Proteins play a crucial role in almost all biological processes. Developing a complete understanding of the link between their structure, dynamics and function is goal of many areas of scientific research. One tool with which the protein structure has been investigated is infrared (IR) spectroscopy. IR is a useful probe of protein structure because the amide I region (1600-1700 cm-1) is sensitive to the secondary structure elements such as alpha-helices and beta-sheets; different secondary structures give rise to different signatures in the IR spectrum. The drawback of traditional IR is that the amide I region is often broad and featureless and thus difficult to interpret. Two-dimensional infrared spectroscopy (2DIR) can improve on the structural sensitivity of IR by spreading the transitions over a second frequency domain resulting in off-diagonal peaks that quantify the coupling between molecular vibrations. The development of the technique has been greatly aided by computational calculations of 2DIR spectra, such as from molecular dynamics (MD) simulations. In this thesis we apply the exciton method to calculate IR and 2DIR of Leu-enkephalin, a pentapeptide that is involved in the mediation of pain in the body by binding to opioid receptors. The calculated IR show qualitative agreement with both previously calculated and experimental spectra. Previous calculations gave results only for a single structure in the gas phase, and we have expanded this work to including spectra from MD simulations. Our work contributes calculated spectra to aid further understanding of the dynamics of Leu-enkephalin, which may help in the search for more effective opioid analgesics. In addition to enkephalin, we investigated four variants of the enoyl-acyl carrier protein reductase enzyme InhA found in Mycobacterium Tuberculosis, the bacterium responsible for tuberculosis, which is a threat to global health. In particular, we investigate both wild type variants and mutant known to exhibit resistance to isoniazid, one of the front-line treatments for tuberculosis. Due to the size of the protein, it is currently not a suitable candidate for theoretical 2DIR calculations. Instead we used data extracted from the one exciton Hamiltonian to probe the structural dynamics of the different variants. Our work supports previous experimental results, in particular work that had suggested the importance of a ~20 residue binding loop in mediating isoniazid resistance, and suggest several residues as potential candidates for isotope-labelled 2DIR experiments. The work in this thesis provides a starting point for further investigation of the dynamics and calculated spectroscopy of both Leu-enkephalin and InhA.

This thesis details the design and synthesis of a small library of 19F MRI MMP probes containing an MMP substrate bound to a paramagnetic agent (gadolinium contrast agent) with a fluorine containing group at the opposite terminus. The largest obstacle in the synthesis of the probe was the conjugation of the paramagnetic agent. Multiple different modifications to the GdIII chelators AAZTA (6-amino-6-methylperhydro-1,4-diazepinetetraacetic acid) and DOTA (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid) are detailed, as well as methods for the conjugation. Once a successful method was achieved the probes were carried forward for in vitro testing. Whilst the agent was whole or non-cleaved the paramagnetic relaxation effect (a distance dependent effect) of the GdIII eliminated the 19F NMR/MRI signal. The introduction of an MMP to the probe caused the probe to be cleaved. This cleavage then resulted in the increased distance between the GdIII complex and the 19F containing group, which in turn reduced the PRE, resulting in the emergence of a 19F NMR/MRI signal. This enzymatic activity was visualised using high field NMR (600 MHz AV(III)400) by the increase in signal peak height with 19F NMR, as well as the changes in the T1 and T2* (observed as a change in linewidth) values observed before and after cleavage of the probe showed a diagnostic change in the magnetic properties of the probe. Reaction rates were ascertained whilst altering the solvent (H2O vs D2O), and the temperature of the reaction. A reduction in the rate of cleavage was noticed as the D2O concentration was increased as expected, however, the change in temperature did not always follow an expected reaction profile change.

USP20 is a deubiquitinating enzyme that is involved in a number of important cellular pathways, including thyroid metabolism, hypoxic response, seven transmembrane receptor signalling, NF-κβ signalling, centrosome homeostasis and DNA repair. Of recent, it is becoming a major deubiquitinase involved in regulating the DNA-damage response pathway and cell cycle checkpoints. The protein consists of a zinc finger domain, catalytic domain and two ‘domain present in USP’ (DUSP) domains; an architecture shared only with its paralogue USP33. There is no structural information on any of the domains of USP20, so crystallisation trials of the domains of USP20 were performed in order to solve their structures by X-ray crystallography. In addition, yeast two-hybrid (Y2H) and in vitro assays were used to further characterise known and putative interactors of USP20. Finally, the zinc finger domain and DUSP domains were used in pull down assays to identify USP20-interacting proteins from HEK293 lysate. Two stable and well-expressing constructs of the zing finger domain (USP20 1-101 and 1-108) were purified and set up for crystallisation trials. Buffer screens were also performed on the USP20 1-101 construct to increase its stability for crystallisation. Monodisperse, pure protein of any catalytic domain-containing construct of USP20 was unobtainable; only a trigger factor-tagged full length USP20 was purified and active. Two constructs containing the double DUSP domains were produced (USP20 686-914 and 686-894), and both suffered from a low solubility limit. Buffer screening was used to increase its stability, which identified ethylene glycol as a stabilising additive. Due to the nature of commonly used solubility tags, novel tags were designed that would potentially benefit the crystallisation of the fusion construct. Identified from the PDB and literature searches, the calponin homology domain from human β-spectrin (PDB code 1BKR) and the receiver domain from Myxococcus xanthus social motility protein frzS (PDB code 2GKG) were used. Both new tags, as well as MBP were fused to the N-terminus of the DUSP domains (USP20 686-894) to enhance solubility and crystallise the DUSP domains. 2GKG was an effective solubility tag, increased the solubility of the DUSP domains to near that of the MBP fusion. 1BKR, however, was only marginally useful as a solubility tag. In total 97 crystallisation trials were set up for all constructs of USP20, but no crystals containing USP20 protein formed. Y2H assays were used to investigate the interaction between USP20 domains and Β-arrestin-1, TRAF6, RAD17 and PLK1. Of these, only and interaction between USP20’s DUSP domains (residues 686-894) and full length PLK1 was observed. Interestingly, further Y2H and ELISA showed a non-canonical, binary interaction between the poloboxes of PLK1 (residues 367-603) and the DUSP domains. Pull down assays produced a list of possible novel interactors for USP20. These include proteins implicated in processes known, and unknown, to involve USP20. Finally, using ELISA, thermal shift assays and ITC, it was shown that the zinc finger domain of USP20 does not bind to ubiquitin.

The protective effect of female sex in renal ageing and cardiovascular function is widely accepted, but poorly understood. Previous evidence has suggested a role for the nitric oxide and renin-angiotensin systems, though the precise mechanisms by which they elicit these effects remain elusive. Female animals and humans have increased nitric oxide bioavailability with age in comparison to males, and this effect can be negated by ovariectomy surgery, suggesting an interaction between ovarian steroids and nitric oxide. In addition, studies have shown an upregulation of the angiotensin II type 2 receptor (AT2R) in aged females in comparison with males. Whilst incompletely understood, the AT2R is known to mediate vasodilation, nitric oxide release, and can be modulated by oestrogen. Work in this laboratory has shown that the expression of AT2R, renal ageing, and blood pressure may all be sensitive to the nutritional environment encountered during foetal development. This thesis aimed to elucidate some of the mechanisms mediating this ‘protective effect’ of female gender in a rat model of developmentally programmed hypertension and accelerated renal ageing. It was hypothesised that ageing would result in decreased renal function and increased blood pressure. These effects would be significantly altered by sex steroid modulation, and negative effects exacerbated by exposure to a low protein diet during gestation. The mechanisms driving these effects would be, at least in part, linked in changes to renin angiotensin system-regulated nitric oxide release. The data obtained suggested that the nitric oxide system did not significantly change with sex steroid exposure, or in response to maternal diet. Unexpectedly, ovariectomy alone did not change physiological responses as has been described previously. Instead, a significant interaction was observed between exposure to a low protein diet during gestation and ovariectomy. Offspring from mothers fed a low protein diet had impaired responses to removal of ovarian steroids. In addition, low protein offspring had altered vascular reactivity in response to targeted agonism and antagonism of angiotensin II receptors. In conclusion, this work has shown that the protective effect of female gender is more complex than previously described. The data did not support the hypothesis that nitric oxide mediates the beneficial effects of female sex, and targeted stimulation of the AT2R is not an effective means of altering this. Moreover, these data suggest that foetal exposure to a low protein diet may permanently programme altered vascular function, and can significantly affect response to sex steroids.

The deubiquitinating enzyme USP15, a member of the USP family, reverses the process of ubiquitination thereby altering the fate of a plethora of substrates. As such, USP15 has been implicated in a numerous important cellular pathways including cell cycle progression, transcriptional modification and DNA damage repair. The spliceosomal subunit recycling protein SART3 has been shown to bind to USP15 enhancing its deubiquitination of histone H2B thereby providing histone dimers for reassembly during subsequent rounds of transcription and splicing. Furthermore SART3 has also been shown to bind to USP4, a close homologue of USP15. In addition to this histone chaperone activity, SART3 primarily functions by mediating the re-annealing of the U4 and U6 snRNPs to facilitate consequent rounds of splicing in addition to its implication in several disease states including numerous cancer types and HIV through interactions with various cellular proteins. In this thesis ITC, ESI-MS, analytical size exclusion, X-ray crystallography, SAXS and pull-down assays are used to characterise the interaction between USP15 and SART3, solve the structure of the N-terminus of SART3 and identify novel binding partners of the USP15-SART3 complex. The structure of SART3’s N-terminus (residues 96-574) has been solved to 3.04Å, revealing that it is a homodimer comprised of a series of anti-parallel α-helices that form a shape reminiscent of a bowtie. This dimeric arrangement is retained in solution and can bind two molecules of USP15DU. The DU-finger of USP15 plays a pivotal role in co-ordinating the binding interaction with SART3, as small changes to this region can affect the nanomolar affinity interaction between USP15 and SART3. In addition the USP15-SART3 complex appears to interact with several cellular proteins which could have significant impact in several disease states.

Microbial polysaccharides are widely used in many industrial and therapeutic applications as bioactive compounds, drug release and encapsulation materials for pharmaceuticals, and as natural ingredients or additives in food applications. To increase the utilisation or to improve for the food and healthcare applications of these polysaccharides, it is critical to have the detailed knowledge of macromolecular properties and interactions with other polymers. In this study, to properly understand the relationship between physical properties and functionality, the microbial polysaccharides-namely xanthan, gellan, microbial hyaluronic acid (HA) and schizophyllan, which have widely been used for many food and medical applications, were investigated. Taking advantage of recent advances in hydrodynamic techniques, including analytical ultracentrifugation (AUC) based on sedimentation velocity and equilibrium, size exclusion chromatography coupled to multi angle light scattering (SEC-MALS) and viscometry, the solution properties of these polysaccharides were studied. Since conformation analyses are important for macromolecular structure function relationships, the conformation of these microbial polysaccharides were examined using the accomplished method-Multi-HYDFIT. For schizophyllan and gellan, the samples were depolymerised using heat treatment to estimate the conformation. The results indicated that, the three microbial polysaccharides (xanthan, gellan, schizophyllan) adapted the rigid rod conformation whereas another polysaccharide-microbial HA adapted semi-flexible coil structure. Then, the possible interactions of these polysaccharides were investigated with chitosan of different degrees of acetylation using the principle of co-sedimentation in AUC. Xanthan-chitosan mixture presented strong evidence by proving a clear interaction on the basis of changes in the distribution of sedimentation coefficients in the solution form of mixture whereas the mixture of chitosan schizophyllan and chitosan-HA did not show or prove any clear interactions at studied concentrations and solution conditions.

Work presented in this thesis shows how chemical modifications of ssRNAs can increase their nuclease resistance, resulting in potent gene inhibition. The first part of thesis describes how a range of chemical modifications were introduced into the ssRNAs to test their effect on oligonucleotide stability and ability to inhibit gene expression. The ssRNAs were tested in a cancer cell line that stably expresses firefly luciferase. This work demonstrates that DNA nucleotides can be incorporated into ssRNA sequences without loss of potency in the RNAi mechanism. By using additional chemical modifications (2’-OMe and 2’-F) further improvement of ssRNA stability and a significant inhibition of firefly luciferase activity, reaching 91%, was achieved. Moreover, we show that the level of inhibition can be improved when the DNA dinucleotide linked by the (E)-vinylphosphonate is used as a protection of the 5’-end of the ASO (95% KD). The use of ssRNAs rather than dsRNAs is beneficial, since the possibility of off-target effects caused by the remaining sense strand is eliminated and the cost of production of such a medicine is reduced. The second part of this thesis focuses on development of an experimental system that will allow detection of the inhibition of a disease relevant target. Several approaches were used in order to detect KD of the target by the unmodified dsRNAs. First a western blotting technique was employed for detection. Although different ASOs and concentrations were used no KD was observed. More attempts were taken in order to generate target KD, but unfortunately they were unsuccessful so far.

The 76 amino-acid protein ubiquitin is peculiar in its ability to covalently modify substrate proteins. The complexity and diversity in structure and function, of this posttranslational modification is increasingly evident and has been extensively scrutinized. However, the existing notion for ubiquitin to be functionally relevant when covalently linked to substrates was recently found to be dispensable. Pioneering work by Chen’s group has shown that unanchored (non-substrate bound) polyubiquitin chains are just as effective in regulating various cellular processes and presented compelling evidence about the influence of unanchored polyubiquitin chains in regulating cell signalling. That said, not enough emphasis has been placed in developing tools for the direct isolation and characterisation of endogenous unanchored polyubiquitin chains. Our group had previously developed a platform, which employs the Znf-UBP (BUZ) ubiquitin-binding domain of human Isopeptidase T enzyme, to (for the first time) directly purify cellular unanchored polyubiquitin chains of up to 15 ubiquitin moieties long from rat skeletal muscle tissue, also confirming the presence of K48 isopeptide linkages in such chains. Following on, the work described here details further refinement and optimisation of the purification protocol specifically from eukaryotic cell line (HEK293T). We could confirm by western blotting, the presence of K48 and K63 linked endogenous unanchored polyubiquitin chains, some 10 or more ubiquitin moieties long. Our next step was to optimise existing standard protein processing protocols specifically for the study of cellular unanchored ubiquitome. Considering ubiquitin being resistant to trypsin digestion, we found a general lack of consensus in protocols best suited for processing and analysis of the cellular ubiquitome. Extensive optimisations of various protein digestion conditions were performed to arrive on the protocol best suited for unanchored polyubiquitin. The protocol was designed to address the specific problem of maximising (efficient) digestion of ubiquitin, minimising losses in the process to produce samples for quantitative analysis. The purchase of a QqQ mass spectrometer by the group enabled quantitative estimation of polyubiquitin linkages by using AQUA standards although this required significant optimisations as well. Subsequently, we could successfully detect and estimate the abundance of isopeptide linkages purified from HEK293T cell line to be K63 (83%) > > K11 (9%) > K48, K27 (3.7-4.0%) at basal conditions. Moreover, during the course of trying to attain self-sufficiency in this area of work, our knowledge about the pitfalls and caveats in the area also improved. Following the development of a robust in-house platform for the isolation and characterisation of unanchored polyubiquitin chains, we were eager to apply the protocol to study the endogenous unanchored polyubiquitin makeup of cells at basal conditions, and take a step further to determine the change in their levels upon activation of signalling, manipulation of possible regulators and in response to stress. While our attempt at absolute quantification of linkage abundance was not entirely successful, we present data which confirmed increase in K48 and K11 (but not K63) linked unanchored polyubiquitin chains upon proteasomal inhibition and also a strong indication of accumulation of all three linkage types upon suppression of IsoT. Finally, the study has improved our understanding of the regulation of the cellular unanchored ubiquitome.

Chapter One describes several natural products derived from the amino acid tryptophan including the structures of the new natural products hyrtioseragamine A and B and tintamine. Chapter Two describes the work towards hyrtioseragamine A and B. It was intended that the unknown furo[2,3-b]pyrazine-2(1H)-one core would be synthesised from a 1,4-dicarbonyl and this has been synthesised from the amino acid ornithine and the tryptophan derivative kynurenine. Cyclisation of this material to give the 2,3-amino-furan was unsuccessful and thus a number of model systems were synthesised to examine the barrier to furan formation. These models suggest formation of a furan with nitrogen at its 2- and 3-positions was not possible by this route, and further attempts to install a nitrogen to a 2-amino-furan also proved unsuccessful. Chapter Three describes work on the synthesis of the marine natural product tintamine. This novel compound contains a unique tropo-1,2-dihydro-3,6-phenanthroline core and our synthesis was based on biomimetic principles. A Friedlӓnder reaction allowed successful formation of the quinoline and the attached seven-membered ring was further functionalised to a tropolone. Formation of the dihydro-phenanthroline was successful; it was also possible to separately functionalise the tropolone-quinoline with the sulfur side chain present in tintamine. Unfortunately several of the key intermediates of this route proved unstable, and at this point a final deprotection to form the natural product has not been successful. Chapter Four contains the synthetic details for the preparation of the novel compounds described in Chapters Two and Three.

Completion of genome duplication by DNA replication catalyzed at stable replisomes is essential for life by facilitating cell division. Replisomes encounter physical blockage and chemical damage to DNA that frequently threatens to derail replication by inhibiting replisome enzymes. Multiple systems support DNA replication, by detection and repair of DNA damage and removal of physical blocks. Homologous recombination (HR) is one example. Archaeal Hel308 and metazoan HelQ DNA helicases are implicated in DNA repair by HR, in response to toxic DNA interstrand crosslinks that block replication forks. HelQ and Hel308 are single-strand DNA (ssDNA) stimulated ATPases with 3’ to 5’ translocase/helicase activity, most effective at unwinding forked DNA. Their helicase activities are likely to be crucial for promoting DNA replication and repair, but little is known about how. Here, I have been able to generate high yields of highly active human HelQ protein, compared to previous published strategies, and used in vitro biochemistry to show multiple oligomeric states of HelQ that are sensitive to reducing agents. I show that HelQ preferentially targets branched DNA molecules for DNA unwinding similarly to existing data for archaeal Hel308. HelQ and Hel308 demonstrated conservation of function between HelQ and archaeal Hel308 in winged helix domains of both proteins that led to a model for bi-modal DNA binding. This highlights how archaeal Hel308 may also be used as a model for HelQ function.