The high mobility group of nucleosome binding proteins (HMGNs) are chromatin architectural proteins that bind specifically to nucleosomes and influence chromatin structure and DNA-dependent functions. However, the mechanisms underlying these events remain largely unknown. HMGN1 and HMGN2 are highly expressed by embryonic stem cells and are downregulated as differentiation proceeds. Nevertheless, embryonic and adult neural stem cells retain elevated levels of these proteins. Chromatin plasticity is essential for the pluri- and multipotency of stem cells and it is achieved by maintaining an open and dynamic chromatin conformation. Conversely, developmental potential seems to be restricted by chromatin condensation. The present work shows that loss of HMGN1 or HMGN2 in pluripotent embryonal carcinoma cells leads to increased spontaneous neuronal differentiation, which is accompanied by a reduction in pluripotency markers and higher gene expression of lineage-specific transcription factors. Inhibition of signalling pathways relevant for neurogenesis does not re-establish the phenotype observed in Hmgn2-knockout cells. Withdrawal of the factors sustaining pluripotency in embryonal carcinoma cells results in higher induction of pro-neural factors in cells lacking HMGN1 or HMGN2. Neural stem cells derived from Hmgn-knockout cells also display higher gene expression of pro-neural transcription factors and increased spontaneous neuronal differentiation. Loss of HMGN2 disrupts the active histone modification landscape, and therefore affects the chromatin structure at local and global levels. The proposition is that the local changes directly influence the transcription rates of pluripotency and lineage-specific transcription factors, while the global changes may restrict chromatin plasticity. The present data support a hypothesis whereby HMGNs contribute to the chromatin plasticity of stem cells by promoting an active histone modification landscape and open chromatin conformation, which are essential for preserving the self-renewal and developmental potential of stem cells.

This thesis describes the development of new methods for the synthesis of 1,2- diamines. Chapter one reviews current methods for the synthesis of 1,2- diamines, and their importance in chemistry. Chapter two highlights attempts to synthesise 1,2-diamines using two nitrogen-containing heterocycles, namely 3- phenyl-1,3,5-triazabicyclo[3.2.0]hept-6-ene-2,4-dione and imidazolin-2-one, which both contain an endocyclic double bond. It includes the synthesis of a novel 1,2-diazetine as well as the functionalisation of imidazolin-2-one via a palladium-catalysed cross-coupling reaction with phenyl iodide. Subsequent hydrogenation and hydrolysis was then utilised to afford 1-phenylethane-1,2- diamine dihydrochloride. Chapter three describes the synthesis and functionalisation of a range of 3-methylene-1,2-diazetidines that were subsequently hydrogenated in an asymmetric fashion, with [Rh(NBD)2]BF4 and ligand Mandyphos M004-1, to yield 1,2-diazetidines with up to 89% ee. Reduction with LiDBB allowed for the synthesis of two carbamate-protected 1,2-diamines in three steps. The first examples of epoxidation, reaction with tetracyanoethylene and 1,3-dipolar cycloadditions of 3-methylene-1,2- diazetidines are reported. Chapter four details the experimental procedures and characterisation data for the novel compounds produced.

In chapter one, recent progress in the synthesis of functional glycopolymers was described. In chapter two, combination of catalytic chain transfer polymerisation (CCTP) with both thiol-ene and copper catalysed alkyne azide coupling (CuAAC) click chemistry has been employed to give a new route to functional glycopolymers for applications in bioconjugation and biological targeting. Ring opening of poly (glycidyl methacrylate) with sodium azide and subsequent reaction with alkyne functional carbohydrates, as prepared by Fischer glycosylation, has been exploited. This combination of a range of efficient chemistry gives a route to multi gram quantities of glycopolymers avoiding the need for living radical polymerisation chemistry. In chapter three, multi-block glycopolymers made of mannose, glucose, fucose and di(ethylene glycol) ethyl ether acrylate monomers were synthesized by Cu(0) wire mediated single electron transfer living radical polymerization (SET-LRP). These highly narrow disperse glycopolymers were then tested for binding and inhibition of DC-SIGN, a protein important for HIV infection. In chapter four, a series of cyclodextrin-based glycoconjugates, including glycoclusters and star glycopolymers, were synthesised via combination of CuAAC click reaction and copper-mediated living radical polymerization. These glycoconjugates show high affinity in binding with human DC-SIGN lectin and could be used as inhibitor to prevent the binding of HIV envelope protein gp120 to DCSIGN at nanomolar concentration. The star block glycopolymer show high loading capacity of hydrophobic anti-cancer and anti-HIV drugs, indicating promising application in HIV-therapeutic and smart drug delivery. In chapter five, a new approach to perform SET-LRP in pure water is described. The key step in this process is to allow full disproportionation of CuBr/Me6TREN (Me6TREN = tris(dimethylamino)ethyl amine) to Cu(0) powder and CuBr2 in water prior to addition of both monomer and initiator. This provides an extremely powerful tool for the synthesis of functional water-soluble polymers with controlled chain length and narrow molecular weight distributions (PDI approx. 1.10), including poly- NIPAM, DMA, PEG acrylate, HEA and glycomonomers. The polymerizations are performed at or below ambient temperature with quantitative conversions attained in minutes. Polymers have high chain end fidelity capable of undergoing chain extensions to full conversion or multi-block copolymerization via iterative monomer addition after full conversion. Activator generated by electron transfer atom transfer radical polymerization (AGET ATRP) of NIPAM in water was also conducted as a comparison with the SET-LRP system. This shows that the addition sequence of Lascorbic acid is crucial in determining the onset of disproportionation, or otherwise. This robust technique was applied to polymerizations under biologically relevant conditions (PBS buffer), a complex ethanol/water mixture (Tequila) and a truly biological condition: sheep blood serum. Finally, double hydrophilic diblock glycopolymers composed of mannose glycopolymer block and LCST PNIPAM or PDEGEEA block were successfully synthesized by this aqueous SET-LRP technique.

The thesis presents the study of the reaction of glyoxal (ethanedial) with polypeptides. This reaction is important in the food industry as well as during ageing and diabetes mellitus. To study this reaction a Fourier transform ion cyclotron resonance mass spectrometer coupled with electron capture dissociation and collisionally activated dissociation was used. Initially this reaction was carried out in the neuropeptide Substance P to set up the reaction conditions, sample preparation, as well as the instrumental parameters in the mass spectrometer. The results in Substance P revealed two compounds, with mass additions assigned as C2O and C2H2O2 from glyoxal, were formed. MS/MS results showed that the modification site for both species could be located at either the arginine residue or at the N-terminus. Thus, in order to distinguish N-terminus from arginine modification the position of the arginine was varied in four model peptides. The results indicated that both mass additions C2O, C2H2O2 were located at the arginine residue. Interestingly, two of those model peptides showed an unusual mass addition of 21.9843 Da, which was assigned as a new type of glyoxal modification at the arginine residue showing the addition of two carbon atoms from glyoxal and the loss of two hydrogen atoms from the peptide (C2-H2), herein referred to as 2-imino-imidazole. In order to assess the involvement of other residues in the reaction with glyoxal a new set of experiments in acetylated and non-acetylated undecapeptides were carried out. Unexpectedly, these experiments revealed that two species with the same mass (16.01092 Da) were being formed in the non-acetylated peptide. One of the species corresponded to diglycation, where the results suggest that the glyoxal binding at the lysine residue is crosslinked with the N-terminus. The second species showing the addition of 116.01092 Da was formed at the arginine residue forming a species, here called a glyoxal dimer, at the arginine residue. The formation of the glyoxal dimer species was also observed in the acetylated peptide. Although is clear that crosslinking between the lysine residue and the N-terminus is not possible in the acetylated peptide, the results seem to indicate that crosslinking between the amino group of the lysine and the amide group of glutamine could occur. However, a systematic study varying the position of the lysine relative to the glutamine residue and also relative to the N-terminus needs to be addressed in the future in order to determine the extent of the involvement of the N-terminus and amide group in the glyoxal glycation reaction.

Electron capture dissociation (ECD) is a powerful and superior tandem mass spectrometry (MS) fragmentation technique in the study of protein post-translational modifications (PTMs) due to its unique features of preserving labile modifications and providing more detailed sequence information, which has been used to study protein platination and disulfide linked proteins. Cisplatin was found cross-linking multiple methionine (Met) pairs on calmodulin (CaM). The cross–linking of cisplatin to apo–CaM or Ca–CaM can inhibit the ability of CaM to recognize its target proteins as proved by a melittin binding assay. To further establish MS strategies to quickly assign the platinum-modification sites, a series of peptides with potential cisplatin binding sites were reacted with cisplatin and then analyzed by ECD. Radical-mediated side chain losses from the charge-reduced M+Pt species (such as CH3S• or CH3SH from Met, SH• from Cys, CO2 from Glu or Asp, and NH2• from amine groups) were found to be characteristic indicators for rapid and unambiguous localization of the Pt-modification sites on certain amino acid residues. Furthermore, the potential of cisplatin as a protein crosslinking reagent was further explored and demonstrated on other peptides and proteins. Many of the inherent features of cisplatin make it an interesting cross-linking reagent, such as targeting new protein functional groups (thioether and imidazole groups), its unique isotopic pattern, its inherent positive charges, its potential of binding to different functional groups, etc. However, it was found that the distance constraints obtained from NMR structures of CaM are inconsistent with the measured distance constraints by cross–linking. Therefore, a newly developed flexibility simulation method was applied to explore whether the flexibility motions of CaM might contribute to the observed Pt-crosslinking on CaM. The flexibility analysis showed that the structural flexibility of CaM is key to cisplatin crosslinking CaM. ECD mechanism of disulfide bonds is still under debate. To further explore the ECD mechanism of sulfur– containing species, a series of disulfide (S–S), sulfur–selenium (S–Se), and diselenide (Se–Se) bond–containing peptides was studied by ECD. The results demonstrate that the radical has higher tendency to stay at selenium rather than sulfur after cleavage of Se–S bonds by ECD and suggest that direct electron capture at Se–Se and C–Se bonds is the main process during ECD of inter–chain diselenide peptides. Last but not least, a new active ion ECD (AI-ECD) method, named Shots-ECD, was developed and applied to improve Top-down ECD backbone fragmentation efficiency of disulfide-rich proteins. The results show that the Shots–ECD approach can not only cleave multiple disulfide bonds but also significantly improve the backbone cleavage efficiency. This strategy is fast, efficient, and with no need of chemical reduction of samples and instrument modification, and therefore can be a powerful approach to improve top-down ECD efficiency of not only disulfide bonded proteins but all proteins by Fourier transform ion cyclotron mass spectrometry (FTICR MS).

Prions diseases are a group of fatal neurodegenerative disorders called the transmissible spongiform encephalopathies (TSEs), which include bovine spongiform encephalopathy in cattle, scrapie in sheep and Creutzfeldt-Jakob disease (CJD) in humans. TSEs are associated with the conversion of normal cellular form of the prion protein (PrPC) to an altered pathological form (PrPSc). An important phenomenon known as the species barrier affects prion transmission, resulting in longer incubation time and lower incidence of disease upon transfer between species. Another feature of prion diseases is diseasemodulating polymorphisms in PrP sequence which can alter individual‟s susceptibility to infection. This thesis investigates two properties of PrP that may elucidate the mechanisms underlying both species barrier and disease resistance; (i) effect of diseasemodulating mutations on folding kinetics of PrP and (ii) impact of diseasemodulating mutations on formation of PrP fibrils. Equilibrium and kinetic folding studies demonstrate that the folding pathway of PrP is affected by mutation Q167R which confers disease resistance, and mutations S170N, N174T and S170N/N174T characteristic for Chronic Wasting Disease in cervids, which are known to increase disease susceptibility. The destabilising effect of Q167R mutation previously observed via equilibrium folding studies was confirmed through direct kinetic observations. Subsequent fibrilisation experiments suggested a possible link between the stability of mouse prion protein and its propensity to form fibrils, elucidating a potential mechanism of increased disease resistance conferred by Q167R mutation. Equilibrium folding studies of S170N, N174T and S170N/N174T revealed a surprising correlation between the structural effects of these mutations and fibrilisation propensity. Based on these findings, a disease resistance mechanism centred on decreased formation of neurotoxic particles in the organism as well as diminished ability of infectious oligomers from both inside and outside to propagate oligomerisation of PrP has been proposed.

Membrane proteins are involved in a wide range of vital cellular processes, responsible for relaying signals and cargo across cell membranes. A deeper molecular understanding of their function is essential to elucidate the mechanisms of numerous diseases and medical conditions. As an example, the carnitine palmitoyltransferase 1 (CPT1) enzymes are responsible for the regulation of mitochondrial fatty acid oxidation thus are central to cell function. Previous work had shown that the transmembrane (TM) domains of the mitochondrial outer membrane protein CPT1 have a significant influence on the enzyme’s kinetics, which is different in the two catalytically active isoforms, CPT1A and CPT1B. It was also shown that TM-domains, and specifically TM2, are involved in driving the oligomerisation of the full length CPT1A. For this reason, the study of TM-TM oligomerisation and their roles in the function of CPT1 are fundamentally important in the design of pharmacological strategies aimed at the modulation of the activities of these enzymes in conditions such as diabetes. The main focus of this PhD research was to systematically investigate the homo- and hetero-oligomerisation of the TM domains to help better understand the structurefunction relationship for CPT1 membrane proteins. In this project, different techniques were used from the areas of chemistry, molecular biology, and biophysics (in vivo oligomerisation assays, chemical crosslinking, circular dichroism and analytical ultracentrifugation), in order to examine these interactions. The membrane spanning sequences of rCPT1A and rCPT1B were found to selfassociate, and the order of oligomerisation was also determined. Sequence motifs likely to be responsible for the interactions of these TM domains were identified using in silico modelling. Mutagenesis analyses confirmed the role of suggested GxxxG(A) motifs in the homo-oligomerisation of rCPT1A TM2 domains. The heterooligomerisation of TM1-TM2 domains was also studied, and was found to be significant in both isoforms. The work presented in this thesis shows that the membrane spanning regions of the CPT1 enzymes are capable of interacting through both homo- and hetero-oligomerisation. The results strongly suggest that these interactions may play a significant role in the complex formation of the full length enzymes and provide further evidence that CPT1 may function as a channel in the outer mitochondrial membrane.

The reaction of salmon calcitonin (sCT) with various linear PEG and PolyPEGs was investigated under a variety of reaction conditions. The optimal experimental conditions for the production of high yield monoconjugated sCT were determined using RP-HPLC, SEC-HPLC, IE-FPLC and MALDI-TOF MS analysis. Semi-preparative IE-FPLC was used to purify high yield mono conjugated sCT-5 kDa linear PEG, sCT -6.5 kDa PolyPEG, sCT-20 kDa PolyPEG and sCT-40 kDa PolyPEG coupled to the N-terminal primary amine. The products were analysed using RP-HPLC, SEC-HPLC, SDS-PAGE and the site of attachment was determined by Tryptic mapping combined with LCMS, RP-HPLC and MALDI-TOF MS. Biological activity was measured invitro, sCT derivates were incubated with T47D cells and an ELISA assay was used to measure the amount of cAMP released. The purified conjugates retained a high degree of biological activity compared to the unmodified peptide. Incubation of sCT and sCTconjugates with proteolytic enzymes showed that for sCT-5 kDa linear PEG and 6.5kDa PolyPEG a high degree (60-70 % of the activity at time zero) of biological activity remained after 30 minutes compared to 0% activity for native sCT.

The three-dimensional conformation of a protein is central to its biological function. Mass spectrometry (MS) has become an important tool for the study of various aspects of protein structure. This project investigates the use of MS for diagnosis of hemoglobinopathies, through primary structure identification, and for threedimensional protein structure analysis, through comparison to established methods and application to protein systems. Travelling-wave ion mobility mass spectrometry (TWIM-MS) was used to investigate the biological significance of gas-phase protein structure. Protein standards were analysed by TWIM-MS. Cross-sections were estimated for proteins studied, for charge states most indicative of native structure, and were found to be in good agreement with those calculated from published X-ray crystallography and nuclear magnetic resonance structures. These results illustrated that the TWIM-MS approach can provide biologically-relevant data on three-dimensional protein structure. TWIM-MS was then used to study the structural properties of the hemoglobin tetramer and its components. Results showed that globin monomers exist in similar conformations whether in apo- or holo- forms and that a heme-deficient dimer is unlikely to be a prerequisite for hemoglobin tetramer assembly. TWIM-MS was used to successfully differentiate between normal and sickle hemoglobin tetramers. The conformational changes occurring in VanS, a histidine kinase, upon autophosphorylation were investigated by TWIM-MS. Results provided insights into the mechanism of autophosphorylation. MS was used to follow the rate of the autophosphorylation and results obtained compared well with those from an established method. This demonstrated that MS offers a simple, reproducible alternative to conventional methods for the study of phosphorylation rates. MS was used to provide positive identification of a range of hemoglobinopathies caused by single point mutations. A high-throughput method was used to screen for hemoglobinopathies in South Asians with and without cardiovascular disease. Results showed a positive correlation between patients with hemoglobinopathies and those with cardiovascular disease.

The endothelium acts to maintain vascular homeostasis, including the regulation of vascular tone, blood fluidity and coagulation. Endothelial dysfunction, a condition largely characterised by reduced NO bioavailability, is an important feature associated with the aetiology of several pathophysiological disorders including type 2 diabetes and cardiovascular disease. AMPK is the downstream component of a protein kinase cascade important in the regulation of cellular and whole body metabolism. AMPK has been demonstrated to mediate a number of physiological responses in the endothelium, including the stimulation of eNOS phosphorylation and NO synthesis; and as such AMPK represents a therapeutic target in the dysfunctional endothelium. VEGF has been established as the prime angiogenic molecule during development, adult physiology and pathology. VEGF stimulates NO production, proposed to be a result of phosphorylation of Ser-1177 on eNOS, a residue also phosphorylated upon AMPK activation in cultured endothelial cells. The present study, utilising HAEC as a model, provides the first demonstration that AMPK is activated by physiological concentrations of VEGF; and furthermore, partially mediates VEGF-stimulated phosphorylation of eNOS on Ser-1177 and subsequent NO production. In addition, the present investigation demonstrates that the upstream AMPK kinase CaMKK is responsible for these VEGF-mediated effects. VEGF is known to increase intracellular calcium levels in endothelial cells via the generation of DAG and IP3. DAG increases Ca2+ influx through a family of non-selective cation channels, whereas IP3 promotes the release of Ca2+ from intracellular stores. High potassium-induced depolarisation, which reduces the driving force for Ca2+ entry through non-selective cation channels in endothelial cells, abolished VEGF-mediated AMPK activation, whereas the IP3 receptor blocker 2-APB was without effect. Exposure of HAEC to a DAG mimetic (OAG) also stimulated AMPK, an effect which was sensitive to the CaMKK inhibitor STO-609 and high potassium induced depolarization. The functional effects of VEGF-stimulated AMPK were also assessed in HAEC. Ablation of AMPK abrogated VEGF-stimulated HAEC migration and proliferation, two key features of the angiogenic process. While AMPK was necessary for VEGF-stimulated endothelial cell proliferation direct activation of the kinase was insufficient to induce this process. AICAR-stimulated AMPK activation has been demonstrated to stimulate fatty acid oxidation in endothelial cells. However, exposure of HAEC to VEGF did not alter fatty acid oxidation in the present study. Together, the current investigation suggests that a VEGF-Ca2+-CaMKK-AMPK-eNOS- NO pathway is present in HAEC, and furthermore, that AMPK is required, albeit insufficient, for the VEGF-stimulated angiogenic response.