotic DNA damage generated as a result of CHMP4C deficiency in the presence of a DNA bridge necessitates repair by the HR machinery, specifically. Intriguingly, we show that BRCA1 is required for abscission delay in response to DNA bridges induced by Topo II inhibition, but not in response to depletion of nucleoporins NUP50 and NUP153. Therefore, our findings reveal important interactions between HR pathway components and the abscission regulator CHMP4C, which could shed a light on potential clinical relevance of CHMP4C polymorphisms in BRCA mutant tumours. For instance, synergistic activity between cancer-associated polymorphisms in CHMP4C and pathogenic mutations in BRCA1 or BRCA2 could further increase susceptibility to breast and ovarian cancer, whereas the hypersensitivity of CHMP4C-deficient tumour cells to DDR pathway disruption might be targeted as part of an anti-cancer treatment.

Development of new systems that mimic the native cellular environment has emerged as one of the main strategies for tissue engineering and future biomedical applications. While nanoparticles (NPs) and nanotubes (NTs) positioned themselves as candidates of high potential in the field, more and more studies emphasize their size-related cytotoxicity, the difficulty of translating them from research laboratories into the clinic, and in conducting large-scale synthesis. Peptoids are peptide mimetics. The only difference from natural peptides is that the side chain is shifted from the backbone alpha carbon on the peptide to the nitrogen atom on the peptoid. This structural difference confers on peptoids several interesting properties such as biostability, and easy and economical synthesis. While the side chain shift deprives the peptoids from chiral centres and backbone secondary structure hydrogen bonding, the incorporation of specific side chains enables the folding of the peptoid chains into organised secondary structures such as nanosheets. This thesis presents the study of two different peptoid systems and their interaction with cells, namely peptoid nanosheets and a peptoid hydrogel. The aim was to create peptoid materials inspired by the mechanisms of extracellular matrix (ECM)-cell interaction to control stem cell growth and differentiation. In the first and major part of the thesis, we focused on the mechanical and biochemical properties. We wanted to mimic the mechanical interaction of the ECM that is displaying biologically relevant peptide ligands with cells. This was carried out by developing peptoid nanosheets (PNS) as a stiff and functionalizable platform and characterizing their effect on cells. Those PNS combine the following advantages: a structure close to the bilayer cell membrane, a peptidic nature and cell size similarity, stiff intrinsic mechanical strength, biocompatibility and the possibility of surface functionalisation with different types of ligands with high degree of control. In the second part, we wanted to capture the dynamic properties of the 3D environment provided by the ECM by beginning to develop a peptoid hydrogel that is capable of changing its mechanical stiffness upon exposure to a stimulus. This thesis describes not only the effect of the peptoid systems on MSCs but also the strategy and steps taken to develop a cell culture system capable of sustaining the integrity of both cells and PNS, as this is the first time the effect of peptoid systems on cells was studied. To summarise, peptoids are biostable, biocompatible and easy to synthesize, flexible to functionalise for developing biomimetics in the nano/micro scale. They have the potential to harness its advantages and at the same time evade the drawbacks of conventional nanomaterials used with biological systems.

Oxygenic photosynthesis plays a pivotal role at the very heart of global ecosystems, providing the food, fuel and oxygen that sustains virtually all life on Earth (Hohmann-Marriott and Blankenship, 2011). Despite the essential and highly intricate role that photosynthesis plays on Earth, the actual process of photosynthesis appears to be remarkably inefficient with only ~ 9-12% of useable solar energy (wavelengths between 400 – 700 nm) being converted to biomass (Zhu et al., 2010). Given the rising demands on the global food chain associated with climate change and a rising population, it is widely recognised that improvements in the efficiency of this vital process will be required to ensure food security for an ever-increasing population over the coming decades (Long et al., 2015; Zhu et al., 2010). Among the multiple targets which have been identified for potential improvement is cytochrome b6f (cytb6f), an integral membrane complex found at the heart of oxygenic photosynthesis. As well as facilitating the rate-limiting step in light dependent electron and proton transfer, the cytb6f complex also plays a key role as a redox-sensing hub in higher plants involved in the regulation of light-harvesting, electron transfer, photosynthetic gene expression and adaptation to environmental stress. Together, these characteristics make cytb6f a judicious target for genetic manipulation to enhance photosynthetic yield and promote stress tolerance in crop plants. While a number of studies show great promise in this regard (Simkin et al., 2017), further progress is hindered by the lack of a detailed understanding of the structure and function of the cytb6f complex from higher plants. Here we present the first structures of the cytb6f complex from a higher plant (Spinacia oleracea), shedding light on the internal mechanics of the Q-cycle and providing new clues as to how this extraordinary complex fulfils its various roles in photosynthetic regulation. We offer further opportunities to explore these insights by presenting a second structure, from the model cyanobacterium Synechocystis sp. 6803. Together these structures provide a number of key mechanistic insights into the cytb6f complex, many of which may now be further explored through structure-based mutagenesis of the Synechocystis complex and molecular dynamics simulations.

Cyclic secondary amines can be used to perform enamine and iminium ion catalysis to modify carbonyl compounds through a broad range of chemical reactions. To utilise these compounds in a chemical biology context requires the catalysis to work in an aqueous and biocompatible environment. To accomplish this, the catalysts can be placed into proteins generating new hybrid organocatalytic artificial enzymes. This thesis concerns two strategies to generate these artificial enzymes and the subsequent characterisation of their activity. Firstly, penicillin binding proteins are covalently modified with carbapenems that carry secondary amine pyrrolidines in their structure. Both the carbapenems alone and the protein-carbapenem hybrids were able to afford a Michael-nitro addition reaction under biocompatible reaction conditions. The second strategy concerns the direct incorporation of secondary amine unnatural amino acids into a protein backbone using genetic code expansion technology. Three unnatural amino acids are incorporated at four different positions in the multidrug binding protein LmrR. The resulting proteins are tested for their activity to perform selective reduction of α,β- unsaturated aldehydes using benzyl-nicotinamide as a hydride source. The catalytic amino acid generating the highest conversion is then incorporated into the natural oxidoreductase dihydrofolate reductase from E. coli at three different positions. The EcDHFR variants are then tested for their activity in the hydride transfer reaction, using NADPH as the hydride source. Two highly active enzymes LmrR Phe93DPK and EcDHFR Ala7DPK are identified and further investigated for their enzyme kinetic character and substrate scope. EcDHFR Ala7DPK is then coupled to a natural enzyme system allowing for the co-factor to be recycled in situ. Mechanistic studies and kinetic isotope effects are performed that strongly suggest an iminium ion mechanism in the hydride transfer reaction. Genetic code expansion relies heavily on orthogonal tRNA synthetases. In this work, pyrrolysine tRNA synthetase form M. bakeri is used as the pivotal in vivo enzyme to incorporate the cyclic secondary amine unnatural amino acids. Although the catalytic C-terminus has been extensively explored, the tRNA binding N-terminus has been less so. Mutations in the N-terminus were performed to determine if they can increase unnatural amino acid incorporation efficiency, however the opposite was found, in contrast to its homolog from M. mazei. Five MbPylRS variants, eight unnatural amino acids and two different culture temperatures are explored.

Asparaginyl endopeptidases (AEP) are cysteine proteases found in mammalian and plant cells. Several AEP isoforms from plant species were found to exhibit transpeptidase activity, which is integral for the key head-to-tail cyclisation reaction during the biosynthesis of cyclotides. AEPs isolated from plants exhibit excellent enzyme kinetics for peptide ligation via a relatively short substrate recognition sequence. Hence, AEPs could offer powerful applications as a biocatalyst in the production of modified proteins and peptides. However, the uptake of AEPs for biocatalytic application is limited by key challenges. In this thesis, an engineered AEP variant from Oldenlandia affinis, OaAEP1-C247A, is employed as a model enzyme to explore strategies which can address some vital limitations to the application of AEP catalysis. Fundamental enzymology of AEPs are discussed and a new protocol which simplifies the recombinant preparation of OaAEP1-C247A is described. To address the challenges posed by the reversible reactivity of the enzyme, a chemo-enzymatic approach to AEP-mediated peptide and protein bioconjugation is reported. Employing an affordable chemical reagent as a scavenger, this method takes advantage of the non-specific substrate recognition of OaAEP1-C247A. Accumulation of active proteases in the cytoplasm can be detrimental to the viability of the host cell. Therefore, strategies to encapsulate OaAEP1-C247A within a self-assembled protein capsid (AaLS-13) were explored. The caged protein complex serves as an artificial organelle for protein and peptide processing which mimics organelles and protein microcompartments found in nature. This synthetic biology approach offers the potential to enable AEP-catalysis in E. coli.

Precise and accessible techniques for measuring metabolic responses to environmental stress are essential to allow the likely impacts of climate and climate change on tick distribution, abundance and phenology to be predicted. Here, in Chapter 2, a series of biochemical protocols employing spectrophotometric methods are described and used to determine the entire energy budget of ticks. Protein, total lipid, neutral lipid, the total water-soluble carbohydrates and glycogen were measured in individual Ixodes ricinus nymphs and adults. Two key trends were identified: in adults, protein was relatively more abundant than in nymphs, whereas in nymphs, glycogen and soluble carbohydrates were more abundant than in adults, with glycogen alone composing 39% of the mass of metabolites in nymphs compared to 15 and 10% in females and males, respectively. The results demonstrate that the spectrophotometric approaches deliver relatively rapid and reliable estimates of the total energetic budget and can be used to quantify the metabolic profiles of individual ticks. The work described in Chapter 3 aimed to investigate the effects of temperature on the rate of depletion of energy reserves by nymphal and adult I. ricinus. A cohort of nymphs, males and females were collected and divided into incubators at a range of temperatures. The protein, total lipid, neutral lipid, soluble carbohydrates and glycogen levels were measured over time. In nymphs, the rate of soluble carbohydrates and glycogen utilisation was higher than in males or females and the concentrations of neutral lipids were affected significantly by higher temperatures. In adults, the concentrations of protein and structural lipid (phospholipid) responded rapidly to changes in treatment time and the ambient temperature. Nymphs and adult I. ricinus were sampled from the field each month from February 2018 to January 2019 and, in Chapter 4, the changes in energy source contents over the year are presented. The data suggest that there exists a well-defined cohort of relatively well-fed nymphal ticks in the early spring, most probably derived from larvae that fed the previous summer and moulted the previous autumn. They start to quest as temperatures rise sufficiently in spring to permit activity. Those that are unable to find a blood-meal continue questing but gradually exhaust their reserves. By mid-summer the only larvae left questing are close to starvation. For females, the population at the start of the year is composed of a cohort that started questing early in the year that were relatively hungry; they may have been derived from nymphs that fed relatively early the previous year moulted to become adults that were unable to feed, so overwintered and started questing early the following year in an already resource depleted state. By April, the cohort of questing females has now been joined by individuals that moulted to become adults late the previous year or early in the year and are in relatively well resourced. However, those that do not feed gradually start to exhaust their resources. In the second half of the year, females with high levels of metabolic resource appear - presumably from nymphs that early in the year of the study, digested their blood meals, moulted to become adults and are now re-joining the questing population. In males the patterns of changing energy sources were much less distinct than seen in nymphs or females. In the final Chapter, the data are discussed generally, and further applications of the spectrophotometric techniques presented here are considered.

Protein self-assembly offers key strategies for nature to generate high-performance and multifunctional proteinaceous materials. Inspired by nature's approach, there is growing interest in the exploitation of self-assembled protein structures as a basis of artificial protein materials. To this end, this thesis aims to gain a better understanding of the nanofibrillar self-assembly of natural proteins and to develop new methods to utilise self-assembled structures for novel protein materials. The approach demonstrated here enabled an enhanced level of control over multi-scale protein assembly, leading to the creation of functional protein hydrogels, fibres and films, making them suitable for a wide range of applications. The first part of the thesis focuses on investigating the molecular self-assembly mechanism of silk fibroin, a natural protein used by silkworms to generate fibres. Through applying kinetic analysis, I demonstrated that the self-assembly process is dominated by a secondary process in which the formation of new fibrils is catalysed by the existing aggregates in an autocatalytic manner. In addition, I showed that hydrodynamic shear applied to protein molecules accelerates self-assembly by promoting primary nucleation. The results provided useful insights into nature's strategies for controlling the assembly of proteins into materials. The second part describes the fabrication of silk-inspired protein-based microfibres using nanofibrils obtained from a milk-derived protein, β-lactoglobulin. The fibres were generated using a miniaturized fluidic device, which enables precise control of the hydrodynamic forces applied to the nanofibrils and thus replicates the natural spinning process of silk. By varying the flow rates, the degree of nanofibril alignment was tuned, leading to an orientation index comparable to that of native silk. It was further confirmed that this increased level of alignment led to enhanced mechanical properties in the fibre. The last part demonstrates a scalable method to induce and control the self-assembly of plant-derived proteins. Plant-derived proteins are attractive protein feedstocks as they can be sourced in a sustainable and low environmental impact manner. However, their poor water solubility poses fundamental challenges for controlling their self-assembly. The method developed in this project enables the dissolution of plant-derived proteins in aqueous solution with concentrations as high as 10w/v%, which can then undergo self-assembly into intermolecular β-sheet rich networks. The approach was further exploited to fabricate nanostructured hydrogels and films with advantageous mechanical and optical properties.

Coordination capsules offer a simple synthetic route to complex structures. These systems can be used to encapsulate a variety of different species, giving them potential uses in biology, as single molecule magnets, in gas storage as well as in catalysis. Almost all current catalytic approaches utilise binding the substrate in the capsule cavity in an analogous fashion to biology's catalysts, enzymes, which activate substrates upon binding in an active site. Substrate encapsulation allows the cavity's microenvironment to influence various reaction factors such as enantio- and regioselectivity. The drawbacks of this approach include narrow substrate scope and the frequent occurrence of product inhibition, which limits genuine catalysis to quite specific reaction types. In biology, some enzymes require the binding of a second species, a cofactor, to generate the active catalyst. Rather than the encapsulated guest acting as the substrate, the bound species can instead be treated as a cofactor. Varying the functionality of the guests generates a modular approach to accessing a multitude of capsule-guest systems with different, emergent catalytic properties. Chapter 2 outlines the initial attempts at using p-quinones as cofactor mimics, whereby alteration of the bound guest led to the transformation of the bound p-quinone into a hydroxylated species. The formation of this species, along with the stabilisation of its conjugate base, means that each host-guest system generates two equivalents of protons, which are liberated into the bulk for catalysis. Chapter 3 covers using the Pd2L4 capsules to stabilise the conjugate bases of weak acids to promote Brønsted acid catalysed reactions in the bulk, primarily the cyclisation of citronellal. The hydrogen bond donor ability of the guest is enhanced by encapsulation, while also moving the substrate into the cage's microenvironment. Chapter 4 focusses on increasing the redox properties of electron deficient p-quinones upon formation of host-guest complexes with Pd2L4 capsules to promote electron transfer catalysis in the bulk. This is further built upon in Chapter 5, whereby p-quinones with extended conjugation, encapsulated in Pd2L4 capsules, generate complexes that can be activated by light to act as photoredox catalysts.

DELLA proteins are master growth regulators that repress responses to the phytohormones gibberellins (GAs). Manipulation of DELLA signalling was instrumental in the development of high‐yielding crop varieties that saved millions from starvation during the "Green Revolution". In order to infer how DELLA signalling has evolved within the land plant lineage, the functions of DELLA proteins in the bryophyte Physcomitrium (Physcomitrella) patens were examined in the present study. It was found that PpDELLAs are regulated differently from flowering DELLA proteins, as they are not degraded by diterpenes, they do not interact with GID1 homologues, they are not directly regulated by the 26S proteasome and they do not regulate abiotic stress responses. However, similarly to flowering plant DELLAs, they are involved in the regulation of reproductive development and they act as transcriptional 'hubs', possibly interacting with transcription factors belonging to similar families as flowering plant DELLA interactors. Using IP-MS and yeast two-hybrid, three photoreceptors were identified as PpDELLA protein-protein interactors, however no involvement of PpDELLAs in light responses was observed. In addition, while the Arabidopsis DELLA AtRGA1 did not interact with the seed dormancy regulator AtDOG1, PpDELLAs interacted with PpDOG1s, and the two proteins displayed similar functions in spore germination. This work shows that DELLA proteins in P. patens have both conserved and divergent functions. Unravelling how PpDELLAs function at the molecular level may provide novel ways of genetic modification that can be used to engineer better crops and contribute towards mitigating the effects of global warming and achieving global food security.

Domains are functional and evolutionary units of proteins that typically fold into stable globular structures. A small subset of natural multidomain proteins contain large arrays of nearly identical domains repeated in tandem, challenging some of our assumptions about protein folding and evolution. In this study, I aim to discover new tandem domain repeats, characterise their sequence and structural properties and understand their roles in the function of proteins. I start by using computational sequence analysis tools across large datasets of proteins and bacterial genomes to survey the prevalence and distribution of tan- dem domain repeats across organisms and domain families. Next, I computation- ally analyse and compare structures of domains found as tandem repeats, several of which have been experimentally determined by our collaborators in the course of this study. I finally develop two computational methods to systematically model the structure and misfolding energetics of tandem domain repeats. Nearly identical tandem domain repeats are rare in natural proteins (below 0.1%) and their sequences are highly biased in amino acid composition. Many of them have structural roles in bacterial surface proteins implicated in biofilm for- mation and host colonisation; new examples of such proteins, named "Periscope proteins", show rapid domain repeat number variation, a molecular mechanism used to modulate bacterial phenotype. Tandem domain repeat structures reveal unusual structural malleability, with numerous cases of domain atrophy (loss of core secondary structures) and elaboration. They are also predicted to be more resistant to misfolding via tandem domain swapping, with potential misfolding- resistant mechanisms such as the domain topology and length. This study improves our understanding of the prevalence, type and function of tandem domain repeats in proteins, in particular their role as structural ele- ments in bacterial surface proteins, and suggests new protein and domain targets for further experimental characterisation. It also has important implications for protein misfolding and for the design and engineering of multidomain proteins.

The majority of Nuclear Envelope Transmembrane Proteins (NETs) are tissue specific and many of these facilitate tissue-specific genome organization. Genome organization changes dramatically during differentiation and these NETs impact this process: muscle-specific genome-organizing NETs NET39, WFS1 and TMEM38A are important for myogenesis (Robson et al, 2016) while fat-specific genome-organizing NETs TMEM120A and B are important for adipogenesis (Batrakou et al, 2015). Although during lineage specification of mouse embryonic stem cells (Peric-Hupkes et al, 2010), we do not yet understand the temporal dynamics of these changes nor the components of the nuclear envelope that orchestrate these changes during early stages of exit from pluripotency. In this thesis, I investigate the temporal dynamics of genome organization changes during pluripotency exit stimulated by LIF withdrawal. Using Fluorescence in-situ Hybridization (FISH) to label DNA, I demonstrate that some of the earliest changes in genome organization occur within the first hour of exit from pluripotency with the relocation of a locus containing three genes Triml1, Triml2 and Zfp42 (that encodes REX1, a well-known marker of pluripotency) from the nuclear interior to the nuclear periphery. The RNA and protein levels of these genes persist for several hours post exit, suggesting that reorganisation of the genome is among the very first of events occurring during lineage specification and is perhaps a higher order mechanism controlling differentiation as a change in genome organisation could affect the transcriptional profile of these cells. To try and identify the proteins involved in tethering the locus and the mechanism of release I also investigated the changes in the nuclear envelope composition as cells undergo an exit from pluripotency. I show that while certain proteins undergo post translational modifications such as phosphorylation, other new proteins are synthesised during the first two hours of exit. Using phospho-null mutants for LBR and LAP2α, I show that these play a role in the relocation of this genomic locus. Finally, I introduced tissue-specific genome-organizing NETs such as NET39 (muscle), TAPBPL (blood) and TMEM120A(fat) into embryonic stem cells and found that their introduction causes a forced exit from pluripotency. Interestingly, these NETs show specificity in their ability to affect the position of genomic loci encoding pluripotency factors like Rex1 and Nanog, strengthening the idea that these tissue specific NETs act as tethers to very specific genomic regions in order to maintain a tissue specific genome organization. The results discussed here present for the first time, a temporal view of the changes in genome organisation during such early stages of in vitro differentiation. While Rex1 repositioning has been studied in greater detail in this thesis, a more comprehensive study over the early stages of exit might reveal additional genomic loci that reposition during this phase. The rapid reorganisation of the genome following LIF withdrawal highlights the importance of tightly controlling and maintaining appropriate culture conditions for the study of pluripotency using embryonic stem cells as a model system. The study leads to conceptual advancement in stem cell biology by describing early events following exit from pluripotency and in the field of nuclear biology by identifying the NE composition in ES cells. Collectively the results demonstrate the role of the nuclear envelope in the maintenance of pluripotency and in orchestrating genome organisation changes during exit.

Reactive oxygen species (ROS) play important roles for the regulation of normal functions such as proliferation, differentiation, migration and cell death. At low doses they participate in the redox balance, but an excess of these species leads to damage to proteins, lipids or DNA. ROS are involved in the onset and progression of several degenerative diseases (e.g., cancer, neurological disorder, etc). Cancer cells are highly susceptible to ROS-mediated damage and several chemotherapy agents achieve cytotoxicity by inducing oxidative stress. Sensing the variations of different intracellular ROS is crucial for real time assessments of anticancer treatment efficiency. Yet, no sensor currently allows simultaneous and independent monitoring of different ROS live cells. Indeed, existing sensors monitor either the total levels of ROS or the levels of single species (i.e., sensors such as diphenylanthracene, peroxy yellow, anthrafluorescein, etc.). The need to optimise and personalise treatment regimens and for unravelling the mechanisms underpinning ROS-induced cell death requires the introduction of a new set of tools able to provide a real-time report of intracellular ROS levels in response to a given intervention. In this project we developed new fluorescent nanosensors able to respond to different ROS. This was achieved through the synthesis of three conjugatable molecular probes able to respond to individual ROS, namely; an anthrafluorescein-based probe for superoxide anion, a dimethylanthracene-based probe for singlet oxygen and a fluorescein-based probe for hydrogen peroxide. The new probes were grafted onto poly(lactic-co-glycolic acid) (PLGA) and formulated as nanoparticles containing either conjugated or encapsulated sensors. We characterised the fluorescent response of the probes, conjugates and nanospecies in the presence of the target ROS analytes. Lastly, we demonstrated the ability of the nanosensors to enter cells and their potential of to be used as intracellular ROS sensors.

The lipid phosphatase PTEN is most well-known for its role in dephosphorylating PIP₃ to negatively regulate the Akt pathway. This pathway is important in tumour suppression, and has more recently been shown to be involved in synaptic plasticity. PTEN is mutated in various cancers and dysregulated in Alzheimer's Disease, suggesting it may be a useful therapeutic target. PTEN activity is controlled by a range of post-translational modifications (PTMs) including SUMOylation, which has been reported to regulate the localisation, function, stability and coupling to the Akt pathway. The initial aim of this study was to assess if, and how, PTEN SUMOylation affects its reported functions. To achieve this, I optimised protocols to detect SUMOylation and characterise a novel "PTEN-3KR" mutant, in which all three known SUMO acceptor lysine residues are substituted to arginine. Unexpectedly, PTEN-3KR displayed both enhanced SUMOylation and ubiquitination. While surprising, this provided an interesting tool to investigate the consequences of these increased levels of PTEN modification. I found that PTEN-3KR is unable to supress the Akt pathway, but its turnover and ability to dimerize were unaffected. I also investigated how PTEN-3KR impacted on AMPAR trafficking and retromer function. PTEN-WT, but not PTEN- 3KR, overexpression upregulated total protein levels of retromer components SNX27 and VPS26. Compared to PTEN-WT, PTEN-3KR also bound less to SNX27 and decreased surface expression of a prototypic retromer cargo GluT1. These results confirm the importance of SUMOylation and/or ubiquitination in the regulation of PTEN activity. Moreover, they demonstrate a novel role for SUMOylation of PTEN in regulating the retromer complex and protein trafficking.

The de novo design of porphyrin binding four-helix bundles (maquettes) is a proven strategy to both mimic and expand on the natural functionality of heme-containing proteins, including electron transfer, oxygen transport and catalysis. The rational design of maquettes utilizes simple principles to construct robust and adaptable heme-binding helical structures which can be iteratively engineered towards new functions. However, maquette designs have historically adopted highly dynamic conformations. Although this does not impede or may be integral to their function, enabling the construction of flexible structures capable of accessing diverse oxidoreductase catalysis, the lack of a unique conformation in these designs hinders structural determination at atomic resolution by techniques such as crystallography or NMR. In contrast, computational protein design is advancing to a stage where a range of protein structures can be routinely constructed with atomic accuracy, a level of precision which could be invaluable to engineer maquette designs towards novel properties. In this project, computational design approaches were integrated into the construction of maquette structures, combining the strengths of functional, rational design strategies with precise computational techniques. The crystal structure of a two-heme binding four-helix bundle maquette was obtained by modification of the previously reported D2 peptide into a single-chain, in vivo expressed design. Computational techniques were employed to construct single and multi-heme binding variants which retain the architecture and stability of this initial structure by the design of precise protein interactions in a hydrophobic core or extending the helical bundle into a four-heme molecular wire, with further structural insights gained by NMR spectroscopy and cryoEM. Furthermore, computational design was harnessed to construct novel c-type cytochromes from scratch, inspired by the recent design of the artificial peroxidase, C45. This work presents novel structures for the continued engineering of oxidoreductase functionality in maquettes and a platform for precise investigation of the fundamental properties of protein-porphyrin complexes.

The aim of this project was to study the role of TRAF6 and TRAF6-mediated signals in immune cells. Initially this project focused on characterising the phenotype of a TRAF6 mutant mouse model in which an E3 ligase inactive mutant of TRAF6 was knocked in to all cells of the body (TRAF6[L74H]). Mice lacking TRAF6 E3 ligase activity were shown to develop a severe inflammatory autoimmune phenotype, characterised by systemic organ inflammation, elevated serum titres of immunoglobulins, autoantibodies and dysregulation of the peripheral T cell populations. While thymic T cell populations were comparable to wildtype, peripheral T cell populations were dysregulated, with increased numbers of T follicular helper cells, T regulatory cells and CD8+ T cells observed. These data therefore highlight an important role for TRAF6 E3 ligase activity in immune homoeostasis. The second part of this thesis examined how cytokines that activate TRAF6 control CD8+ T cell function. Using CD8+ cytotoxic T cells as a model system, we observed that the IL1 family cytokines IL18 and IL33 modulate CTL metabolism. Acting in combination with IL12, which induces IL1 family receptors, we observed that IL18 and IL33 induce a short-lived acute metabolic response. Characterisation of this response using high resolution mass spectrometry revealed that IL18 and IL33 controlled core metabolic proteins including translational machinery such as the EIF complex, as well as a repertoire of essential nutrient transporters including those responsible for the uptake of Glucose. We further show that underlying this ability of IL18 and IL33 to control CTL metabolism is their ability to sustain expression of key nodes in CTL metabolism such as the transcription factor Myc and activity of mTORC1. Taken together, these results highlight how alarmins control CD8+ effector T cells as part of the immune systems response to infection or damage.

Common cancer treatments, such as radiotherapy, primarily involve the induction of DNA damage. Like healthy cells, the cancer cells rely on the highly conserved DNA damage response (DDR) checkpoints for survival, leading to therapeutic resistance. There are established DDR checkpoints in each interphase stage of the cell cycle, which impede the cell cycle until the damage has been sufficiently repaired. However, there is limited evidence of a specific DDR checkpoint in mitosis. The spindle assembly checkpoint (SAC) is a protective mechanism that regulates mitotic spindle attachment to the centromere, regulating metaphase-anaphase transition. There is emerging evidence to suggest crosstalk between the DDR and SAC pathways in damaged conditions. I have consistently observed a delay in mitotic transit time following the induction of DNA damage by various agents including irradiation, chemotherapeutics and H2O2. Overall, this indicates the existence of a mitotic DNA damage checkpoint (MDDC), which may contribute to treatment resistance. This thesis aims to identify proteins involved and begin to establish a pathway for this novel checkpoint. Firstly, an examination into the antioxidant enzyme superoxide dismutase 1 (SOD1), which was highlighted in a preliminary siRNA DNA damage screen of potential proteins involved in the checkpoint. I observed the MDDC was dependent on SOD1 and revealed SOD1 regulates phosphatase PP2A activity to activate the MDDC. Furthermore, a role of SOD1 in DNA damage repair was also identified. Then a mass spectrometry screen was performed to reveal novel interactors of SAC protein BubR1, identifying the nonsense mediated decay (NMD) factor, Upstream frameshift 1's (UPF1) involvement in the MDDC. Overall, an initial pathway for the MDDC has been established and future direction determined, which will provide potential drug targets to reduce therapeutic resistance of cancer cells to conventional treatments.

Studying the role of various factors like pressure both osmotic and hydrostatic, temperature, and chemical denaturants on protein structure, function, solubility and stability is very helpful in understanding many biochemical processes. Franz Hofmeister has studied the salting out properties of salts on proteins and has derived a lyotropic series of ions. According to the structure maker/breaker theory, kosmotropes withdraw water from the protein decreasing solubility and increasing stability, whilst chaotropes are weak in withdrawing water, increasing solubility and decreasing stability. Jordan Bye and coworkers refined the structure maker/breaker theory and focused on the water organizing ability of ions rather than direct interactions with protein surfaces. Their findings confirm that sulphate stabilizes whilst thiocyanate destabilizes proteins. It is believed that osmolytes like TMAO, ectoine, and betaine that are produced and accumulated by marine organisms as an intracellular solute, to counteract the destabilizing effects of osmotic stress, denaturants such as urea, and high pressure can counteract protein destabilizing effects and enable the cells to carry out cellular functions without affecting protein structure. We have tested the hypothesis that osmolytes work in a similar manner to Hofmeister series ions: that they interact strongly with water and compete with the protein for water binding, which is how they counteract the effect of urea. To find this we have studied the effect of TMAO, ectoine, and betaine on barnase by ¹⁵N HSQC, ¹³C HSQC, and 2D-HNCO experiments and compared them to the effect of sodium thiocyanate on barnase. We have also studied the effects of osmolytes on protein denaturants like urea by ¹H NMR. DSC results suggest that TMAO may not strongly stabilize the protein by itself but can counteract the destabilizing effects of urea. NMR data show that the effects of osmolytes on NMR spectra of the protein barnase are smaller than those of typical Hofmeister ions such as sulphate and thiocyanate. NMR Δδ also show that ectoine binds very weakly and betaine does not bind, TMAO binds stronger than other osmolytes and chloride but weaker than sulphate and thiocyanate. Osmolytes have a clear influence on protein stability, but they clearly do this without significant protein binding. Data from mixed osmolyte solutions also suggest that osmolytes do not bind to urea either. All this data suggests that binding may not be necessary to cause chemical shift changes and effects of ions on protein is independent of their ability to bind. All this implies that osmolytes affect the protein stability by their ability to withdraw water from the protein surface – and therefore counteract the perturbations induced by other solutes.

Redox reactions play important roles in the biological processes, especially in cellular respiration. The enzyme with redox-active metal cofactors are the main players in the processes. Despite the various metal cofactors are available, the nature selects limited number of them for electron transfer (ET) processes including cytochromes, iron–sulfur clusters, and cupredoxins. For this reason, the metalloenzymes involving in electron transfer were investigated structurally and biochemically. The first enzyme is CuNiR in denitrification pathway and the other is cytochrome bc1 in cellular respiration. First part of the thesis covers the purification of cyt bc1 from Plasmodium falciparum. The parasite Plasmodium falciparum is the cause of one of the most dangerous disease malaria, which has been killing people over centuries. The emergence of drug resistance parasite strain triggers an urgent development of new drugs against malaria. Cyt bc1 from P. falciparum has been an approved drug target for malaria as its function involves in pyrimidine synthesis rather than energy production. The aim of this part is to purify and identify cyt bc1 from P. falciparum. The purification methods including DEAE, DEAE-gel filtration, and oxidized cyt c resin were performed on the membrane fraction extracted from Plasmodium falciparum cell lysate. The low specific activity of cyt bc1 is present in protein purified by oxidized cyt c resin. The identity of cyt bc1 from P. falciparum was confirmed by mass spectrometry and only 3 out of 7 subunits of P. falciparum cyt bc1 can be detected from the protein purified by DEAE-gel filtration. Copper nitrite reductase (CuNiR) catalyses the reduction of nitrite to form gaseous nitric oxide (NO) as the product. CuNiR is the key enzyme in the denitrification pathway, which the enzyme produces NO, the precursor of the potent greenhouse N2O gas. Recently, the new class of CuNiRs called three-domain CuNiRs have been characterized in which an extra cyt c domain or cupredoxin domain is tethered at either at C-terminus and cupredoxin domain at N-terminus of the enzymes. The structures of cyt c tethered CuNiRs from two species: Ralstonia pickettii (RpNiR), and Pseudoalteromonas haloplanktis (PhNiR) have been elucidated. The unique characteristics of CuNiRs in this class is that the substrate channel is blocked by a tyrosine residue, in which the mechanism for activating the blocked Tyr is elusive. Comparing to the prototypic CuNiR, the knowledge on three-domain cyt c is less limited. The aim of this work is to investigate the RpNiR mutants that are not located at the active site and perturb the enzymatic activity of the enzyme. The strategy was performed by site-directed mutation as followed: The first group of mutants involves in electron transfer as followed A. Mutation of axial ligand T1Cu, Met148 to Leu (Met148Leu), B. Mutation of the residue locating at the linker loop between cyt c domain and core domain, Ser321 to Met (Ser321Met). The second group of mutants involves in protron transfer as followed: C. Mutation of the residue at the constraint hydrophobic proton cavity, Phe295 to Leu (Phe295Leu) D. Mutation of the residue locating in the primary proton channel, Gln262 to Asn262 (Gln262Asn). The structure-activity relationship of each mutant was also investigated. The structure of NO bound RpNiR can be obtained from the two mutants Met148Leu, and Phe295Leu, however the nitrite bound RpNiR cannot be achieved in this work. The significance of each residue (Met148, Phe295, Ser321, Gln262) has been proposed, which implicated in electron transfer from T1Cu to T2Cu, the limitation of substrate/small molecules cavity, and the functional primary proton channel in three-domain CuNiR.

South Asians are known to have an increased risk of type 2 diabetes, at lower levels of adiposity and at a younger age, in comparison to Caucasians. In addition, there is evidence of insulin resistance amongst South Asians, from early years of life. The exact mechanisms for this increased risk are not fully understood. Incretins are insulinotropic intestinal hormones accounting for a significant amount of postprandial insulin release. Both GLP-1 and GIP are now recognised as key hormones responsible for the incretin effect. The incretin effect is impaired in type 2 diabetes and remains a pivotal therapeutic target with the advent of incretin-based drugs. There is emerging evidence that incretin hormones may vary amongst Asians. Incretin-based drugs have been found to be more efficacious amongst Asians including South Asians. However, there are very limited mechanistic incretin studies amongst South Asians. We compared two key aspects of the incretin hormone responses between normal glucose tolerant South Asians and Caucasians. This included comparing the incretin effect as well as total GLP-1 and GIP responses to isocaloric liquid meals with varying compositions (carbohydrate rich, mixed and fat rich meals). Despite the presence of hyperinsulinaemia, normal glucose tolerant South Asians had a comparable incretin effect following a 50-gram glucose load. To the best of our knowledge, this is the first study making contemporaneous comparisons between incretin hormone responses between normal glucose tolerant South Asians and Caucasians.