There is an urgent need for new chemotherapies against human African trypanosomiasis (HAT), caused by the protozoan parasite Trypanosoma brucei. It is anticipated that the parasites’ divergent biochemistry will enable development of novel therapies. To study the behaviour of a complex network as metabolism, one can employ mathematical models. In this thesis, metabolism of bloodstream form T. brucei was investigated. Cellular metabolism consists as a complex system connecting enzymes with metabolites, and to study such a network one can construct mathematical models that describe the connections within the biological system. A previously published, and well-curated model of glycolysis in bloodstream form T. brucei (Bakker BM, et al. (1997) J Biol Chem 272:3207 15), was extended here with the pentose phosphate pathway (PPP), the second major pathway in glucose metabolism in most life forms. Several hypotheses were derived during the model building process and these were tested experimentally. It became apparent that the glycosomal bound-phosphate balance is essential for the parasite. Extension of the glycolytic model with the PPP introduced the risk of a so-called ’phosphate leak’, where bound-phosphates are depleted in the glycosome. Two hypotheses were investigated in silico, while one hypothesis could also be tested experimentally; (i) a glycosomal ATP:ADP antiporter was proposed, but in silico analysis indicated that the activity of such an antiporter requires tight regulation. (ii) A glycosomal ribokinase was investigated both in silico and experimentally. Genetic mutants indicated that ribokinase is essential to bloodstream form T. brucei, albeit at low levels. Additional analysis of the generated models indicated that ablation of 6-phosphogluconate dehydrogenase (6PGDH) in T. brucei is lethal by a different mechanism as seen in other organisms. Overall, extension of the glycolytic model with the PPP demonstrated the fragility of the model regarding the bound-phosphate balance and indicated that future analysis on glycosomal metabolism should be focused on this. Important in the use of mathematical models of metabolism is that the underlying stoichiometry of the model reflects (albeit with simplification) the in vivo system. It is therefore paramount to know what enzyme activities are present in the organism of interest. In this thesis a metabolomics approach was used to elucidate the function of three T. brucei genes. These genes were putatively annotated as arginase (ARG), N-acetylornithine deacetylase (NAO) and nicotinamidase (NAM). The results suggested that ARG has catalytic activity as tryptophan monooxygenase (EC 1.3.12.3), while substrate promiscuity was indicated for NAO and NAM. The work presented in this thesis has provided us with new insights on trypanosomal metabolism. The extended model now allows us to research a larger part of T. brucei metabolism with mathematical modelling, and will thereby aid in the identification and further investigation of (proposed) drug targets.

Polymicrobial diseases arise when multiple microorganisms colonize a host and form multi-species biofilms. Within polymicrobial communities bacteria, fungi, viruses and/or parasites directly and indirectly interact with one another in a multitude of ways. The composition and the interactions between organisms within polymicrobial biofilms govern disease severity and patient outcomes. Polymicrobial infections are of significant interest because of the escalating development of antimicrobial resistance and the increasing involvement polymicrobial biofilms in chronic and systemic infections. The Gram-positive bacteria Staphylococcus aureus and dimorphic fungi Candida albicans have been shown to coexist within the human host in polymicrobial biofilm communities which often result in increased disease severity and mortality. Both of these commensals are opportunistic human pathogens that cause a plethora of infections ranging from relatively non-lethal local infections to life-threatening systemic infections in immunocompromised individuals. S. aureus and C. albicans have been co-associated with a number of polymicrobial diseases including cystic fibrosis and polymicrobial sepsis. Furthermore, S. aureus and C. albicans dual-infections have been associated with increased virulence and antimicrobial resistance. Although an effort has been made to unravel the relationship between S. aureus and C. albicans, metabolomics offers a powerful analytical tool to gain a better understanding of the interactions between this bacteria and fungus. To gain a better understanding of these interactions novel methods must be developed to modulate biofilm growth. Metabolomics is intended to analyse the complete small molecule component of a biological system. Analytically, the diversity present in these compounds presents huge opportunities for improvement. The overall aim of this research was to develop novel metabolomics methods and to apply these methods to the analysis of a S. aureus/C. albicans dual species biofilm to aid in the understanding of the relationship between this bacteria and fungi. Characterisation of the S. aureus/C. albicans biofilm in comparison in to the mono-species was carried out using a number of techniques, including fluorescence microscopy, SEM imaging, qPCR and transcriptional analysis, which indicated that these two organisms interact with each other on a physical and molecular lever. Although the presence of C. albicans facilitates S. aureus biofilm formation in sera, the presence of the bacteria reduced the number of C. albicans within the dual-species biofilm compared to the fungal mono-species and caused ‘crinkled’ hyphae which suggested possible antagonistic behaviour towards the fungi. An untargeted liquid chromatography-mass spectrometry separation method was developed that effectively retained both polar and nonpolar compounds by serially coupling a reversed-phase liquid chromatography (RPLC) column to a hydrophilic interaction liquid chromatography (HILIC) column via a T-piece. Two independent pumps were incorporated into the system to allow independent gradient control of the two columns. The high dilution between the columns, achieved by the difference in flow rates, enabled the retention and separation of both polar and nonpolar standards and numerous polar and non-polar metabolites extracted from beer. Good peak shapes and retention time reproducibility was achieved across all compound classes analysed. Next, a targeted ion-chromatography mass-spectrometry method was developed for the analysis of central carbon metabolism intermediates, specifically those involved in glycolysis, the tricarboxylic acid (TCA) cycle and the Electron Transport Chain (ETC). A total mix of all of the energy metabolites standards analysed were able to be separated and detected using IC-MS, with the exception of DHAP, G3P, oxaloacetate, acetyl-CoA, succinyl-CoA, NAD and NADP. The method displayed good reproducibility and limits of detection. The complexity of the extracted biofilms proved challenging to the IC-MS. Sample variation and low intensities in some sample groups (particularly the S. aureus samples) resulted in lower detection than expected. The RPLC/HILIC method provided hundreds of metabolite detections, but suffered in comparison to the conventional HILIC method, likely due to far greater optimisation of the original technique, leading to the utilisation of the routine pHILIC method in place of the serially combined method. Untargeted metabolomics analysis highlighted significant changes in a number of metabolic pathways including purine, pyrimidine, methionine and cysteine metabolism between the S. aureus and C. albicans mono- species and the dual-species biofilms. The differences detected within individual pathways suggest a difference in behaviour when the microorganisms are cultured with one another. The dramatic downregulation of a large portion of essential metabolites within purine, pyrimidine, cysteine and methionine pathways is indicative of the bacteria struggling to proliferate and form strong biofilms in sera. Down-regulation of many of the pathways in the dual-species biofilm compared to the C. albicans mono-species biofilm suggests that the presence of S. aureus within the biofilm could be having an adverse effect on the C. albicans. The results and conclusions herein provide greater understanding of the clinically important interaction between S. aureus and C. albicans. Microscopic and molecular characterisation enabled visualisation of the dual-species biofilm. The development and application of metabolomics techniques highlighted changes in metabolism between the mono- and dual-species biofilms, indicating that the relationship between S. aureus & C. albicans may not be completely synergistic, as previously suggested. Although the metabolomics methods developed during this study performed well, with regards to the separation of simple standard mixes and the complex beer sample, were not suitable for biofilm analysis. Through continued sample preparation and chromatographic optimisation these novel methods could offer relatively simple alternatives to more time consuming and complex chromatographic procedures.

Transcription factor E2F plays in important role in growth control by co-ordinating early cell cycle events. In addition, certain E2F family members, including E2F-1, are endowed with apoptotic activity. E2F-1 is regulated during cell cycle progression and inducible by cellular stress, such as DNA damage. Within the DNA damage signalling pathway, checkpoint kinases act as effectors of the damage response through phosphorylating key substrates involved in growth control. Here, I report that checkpoint kinase Chk2 regulates E2F-1 activity in response to etoposide. A Chk2 kinase phosphorylation site resides in E2F-1, and undergoes physiological phosphorylation in response to DNA damage. Phosphorylation of E2F-1 by Chk2 leads to protein stabilization, increased half-life and transcriptional activation, and phosphorylated E2F-1 resides in discrete nuclear structures. A dominant-negative derivative of Chk2 blocks the induction of E2F-1, and prevents E2F-1-dependent apoptosis. Moreover, E2F-1 fails to be induced by etoposide in tumour cells that carry mutant chk2. Chk2 therefore phosphorylates and activates E2F-1 in the cellular response to DNA damage, and the damage induction of E2F-1 leads to apoptosis. The results suggest a role for E2F-1 in response to stress, perhaps in checkpoint control, and provide a plausible mechanistic and physiological explanation for the tumour suppressor activity of E2F-1.

Food security is the idea by which a population has enough food to sustain itself without famine. A large number of factors can influence the stability of food production, including diseases caused by microorganisms. Pectobacterium spp. is one of the leading causes of soft rot disease, resulting in crop losses both pre- and post-harvest. Bacteriocins are potent narrow spectrum protein antibiotics which target closely related bacteria to the producing strain. Ferredoxin-containing bacteriocins produced by and targeted towards Pectobacterium species have both a different domain organisation and uptake mechanism to all known Gram-negative bacteriocins. This work has shown that pectocins are able to pass through the outer membrane of Pectobacterium spp. by parasitising the ferredoxin uptake receptor, FusA. This uptake requires the pectocins to be flexible in order to pass through the lumen of the barrel and enter the periplasm. This work has shown an interaction between FusB and pectocin M1, suggesting a novel mechanism of uptake. Streptococcus agalactiae is the causative agent of disease in a wide range of hosts, ranging from human neonates to farmed Tilapia. S. agalactiae infection has a detrimental effect on the dairy industry each year as it is the leading cause of mastitis in cattle. As well as this, the prevalence of S. agalactiae in farmed fish has resulted in large numbers of infected fish and subsequently the infection of consumers. Bacteriocins produced by Gram-positive bacteria are often small modified peptides which target the peptidoglycan layer or cytoplasmic membrane of the target cell. However, a small number of protein bacteriocins produced by Gram-positive bacteria have been discovered, with the best characterised of these being lysostaphin. It has been shown that bacteriocins similar to lysostaphin are also produced by other Gram-positive bacteria, such as zoocin A produced by Streptococcus zooepidemicus. Prior to this work a novel protein bacteriocin produced by and targeted towards S. agalactiae, named agalacticin A, was discovered. It was predicted that this bacteriocin was similar in structure and function to zoocin A. This work has gone some way to structurally characterising agalacticin A, showing a two-domain structure joined by a flexible linker region allowing for the two domains to move independently. As well as this it has shown the importance of the histidine residues at the predicted active site confirming the similarities between agalacticin A and zoocin A. Together this work has gone some way to showing the potential of agalacticin A as a novel therapeutic. Altogether this work has characterised three novel bacteriocins active against pathogenic bacteria to gain a better understanding of their structure, mechanism of action and uptake.

The apicomplexan parasite Plasmodium is the causative agent of the devastating tropical disease, malaria. The World Health Organisation reported that in 2010 there were an estimated 219 million malaria cases and about 660,000 deaths. Sub-Saharan Africa is the worst affected, with 80% of malaria deaths and 90% of cases occurring in this area of the world. P. falciparum causes the most severe form of the illness and accounts for the majority of malaria deaths. Parasite resistance to antimalarials including the most effective drug, artemisinin, is becoming an increasing problem, thus research into new drug targets is vital. The pyruvate dehydrogenase complex (PDC) is one of the alpha-ketoacid dehydrogenase complexes, which are involved in energy and amino acid metabolism. The PDC catalyses the transfer of the acetyl group from pyruvate to coenzyme A (CoA) to form acetyl-CoA. The complex comprises three enzymes; pyruvate dehydrogenase (E1), dihydrolipoamide acetyltransferase (E2) and dihydrolipoamide dehydrogenase (E3). E2 has a multi-domain structure where the C-terminal catalytic domains (CD) of several E2 subunits interact to form the core of the complex. This core is an icosahedral 60-mer in mammals, plants and Gram-positive bacteria and an octahedral 24-mer in Gram-negative bacteria. The N- terminal part of E2 consists of the sub-unit binding domain (SBD) and 1 to 3 lipoyl- domains (LD) creating the ‘swinging arm’ of the PDC, which facilitates substrate channelling. E1 (a heterotetramer) and E3 (a homodimer) bind to the SBD to form the functional PDC. In humans, PDC converts pyruvate to acetyl-CoA, which leads into the citric acid cycle located in the mitochondrion. P. falciparum PDC, however, is found only in the apicoplast and produces acetyl-CoA for fatty acid biosynthesis. Plasmodium PDC has recently been shown to be important for parasite progression from the asymptomatic liver stage to the symptomatic erythrocytic stage. Thus, inhibiting PDC could prevent development of malaria. This study focuses on the role of the PDC in P. falciparum blood stages and the identification and characterisation of structural and biochemical differences between parasite and human PDC that may be ultimately exploited for drug or vaccine development. I have optimised the expression and purification of soluble recombinant mature-length P. falciparum (Pf) E2 (His-rPfE2m), truncated PfE2 consisting of the SBD and CD (His- rPfE2bc) and mature-length apicoplast PfE3 (His-rPfaE3) to obtain mg amounts of protein for biochemical and structural analyses. Each of the recombinant proteins was catalytically active. Analytical ultracentrifugation (AUC) showed that His-rPfE2m forms the typical trimer building blocks required for the large E2 core formation. Sedimentation velocity (SV) experiments showed a main species with a sedimentation coefficient of 6.6 ± 0.1 S, which corresponded to a species consistent with the PfE2 trimer size observed in sedimentation equilibrium (SE) studies. However, no 24-mer or 60-mer core species were detected in SV experiments. SE analyses did show some larger molecular mass species (> 1.5 MDa), however, whether these represent a PfE2 core complex was inconclusive due to interference by aggregated protein in the sample. My conclusion from these data is that PfE2 may form a large core structure but that this is very unstable compared with E2 multimers from other organisms, where the core complex is readily formed and maintained. Similar results were obtained from SV and SE analyses of His-rPfE2bc; only trimers are present. Small angle X-ray scattering (SAXS) was used to further analyse the trimer structure. The solution structure obtained for His-rPfE2bc revealed the linker between the SBD and the CD is extended and partially flexible. This conformation, which would allow SBD interaction with E1 and E3, is supported by cryo-electron microscopy images obtained by others for E2 from other organisms. AUC on His-rPfaE3 showed a main species with a sedimentation coefficient of 6.1 ± 0.1 S and SE analyses of the protein confirmed it to be a dimer as expected. These findings were further corroborated with the solution structure obtained using SAXS. Deletion of the PfE2 gene from P. falciparum was unsuccessful, as the correct gene locus was not targeted. This could be due to the gene being required for parasite survival in the blood stages. However, it could also be due to the difficulty of P. falciparum gene manipulation and hence, further attempts to delete PfE2 using different approaches will be required. The PfaE3 gene was successfully deleted, thus as shown in previous studies with murine malaria species, the gene is not essential in the human malaria parasite, P. falciparum. The PfaE3 deletion mutants did not show increased susceptibility to oxidative stressors compared with the wild type strain, suggesting that PfaE3 does not play a role in defence against oxidative stress. However, the mutants were less susceptible to triclosan, an inhibitor of the fatty acid biosynthesis enzyme FabI. In addition, the mutant parasites maintained synchronicity over more than four replication cycles as opposed to WT parasites, which gradually lost synchronicity over this period of in vitro culture. Further work will need to be carried out to fully characterise the role of PfaE3 in P. falciparum.

The colicin and lectin-like bacteriocins are a broad class of antimicrobial proteins produced by Gram-negative bacteria. They are generally narrow spectrum, killing or inhibiting the growth of closely related bacteria. Numerous Gram-negative bacteria that are important pathogens of both animals and plants produce and are susceptible to these bacteriocins. As such, these proteins represent an attractive alternative to traditional small molecule antibiotics for controlling bacterial infection. Very little is known about bacteriocins produced by Gram-negative plant pathogens and so the aim of this work was to discover novel bacteriocins active against globally important plant pathogens from the genera Pectobacterium and Pseudomonas. The bacteriocins discovered in this study were then structurally and functionally characterised and assessed for their ability to impart disease resistance when expressed in a model transgenic system. This study presents the discovery and characterisation of the bacteriocins syringacin M, syringacin L1 and pyocin L1 from the genus Pseudomonas, As well as the discovery and characterisation of the unusual ferredoxin containing pectocins from the genus Pectobacterium. Also presented is the discovery of a novel virulence related ferredoxin/iron-uptake system in Pectobacterium, which is parasitised by the pectocins for cell entry. Additionally, the transgenic expression of the bacteriocin putidacin L1 in both Arabidopsis thaliana and Nicotiana benthamiana was shown to provide these plants with resistance to infection by strains of the plant pathogen P. syringae.