<p><em> Chlamydia pneumoniae</em> is a Gram-negative, intracellular bacterium which utilizes a type III secretion system for virulence. This system injects virulence associated proteins into a host cell and ultimately hijacks host intracellular machinery required for the bacteria to propagate and reproduce. Chlamydia outer protein N or CopN (<em>Cpn0324</em>), is a member of a family of proteins found in pathogenic bacteria, which inhibits premature secretion of effector proteins by plugging the base of the injectisome. The lack of a genetic system to manipulate the bacteria hampers the identification of proteins within the T3S field. The work presented in this thesis establishes the role of CopN as the plug protein of <em>Chlamydia pneumoniae</em>, and examines protein interactions within the individual CopN domains.</p> <p>The structure of CopN was first explored by limited proteolysis to establish the domain boundaries. We found three domains, an N-terminal, central domain, and C-terminal domain. Next, we used the full length protein as well as a series of truncations to examine the interactions within each domain. Using a subset of the known protein interactions between CopN and other T3S proteins, we found that the proposed chaperones Scc1 and Scc4 bind in the N-terminal region. There were no apparent interactions in the central domain, whereas FliI, FliF, CopD1_158-206 and Scc3 all bound within the C-terminal region of CopN. Finally the secretion of CopN in a HeLa cell model was addressed throughout the course of an infection. CopN was detected in the host cell immediately after infection, and then was not detectable again until late infection. Overall, I have characterized the individual domains of CopN and present data to support the role of CopN as the plug protein in the T3SS of <em>Chlamydia pneumoniae</em>.</p> / Master of Science (MSc)

<p>A variety of short oligoribonucleotide sequences were synthesized using the phosphotriester chemical synthesis developed in Neilson's laboratory. These sequences were designed to incorporate a variety of features that would aid in the study of perturbations of helix structure and stability. Variable temperature proton nuclear magnetic resonance (NMR) spectroscopy was used in this study and provides a powerful technique for the study of nucleic acid conformations and for the investigation of the effects of the mispairing and single stranded regions caused by the helix imperfections introduced.</p> <p>The assignment procedure for NMR spectra was improved through the study of a series of related sequences. This study determined the effects that the addition of a nucleotide to the terminus of a sequence or the insertion of a nucleotide into the middle of a sequence would have on the chemical shifts from the rest of the sequence.</p> <p>A series of self-complementary pentaribonucleotides, with a central non-base paired opposition (AGXCU, where X ≡ A, G, C or U), was studied to determine the effects of small loops on duplex stability. In contrast to earlier results, these pentamers formed stable duplexes when X ≡ A or C, although the duplex Tm's were significantly, reduced. These sequences also provided the opportunity to study the sequence effects of adjacent internal G.C base pairs on duplex stability when the middle base pair was A·U, G·C or G.U. Comparison with earlier results using corresponding A·U base pair neighbours, demonstrated the enhanced stabilization of G·C base pairs.</p> <p>The effects of terminal non-base paired (dangling) adenosines were more closely investigated and found to contribute an average of 11°c to the thermal stability of the duplex formed. This study also demonstrated that 5'-dangling adenosines contribute less to overall stability than do 3'-dangling adenosines. However, this effect did display some sequence dependence.</p> <p>The triribonucleotide GpCpA was the first trimer shown to form a stable RNA duplex (Tm = 33°C). The duplex consisted of two G·C base pairs and two 3'-dangling adenosines and had a stability equal to that of the tetramer duplex UpGpCpA with four Watson-Crick base pairs.</p> <p>The influence of base stacking on duplex formation was studied and it was discovered that the direction of base stacking had a definite influence on helix stability. Stacking in the 5'→3' direction was more favourable to duplex formation than stacking in the 3'→5' direction.</p> <p>Lastly, the significance of invariant adenosines at position 14 and 21, in the D-stem of tRNA, was investigated through a model study that suggested the adenosines contributed to D-stem stability.</p> <p>Most of the results presented in this thesis have been published or accepted for publication in:</p> <p>1. Jeremy R. Everett, Donald W. Hughes, Russell A. Bell, Dirk Alkema, Thomas Neilson and Paul J. Romaniuk. "Nearest-Neighbour and Next-Nearest-Neighbour Effects in the Proton NMR Spectra of the Oligoribonucleotides ApGpX and CpApX." (1980) Biopolymers 19, 557.</p> <p>2. Thomas Neilson, Paul J. Romaniuk, Dirk Alkema, Donald W. Hughes, Jeremy R. Everett and Russell A. Bell. "The Effects of Base Sequence on the Stability of Short Ribonucleic Acid Duplexes." (1980) Nucleic Acids Res. Sym. Series No. 7, 293.</p> <p>3. Dirk Alkema, R.A. Bell, P.A. Hader and T. Neilson. "Triplet GpCpA Forms a Stable RNA Duplex." (1981) J. Am. Chem. Soc. 103, 2866.</p> <p>4. Russell A. Bell, Jeremy R. Everett, Donald W. Hughes, Dirk Alkema, Paul Hader, Thomas Neilson and Paul J. Romaniuk. "Nearest-Neighbour and Next-Nearest-Neighbour Effects in the Proton NMR Spectra of the Oligoribonucleotides ApXpG, CpXpG, CpApXpUpG, ApGpXpC and ApGpXpCpU." (1981) Biopolymers 20, 1383.</p> <p>5. Dirk Alkema, Russell A. Bell, Paul A. Hader and Thomas Neilson. "Short RNA duplex Stability: Contribution from non-base paired residues to the direction of stacking." (1981) in "Biomolecular Stereodynamics" R.H. Sarma, ed., Adenine Press, Elmsford, NY., p. 417.</p> <p>6. Dirk Alkema, Russell A. Bell, Paul A. Hader and Thomas Neilson. "Invariant Adenosine Residues Stabilize tRNA D-steins." (1982) Accepted for publication in FEBS Letters.</p> <p>7. Dirk Alkema, Paul A. Hader, Russell A. Bell and Thomas Neilson. "Effects of Flanking G·U· Base Pairs on Internal Watson-Crick, G·U and Non-Bonded Base Pairs Within a Short RNA Duplex." (1982) Accepted for publication in-Biochemistry.</p> <p>Two additional publications will appear shortly:</p> <p>1. Paul A. Hader, Thomas Neilson, Dirk Alkema, Eric C. Kofoid and M.C. Ganoza. "Sequencing of Short RNA Oligomers by Proton Nuclear Magnetic Resonance." (1982) Accepted for publication in FEBS Letters.</p> <p>2. Paul A. Hader, Dirk Alkema, Russell A. Bell and Thomas Neilson. "Parameters for Proton Chemical Shift Prediction in Oligoribonucleotides." (1982) J. Chem. Comm. (in press).</p> / Doctor of Philosophy (PhD)

Neuropeptides regulate diverse physiological processes, including homeostatic metabolism, behavior, reproduction, and development. The neuropeptide Corazonin (Crz), was first isolated from American cockroach, P. americana, as a potent cardioactive substance, and has been shown to exert diverse functions in different insects. In Drosophila, Crz expression is limited to three groups of neurons; totaling only 26 neurons out of ~10,000 neurons in a third instar larval central nervous system (CNS). In adults, Crz is expressed in 6-8 pairs of protocerebral neurons and 2 pairs of male specific abdominal ganglion. To gain insight into such tight regulatory mechanisms of Crz gene transcription, Crz promoter activity was dissected in vivo. The promoter bashing experiments yielding various 5‘-upstream sequences show that there are separate cis-acting elements that are highly conserved phylogenetically, which speaks to its functional significance in the activation of Crz transcription. In larval stage, a 504-bp upstream region is sufficient to activate Crz in all endogenous neurons. Further dissection revealed two important regions; one between -419-bp and -504-bp region for the expression in dorsal medial neuron (DM). The other located between - 241-bp and -380-bp is responsible for dorsal lateral (DL) and ventral nerve cord (VNC) expression. The latter region can be subdivided into three DL-specific and two VNCspecific cis-acting elements. For DL-specific expression, two out of any three combination were needed; however, VNC needed two elements altogether. Interestingly, basal transcription factor binding site TATA box showed minor role for Crz expression. In contrast to the larval expression, 321-bp upstream region is sufficient to activate Crz in all adult neurons. For the male-specific abdominal ganglion (ms-aCrz) expression, the cis-acting element was found to be in a region between -250-bp and - 290-bp. Overall, the data show that transcriptional regulatory mechanisms for Crz expression are not uniformed among Crz-containing neurons, which further indicates that their neuronal functions might be different. To identify the roles of Crz in Drosophila, several fly behaviors were tested; ethanol-related responses, olfactory sensing responses and circadian rhythmic behaviors. Crz cell deficient (Crz-CD) flies and Crz receptor knock down (CrzR-KD) flies showed significantly delayed recovery from ethanol-induced sedation compared to control flies. Such hangover phenotype was ethanol specific. This result suggests that Drosophila Crz involves in ethanol-related responses. Further analyses suggest that Crz-CD, CrzR-KD and CrzR mutation did not affect aldehyde dehydrogenase (ALDH) at transcription level, but reduced ALDH enzyme activity. Crz is also associated with olfactory signaling, as Crz-CD and CrzR-KD flies are unable to find odor source, such as live yeast paste. Previously, Crz neurons located in the vicinity of nerve terminals originated from circadian pacemaker Pdf-expressing neurons, which indicate Crz neurons as part of circadian circuit. However, circadian locomotor rhythmic behavior of Crz-CD, Crz over-expression and CrzR-KD flies show normal circadian rhythmic behavior.

Pregnancy is dependent on the development of maternal immune tolerance to the genetically foreign fetus. During pregnancy the mother's immune reactivity and energy metabolism undergoes significant changes and the levels of certain hormones in peripheral blood are significantly increased. Hormones are important regulators of the functional activity of the immune system and immune cells within. Hormones secreted by the placenta, protect the fetus from the maternal immune response of the mother, emphasizing their immunomodulatory effects. Therefore, hormonal regulation is essential for the functional activity of immune cells. There is evidence that the hormone, kisspeptin, plays a role in the development of immune tolerance during pregnancy based on its role in the regulation of the adaptive T regulatory (aTreg)/T-helper 17 (Th17) cells, induction of the enzyme indoleamine 2,3-dioxygenase (IDO) and regulation of monocyte function during pregnancy. In addition, kisspeptin has been implicated in the regulation of specific cytokines during pregnancy. It is crucial to maintain an appropriate cytokine balance at the maternal– fetal interface as well as in circulation. Several pregnancy-related disorders have been associated with a variation in Th1/Th2/Th17 cytokines and aTreg cell subsets. Kisspeptin has been implicated in regulating cytokines IL-10 and IL-17A as well as aTreg and Th17 cells which are significant role players in immune tolerance during pregnancy. However, its effect on other pro- and anti-inflammatory cytokines remain unknown. Therefore, more research is required to better understand the role of kisspeptin in the development of immune tolerance during pregnancy. The hypothesis of this study is that kisspeptin alters the expression of anti-and pro-inflammatory cytokines and may thus influence the establishment of immune tolerance in pregnancy. To test this hypothesis, we used a previously established in vitro peripheral blood mononuclear cell (PBMC) Mycobacterium tuberculosis (Mtb) infection assay model as well as a newly established in vitro infection model using lipopolysaccharide (LPS)-stimulated whole blood. Protein expression analysis of selected pro- and anti-inflammatory cytokines was performed on PBMC infected with Mtb and on whole blood cells stimulated with LPS in the absence and presence of kisspeptin-10 for different times. The cytokines levels were measured by luminex multiplex assay and sandwich ELISA, respectively. Results from the PBMC infection assay showed a varied but not statistically significant effect of kisspeptin-10 on selected pro- and anti-inflammatory cytokine expression at 2 hours post-infection. However, there was a suggestion of an inhibitory effect of kisspeptin-10 on selected pro- and anti-inflammatory cytokine expression, macrophage inflammatory protein (MIP)-1α, MIP-1β, tumour necrosis factor (TNF)-α, granulocyte macrophage colony stimulating factor (GM-CSF) and interleukin (IL)-10, after 24 hours which was not observed at 6 days post-infection. Results from the whole blood stimulation assay suggested an inhibitory effect of kisspeptin-10 on selected LPS-induced pro-inflammatory cytokines (IL-1β, TNF-α, IL-6) whilst generally not having an effect on selected anti-inflammatory cytokines (IL-10). Overall this study suggests, based on the lack of statistically significant data, a potential immunomodulatory effect of kisspeptin-10 based on the observed inhibition of pro-inflammatory cytokines. Investigating and developing an understanding of key regulators and mechanisms of maternal immune tolerance may help researchers understand the pathophysiological mechanisms underlying certain pregnancy-related disorders. This was a pilot study aimed at characterising the effect of kisspeptin stimulation on cytokines and chemokines responses. Manipulation of regulatory hormones such as kisspeptin could represent a potentially novel approach in the treatment of various pregnancy-related disorders including preeclampsia and unexplained recurrent miscarriage.

Includes abstract. / Includes bibliographical references. / In South Africa, resource-poor farmers mainly depend on livestock farming for their livelihoods, with cattle production being the most important livestock sector. As a consequence of natural selection in stressful conditions, Nguni cattle have been reported to be metabolically superior to other cattle breeds under unfavourable conditions. Using proteomics, with mass spectrometry at the core of the analysis, the objective of this study was to establish a reliable set of methods for the protein profiling of Nguni cattle livers. To achieve this several alternative technologies were employed and their outcomes compared namely, two-dimensional electrophoresis, fractionation by solution phase iso-electric focusing-reversed phase chromatography (IEF-RP), offline strong cation exchange- low pH reversed phase chromatography (SCX-RP) and offline high pH reverse phase-low pH reverse phase chromatography (RP-RP). All solution based methods were coupled to a tandem mass spectrometer. Protein identification was performed using the ParagonTMAlgorithm of Protein Pilot v4.0 as well as PEAKS v6. The IEF-RP and RP-RP methods achieved similar results in terms of number of proteins identified. In addition, proteins that play a role in the urea cycle (which is believed to contribute to the Nguni cattle’s enhanced metabolic ability) were all identified with both techniques. The RP-RP method was selected as the most appropriate method for future research linked to this work and will be used in the next phase of this project, on the basis that it is easier to automate compared to the IEF-RP method. It will be used beyond the scope of this work to compare levels of expression and modification of the liver proteins and their isoforms in Nguni and Hereford cattle grown under adverse environmental conditions, in order to identify those that may contribute to enhanced liver metabolism in Nguni cattle. This will be complemented by the identification and characterisation of potential polymorphisms with in such proteins that can be used to select for this trait during breeding.

Includes bibliographical references (leaves 100-110). / AP-I is a dimeric transcription factor comprised primarily of Jun and Fos family proteins, that regulates numerous genes involved in cell proliferation, differentiation and oncogenesis. The expression of AP-I is shown to play an important role in many human cancers and plays a key role in the regulation of the E6 and E7 oncoproteins of high-risk Human Papillomaviruses (HPV) that are etiologically associated with cervical cancer. The c-Jun and Jun B components of AP-I were shown to be expressed at higher levels in cervical cancer patients compared to nonnal patient tissue while Jun D levels were largely unchanged. To define the role of AP-I in cervical cancer, the effect of inhibiting AP-I actvity was determined using a dominantnegative deletion mutant T AM67. CaSki cervical cancer cells with a doxycycline inducible T AM67 demonstrated that inhibition of AP-I activity and expression resulted in an altered cell morphology, a significant decrease in cell proliferation and inhibition of colony formation. This was accompanied by a slower progression of T AM67 expressing cells through the cell cycle, with an accompanying increase in G21M phase. An increase in the expression of the cell cycle regulatory protein, p21 CIPI, was observed that appeared independent of p53 expression. siRNA directed at inhibiting individual AP-I components showed that Jun B was an important regulator of CaSki cell proliferation. These results suggest that AP-I is involved in the cell proliferation and tumourigenic phenotype of cervical cancer cells, such as CaSki cells, possibly via a direct repression of cell cycle regulator p21 CIP1

Bibliography: leaves 240-266.

Includes abstract. / Includes bibliographical references (leaves 125-143). / Oesophageal squamous cell carcinoma presents a significant health burden in South Africa. It is one of the most common causes of cancer-related mortality of South African black males, as a result of its asymptomatic progression leading to late diagnosis and poor prognosis. The aim of this study was to identify membrane or membrane-associated proteins that are expressed at different levels in oesophageal tumour tissue when compared to normal tissue. The identification of such proteins would be an important step towards the development of better diagnostic and therapeutic strategies for this disease. Two proteomic approaches, were employed to identify differentially expressed proteins.

The advent of gene expression profiling studies has allowed for the identification of genes with potential as disease markers and therapeutic targets. Our laboratory identified the eukaryotic nuclear importer protein Karyopherin B1 (KpnB1), to be up-regulated in different cancer cell lines, including cervical and oesophageal as well as transformed cells. Inhibition of KpnB1 in these cells using small interfering RNA (siRNA) resulted in significant cancer cell death via apoptosis, suggesting KpnB1 is essential for cancer cell survival. Within our laboratory, we established that candidate small molecules targeted against KpnB1 identified using a rational drug design approach. The outcome of this research is for examine inhibitors of KpnB1 for potential as future anti-cancer agents using. Based on the long-term goal of this research, this particular project was aimed at investigating a small molecule inhibitor identified in our laboratory, known as Inhibitor of Nuclear Import-43 (INI-43) for its potential to bind to the nuclear importer, KpnB1. Using conventional assays as well as cutting edged techniques including circular dichrosim (CD) and isothermal titration calorimeter (ITC), an examination of INI-43 and its interactions with KpnB1 was made. In vitro analysis showed that INI-43 exhibits cytotoxic effects on cervical cancer cells with an IC₅₀ of ≈10μM and induces apoptotic cell death. The NFAT dual luciferase assay measured nuclear import of KpnB1 associated proteins, showing that INI-43 inhibits nuclear import/activity of NFAT in a dose dependent manner. Confocal microscopy of exogenous FRFP-KpnB1 as well as endogenous KpnB1 in the presence of INI-43 showed a change in the localisation of KpnB1 upon drug treatment. Both FRFP-KpnB1 and endogenous KpnB1 appear to be prevented from entering the nucleus, and is retained in the peri-nuclear space and the cytoplasm suggesting that INI-43 inhibits KpnB1 movement into the nucleus. To investigate KpnB1-INI-43 interactions, purified KpnB1 was prepared and used in biophysical techniques. Purified KpnB1 protein was prepared using GST-tagged purification methods and the tagged protein confirmed by mass spectrometry. Purified GST-KpnB1 was used in drug binding studies including circular dichrosim (CD) isothermal titration calorimetry (ITC). CD showed a drug concentration dependant shift in the spectra at around 233nm, indicative of drug protein interaction possibly occurring in a region of KpnB1 containing aromatic amino acids. The purified GST-KpnB1 was used in ITC, which confirmed an interaction between KpnB1 and INI-43, a relatively weak interaction. In conclusion, our data shows that the small molecule, INI-43 kills cancer cells, likely by interfering with KpnB1 associated nuclear import pathways. We show that INI-43 interferes with the nuclear localisation of KpnB1 itself and biophysical assays provide evidence for possible KpnB1-INI-34 interactions. Small molecules such as INI-43 present as promising tools to studying the potential of KpnB1 as an anticancer target.

The mechanisms of thermal stability have been a long studied subject for many years with the aim of enhancing thermal stability of protein molecules to enhance their application in industry. The nitrile hydratases group of enzymes catalyse the hydrolysis of nitriles to amides using an exothermic catalytic mechanism. Understanding and applying specific amino acid residue mutations at specific regions in protein structures has been important for engineering of thermal stability into these often tetrameric thermolabile nitrile hydratases currently used in industry globally. At the near atomic level, the interatomic interaction(s) between specific amino acid residues governs the structure and function of nitrile hydratases. This study investigated several possible interactions responsible for conferring thermal stability to several thermostability-enhanced nitrile hydratase composite mutants generated from the wild type Geobacillus pallidus RAPc8 nitrile hydratase (NHase), namely: L103S+Y127N+F36L+D4G, M43K+T150A+S169R and D96E+D167V+M188V each labelled as 9E, 9C and 8C respectively. The composite mutants were previously developed using error-prone PCR of the wild type nitrile hydratase genes coding for the alpha and beta subunits from Geobacillus pallidus RAPc8. These composite mutants presented an opportunity to understand intramolecular thermostabilising mechanisms in this nitrile hydratase. Each individual mutation found in the composite mutants, was separately introduced into the DNA coding for the Geobacillus pallidus RAPc8 NHase by site directed mutagenesis. These individual mutants were over-expressed from E. coli and purified for further study. Using activity assays and protein melting curves, their individual thermal stability contributions were determined and represented as the difference in free energy of thermal unfolding (change in Gibbs free energy) of the single and composite mutants relative to the wild type nitrile hydratase. The measured residual activity following thermal inactivation was used together with the Arrhenius equation and a three parameter non-linear fit to determine the free energy of thermal unfolding. The change in Gibbs free energy resulting from each thermostabilising mechanism coupled to the analysis of their crystal structures was used to suggest the contributing mechanisms. This study found that intersubunit interactions through hydrogen bonds and salt bridges are especially important for contributing towards thermal stability of tetrameric nitrile hydratases. Hydrophobic interaction through the formation of a water shell around hydrophobic side-chains and packing of hydrophobic side-chains was also observed to contribute to thermal stability. These results suggest a path towards rational design and engineering of thermostabilising mutations into nitrile hydratases. Increased thermostability would improve their large scale application in industry by allowing these enzymes to be more active for longer at higher temperatures and decrease the cost of amide production.