Antimicrobials have been widely using as a major resource to treat bacterial infections for almost a century. However, it is not unusual to see antimicrobial resistance emerges in a bacterial species due to natural selection under the usage of antimicrobials. Moreover, numerous studies show that bacteria can accumulate genes encoding resistance to different classes of antimicrobials and share them with other bacteria regardless of ancestry via a biological process called horizontal gene transfer, causing emergence and fast transmission of multidrug resistance. As such, antimicrobial resistance becomes an urgent and global threat to public health, pushing us backwards to the pre-antimicrobial era. In this thesis, I focus on horizontal co-transfer of resistance genes between bacteria of the same species, which is usually caused by co-localisation of resistance genes in mobile genetic elements, also known as physical linkage between these genes. This kind of linkage plays a pivotal role in the evolution of multidrug resistance, because the mobile elements can translocate, recombine and aggregate, rapidly rendering their host bacteria resistant to a wide spectrum of antimicrobials. By far there is nonetheless not an approach identifying horizontally co-transferred genes in a single bacterial species. Yet most authors of literature reported a few co-mobilised resistance genes each time following biological experiments, and some researchers only applied simple association analysis to representative bacterial isolates of distinct species so as to minimise the possibility that a specific combination of genes is inherited from their most-recent common ancestor. In contrast, intra-species association analysis is severely confounded by strong sample relatedness because of bacterial clonal reproduction. This obstacle leaves a gap between the known high frequency of intra-species horizontal gene transfer and our understandings of this process. This thesis presents a scalable computational approach that uses whole-genome sequencing data to identify co-transferred antimicrobial resistance genes in bacteria collected in a few decades from the same species. Moreover, it demonstrates applications of the approach to three clinically important pathogens and reports key players, patterns and dynamics underlying the horizontal co-transfer of resistance genes within each species. In the first chapter, I provide a background to antimicrobial resistance, horizontal gene transfer, whole-genome sequencing and contemporary bioinformatic techniques. I also summarise outcomes of horizontal gene co-transfer for characteristics that we can utilise for inference of physical linkage. Finally, I compare several statistical approaches determining pairwise association between presence-absence status of genes or alleles in bacteria to justify the necessity of controlling for sample relatedness in association analysis. For the second chapter, I derived a methodology inferring co-transferred genes by integrating gene detection, de novo genome assembly, core-genome and phylogenetic analysis, linear mixed models, hypothesis tests for effects of sample relatedness and evaluation of consistency in pairwise physical distances between resistance alleles in bacterial genomes. This methodology is designed to overcome limitations of existing methods summarised in the first chapter. Moreover, I show interpretations of expected outcomes and discuss constraints of this approach. The next three chapters present an implementation of my methodology and its applications on antimicrobial resistance genes in three clinically important species of Enterobacteriaceae. First, I conducted an empirical study following a simulation-and-validation strategy on finished-grade full genomes of six strains of Klebsiella pneumoniae to find out an optimal method that measures the pairwise physical distances between alleles in de novo genome assemblies. I found that for each assembly graph, the most accurate measurements are obtained via setting up constraints for both the number of nodes in the graph and the maximum of distance measurements. Second, I developed GeneMates, a computational and integrative software package that implements my methods proposed in the second chapter for the identification of physically linked resistance alleles or for analysing associations between a large number of resistance alleles when controlling for individual relatedness. In particular, GeneMates leverages network topology to identify potential physical linkage between the alleles. For validation, I applied this tool to whole-genome sequencing data of Escherichia coli and Salmonella Typhimurium, whose acquired resistance genes and relevant mobile genetic elements have been well characterised in publications. In result sections, I illustrate clusters of physically linked resistance alleles and discoveries of their vectors. For the last result chapter, I applied GeneMates to genomes of a large global collection of K.pneumoniae strains, which are adept to uptake DNA from various environments. Furthermore, I compared structure and contents of co-localised allele clusters across time and geography and discovered patterns underlying the evolution of multidrug resistance in this species. To conclude, I have developed and implemented a network approach that performs association tests on presence-absence of resistance alleles in a large collection of bacterial isolates of the same species and infers potential horizontally co-transferred alleles. I have validated this approach using known co-mobilisable resistance genes and the approach showed higher statistical power than existing methods. The GeneMates package will become a powerful tool contributing to routine surveillance of antimicrobial resistance and identifications of known and novel mobile genetic elements. In addition, applications of this package to other kinds of bacterial genes is also feasible and convenient.

The central nervous system (CNS) constantly responds to changes in environmental stimuli by undergoing structural and functional modifications. Some stimuli induce persistent CNS changes which in turn underpin adaptive behaviours that enable individual animals to function in their unique environmental circumstances. This phenomenon, referred to as neuroplasticity, has been studied predominantly with respect to adaptive neuronal changes, and has focused primarily on synaptic changes and the molecular transduction mechanisms that mediate them. It is increasingly recognised, however, that glial cells can also be modified by external stimuli. Oligodendrocytes – the myelinating glia of the CNS which facilitate efficient nerve impulse conduction and support axonal metabolism – have also been demonstrated to undergo long term changes in response to environmental stimuli. Experience-dependent changes in oligodendrocyte number or myelination could underpin adaptive behaviours via modifications to neuronal metabolism and nerve impulse conduction. The emerging consensus is that stimulation – whether indirectly through modulating sensory, motor, or social experience, or directly through modulating neuronal activity – increases oligodendroglial lineage progression and myelin production. It has further been demonstrated that myelin plasticity is an axon-specific phenomenon whereby, when given the choice, myelin segments preferentially form on axons that are more highly active relative to those that are nearby but less active. The molecular mechanisms that mediate myelin plasticity are not well understood, and studies addressing this question have predominantly focused on the role of extracellular, pro-myelinating signals released by neurons in an activity-dependent manner. Comparatively little is known about the oligodendroglial intrinsic molecular transduction mechanisms that mediate myelin plasticity. This thesis aimed to develop a model system for studying myelin plasticity, including in particular to investigate the molecular transduction mechanisms that are triggered within oligodendroglia to mediate myelin plasticity. In developing such a model, two approaches were employed. First, an in vitro myelinating co-culture model was developed. A standard co-culture protocol was adopted and refined to produce robustly myelinating co-cultures of dorsal root ganglion (DRG) neurons and oligodendrocyte precursor cells (OPCs). To stimulate neuronal activity, both the hM3Dq pharmacogenetic and the channelrhodopsin-2 (ChR2) optogenetic techniques were explored. The pharmacogenetic stimulation technique was ineffective at driving DRG neurons to the levels of activity reportedly required for inducing myelin plasticity. In contrast, the optogenetic stimulation technique reliably drove DRG neurons to fire at the required frequency. Contrary to expectations, optogenetic stimulation did not increase myelin production in co-cultures, nor did it increase the propensity of myelin segments to preferentially form on optogenetically stimulated relative to control axons. The reasons for this are unclear, but are unlikely to be related to phototoxicity and are more likely to be explained by a negative effect of high ChR2 expression on myelination in these co-cultures. Second, an in vivo pharmacogenetic model was employed to drive activity of cortical neurons in juvenile hM3Dq transgenic mice. Contrary to expectations, there was no evidence for an activity-dependent increase in oligodendroglial lineage progression. The reasons for this are unclear, however they could relate to the young age of the animals in this relative to other studies of myelin plasticity or to the large population of neurons undergoing activity manipulation in this relative to other studies of myelin plasticity. The implications for glial plasticity, and for how it is studied, are discussed.

Excitotoxicity is a primary pathological process directing neuronal cell death in both acute neurological disorders and neurodegenerative diseases such as ischemic stroke and Alzheimer’s disease. It is initiated by over-stimulation of ionotropic glutamate receptors (iGluRs), which in turn permit the influx of excessive calcium ions into the cytosol of the affected neurons. The sustained and excessively high cytosolic calcium concentration induces substantial imbalance of the intracellular environment such as depletion of ATP and over-production of reactive oxygen species by inducing dysregulation of specific proteases, protein kinases, and phosphatases. These dysregulated enzymes in turn aberrantly catalyse post-translational modifications of specific neuronal proteins to cause neuronal death. Despite the efforts by many investigators in the past four decades, the cell signalling pathways relaying the neurotoxic signals from the over-stimulated iGluRs and the excessively high cytosolic calcium concentration to direct neuronal death in excitotoxicity, remains unclear. Avenues to chart these neurotoxic signalling pathways include identifying the excitotoxicity-dysregulated proteases, protein kinases, and phosphatases, and delineating the mechanisms of their dysregulation and elucidating their exact roles in excitotoxic neuronal death. In my PhD study, I employed multiple quantitative proteomic approaches and a combination of the neuronal cell-based model and two animal models of excitotoxicity to identify the neuronal proteins aberrantly modified by these excitotoxicity-dysregulated enzymes. Furthermore, these approaches also defined the exact sites of modifications and quantified the degree of modifications. The proteomic approaches I adopted include the quantitative N-terminomic, global and phosphoproteomic approaches. They were used to analyse changes in general protein degradation, limited proteolysis, phosphorylation and dephosphorylation of specific neuronal proteins catalysed by the excitotoxicity dysregulated enzymes in cultured primary cortical neurons and in brain tissues subjected to ischemic stroke and traumatic brain injury. Using an N-terminomics method called “Terminal amine isotopic labelling of substrates” (TAILS), I discovered hundreds of cellular proteins undergoing enhanced proteolytic processing catalysed by the excitotoxicity-activated proteases to form truncated protein fragments in neurons over-stimulated with glutamate. To define the identities of these excitotoxicity-associated proteases, I performed TAILS analysis of neurons co-treated with the excitotoxic level of glutamate and calpeptin, a specific inhibitor of calpains and cathepsins. Results of my analysis revealed that calpeptin abolished enhanced proteolytic processing of the majority of the neuronal proteins identified in the TAILS study of neurons treated with glutamate-only. These findings indicate that most of these identified neuronal proteins were either direct calpain/cathepsin substrates or the substrates of proteases activated by calpains or cathepsins. More importantly, the findings suggest that calpains and cathepsins are the major modulator proteases catalysing proteolytic processing of specific neuronal proteins in excitotoxicity. Using label-free global and phosphoproteomics analyses I found 483 phosphopeptides derived from neuronal proteins undergoing significant changes in phosphorylation level in glutamate-treated primary cortical neurons. Interestingly, global proteomic analysis revealed only a few neuronal proteins exhibiting significant changes in abundance, suggesting that most of the neuronal proteins remained stable up to 240 min of glutamate treatment. These findings suggest that instead of protein synthesis and degradation, phosphorylation plays a major role in modulating the activities and functions of specific neuronal proteins in excitotoxicity. From the identified phosphosites in the neuronal proteins undergoing significant changes in phosphorylation state, bioinformatic analysis predicted PAK1 (Serine/threonine-protein kinase), CK2A1 (Casein kinase II subunit alpha), and mTOR (mammalian target of rapamycin) as the most perturbed kinases in neurons undergoing excitotoxic cell death. Bioinformatic analysis of the neuronal proteins showing significantly changed phosphorylation and/or enhanced proteolytic processing predicted defective axonal guidance signalling as the most perturbed signalling pathway in excitotoxicity. To complement the proteomic analysis in cultured neurons, I conducted phosphoproteomic studies of thebrains in mice suffering from stroke and TBI (traumatic brain injury). Results of my study indicate that neuronal proteins, especially those in synapses are mostly dysregulated. Brain tissue lysates contain proteins expressed in neurons, astrocytes, microglia, oligodendrocytes, and endothelial cells. The high abundance of neuronal proteins identified as the brain proteins exhibiting significant changes in phosphorylation state induced by stroke and TBI suggest that neurons are particularly sensitive to the detrimental impacts of stroke and TBI. The neuronal protein tyrosine kinase Src was discovered by TAILS to be cleaved by calpains to form a truncated Src fragment in excitotoxicity. To examine the therapeutic potential of my proteomic findings, I chose to investigate if blockade of calpain cleavage of Src in neurons could protect against neuronal loss in vivo in a rat model of neurotoxicity. My study showed that stereotaxic injection of a cell membrane-permeable peptide inhibitor of calpain cleavage of Src in rats can protect against neuronal loss induced by over-stimulation of the N-methyl-D-aspartate (NMDA) receptors. Taken together, results of my proteomic analyses have built a conceptual framework for future investigation to decipher the signalling pathways underpinning neuronal death in excitotoxicity. Furthermore, my results provided evidence of the therapeutic potential of the findings from TAILS analysis. Some of the neuronal proteins identified in my study to exhibit significantly perturbed phosphorylation state and/or proteolytic processing, are potential targets for the development of new therapeutic strategies to reduce neuronal loss in ischemic stroke, traumatic brain injury and other neurodegenerative diseases.

Early postnatal life is a critical stage of microbiota establishment and ENS development. While the initial postnatal stage from birth is fundamental for the development of the gut microbiota and ENS, weaning is another key developmental period where there are major changes in diet, behaviour and physiology, and notably, microbiota. Antibiotics are frequently administered to infants and young children, however, recent studies have identified prospective long-term health consequences of early life antibiotic exposure on the developing gut microbiota. Yet, how antibiotic influences short and long-term ENS development remains unclear. Vancomycin is given as a prophylactic to preterm babies and paediatric patients to treat and prevent infections. It is also one of the most commonly used antibiotics on its own or as part of a cocktail in research to induce dysbiosis in mice. The aim of my PhD was to examine how early life exposure to vancomycin during two critical developmental periods affects microbiota and ENS development and whether changes observed during early postnatal life have long-term repercussions. In chapter 3, I investigated if acute administration of vancomycin, during the early postnatal period, influenced gut microbiota and ENS development. A single regimented dose of either water or vancomycin was administered daily to Wnt1-Cre;R26R-GCaMP3 mouse pups from postnatal (P) day 0 to P10/11. These mice contain a genetically-encoded fluorescent Ca2+ indicator in all enteric neurons and glia. At P10/11, vancomycin-fed pups showed significant dysbiosis, reduced myenteric neuron density and altered nNOS and calbindin neuronal subtype proportions compared to water-fed littermates. Using Ca2+ imaging, I showed that vancomycin-fed pups had more neurons responding to electrical stimulation applied to interganglionic connectives and larger amplitudes of train-evoked [Ca2+]i transients. These changes in the ENS contributed to dysmotility of the colon of vancomycin-fed pups. In contrast to the colon, the structure of the ENS and motility patterns of the duodenum were not affected by vancomycin, ruling out drug toxicity effects. P10/11 vancomycin-fed pups also had lower numbers of serotonin (5-HT) positive cells in the colonic mucosa. Altered 5-HT metabolism in these animals were confirmed by performing mass spectrometry on 5-HT biosynthesis intermediates, showing reduced concentrations of the 5-HT metabolite, 5-HIAA and droplet digital PCR (ddPCR) revealing increased gene expression of the 5-HT transporter, SERT. Bypassing tryptophan hydroxylase, by supplementing vancomycin-fed pups with 5-HTP, restored 5-HIAA levels in the colonic mucosa and prevented some of the vancomycin-induced effects on myenteric neurons, colonic motility and gut microbiota. Therefore, vancomycin exposure during the neonatal period induced significant developmental changes to both the gut microbiota and ENS. Some of these changes could be mediated by altered mucosal serotonergic signalling. In Chapter 4, I examined if vancomycin-induced changes on the gut microbiota and ENS observed at P10 were long-lasting. Newborn mouse pups were only treated with water and vancomycin till P10, then pups were left to grow to adulthood. 6-week-old mice given neonatal vancomycin had enlarged caeca, which is an indication of dysbiosis. This suggests that the gut microbiota of vancomycin-fed mice was not fully recovered despite cessation of antibiotic treatment. Adult mice treated with neonatal vancomycin had sustained reduction in myenteric neuron density. However, alterations in the proportions of nNOS+ and calbindin+ neurons observed during the neonatal periods was now restored. In contrast to the heightened [Ca2+]i activity at P10s, adult mice given neonatal vancomycin had lower numbers of neurons responding to electrical stimulation and no change in the amplitudes of electrically-evoked [Ca2+]i transients in their myenteric neurons compared to water-fed controls. Furthermore, there were no treatment-induced changes in colonic motility. Interestingly, faecal water content, which was unaffected in vancomycin-fed pups at P10, was lower in adult mice given neonatal vancomycin compared to controls. These findings indicate that although vancomycin treatment is terminated, the gut microbiota is not fully recovered and significant re-modelling of the ENS occurs, some of which are distinct to changes observed during the neonatal period. In Chapter 5, I explored the effects of vancomycin exposure between weaning and adulthood. From the day of weaning, mice were administered vancomycin or sterile water in their drinking bottles for three weeks. At 6-weeks of age, vancomycin-treated mice had dysbiosis accompanied with enlarged caeca. Similar to vancomycin-treated neonates in Chapter 3, increased synaptic activity exhibited by enteric neurons were mainly observed by larger amplitudes of train-evoked [Ca2+]i transients and increased number of neurons responding to electrical stimulation. However, in contrast to antibiotic exposure during the neonatal period, vancomycin-treated mice displayed significantly slower colonic motility, increased faecal water content and a decrease in the proportions of ChAT+ cholinergic neurons including calbindin and neurofilament-M subtypes in the myenteric plexus of the colon. Moreover, vancomycin treatment between weaning and adulthood had no effects on the serotonergic system in the colonic mucosa. Collectively, these findings suggest that vancomycin exposure from weaning had differential effects on the gut microbiota and ENS compared to administration of the antibiotic during the neonatal period. Together, my study is the first to identify and compare effects of antibiotic exposure on the gut microbiota and ENS during two critical stages of development. While vancomycin did not deplete bacterial diversity and abundance, it caused profound shifts in microbial composition in both developmental periods. Additionally, acute vancomycin exposure in both periods, resulted in dysmotility and alterations of the neuronal circuitry. Although the effects on colonic motility for mice given neonatal antibiotic treatment did not appear to be long-lasting, changes in the ENS and disrupted faecal and caeca weights, which manifested only in adulthood, suggests that early life exposure to antibiotics can have other long-term consequences on microbiota and host gut physiology.

Virtually all functions of the enteric nervous system (ENS) rely on synaptic transmission, which occurs at specialised sites referred to as synapses. Molecular mechanisms behind synaptic transmission at central synapses have been extensively characterised, studies accordingly show that pre- and post-synaptic proteins localized to these synapses regulate transmission. However, little is known about the synaptic machinery involved in regulating excitatory transmission at enteric synapses. There is growing evidence to suggest that patients with synaptic protein associated neurodevelopmental and neurodegenerative diseases, such as Parkinson’s Disease, also display abnormalities in gastrointestinal function. This suggests there is a commonality between the central nervous system (CNS) and the ENS. Excitatory transmission within the ENS is primarily mediated by acetylcholine (ACh) acting on nicotinic receptors, there are also many other putative excitatory neurotransmitters in the system whose roles remain elusive. Therefore, the aim of my PhD thesis was to elucidate molecular and pharmacological mechanisms underlying excitatory transmission in the ENS. In Chapter 2, I localized synaptic vesicle proteins synaptophysin, synaptotagmin-1 and vesicular acetylcholine transporter (vAChT) to enteric varicosities. I developed two high-throughput analysis methodologies to quantify co-expression in varicosities and their close contact with enteric neurons. Using these analysis tools, I found that synaptic vesicle proteins synaptophysin and synaptotagmin-1, described to be ubiquitous in pre-synaptic terminals, are not found in all cholinergic varicosities (vAChT+) in the myenteric plexus. I found that in the submucosal plexus, all cholinergic varicosities contained synaptophysin, but some lacked synaptotagmin-1. This highlights the sensitivity of the analysis tool developed and the disparity in synaptic protein localization at cholinergic varicosities between the two plexuses. Additionally, using 3D rendering I examined close contacts between varicosities expressing synaptophysin and vAChT on neuronal nitric oxide synthase (nNOS+) neurons. I found that nNOS+ neurons receive three distinct classes of input. This includes varicosities that either contain vAChT, synaptophysin or both. Overall, my findings demonstrate that there is molecular heterogeneity in cholinergic varicosities within the ENS, which will likely transpire into distinct modes of cholinergic transmission or ACh release at enteric synapses. Moreover, this study highlights the use of advanced image analysis tools to examine connectivity and mechanisms of transmission within the ENS. In Chapter 3, I described the expression of post-synaptic density protein PSD93 in the ENS using immunohistochemical methods. I found that most myenteric neurons, including subpopulations of cholinergic and nitrergic neurons express PSD93. The wide spread expression of PSD93 in the cytoplasm and axons of enteric neurons indicates that it is an unsuitable marker for identifying excitatory post-synaptic densities in the myenteric plexus. Instead, PSD93 is likely to be involved in other cytosolic processes in addition to any role as a post-synaptic density protein at excitatory synapses. In Chapter 4, I demonstrate importance of α-synuclein (α-Syn) in cholinergic function within the ENS. α-Syn is a synaptic vesicle protein pathologically linked to neurodegenerative diseases. I show that α-Syn is expressed in varicosities and some neuronal somata within the mouse colon, a result described previously in other species. Using the quantitative method described in Chapter 2, I found that most cholinergic varicosities (vAChT+) contained α-Syn. I also investigated the implications of α-Syn deletion for ENS function using α-Syn knock out (KO) mice. α-Syn KO mice have increased proportions of cholinergic neurons in the myenteric plexus. Additionally, cross-sections of mouse colon preparations also show that α-Syn KO mice have increased cholinergic innervation to the circular muscle. Calcium (Ca2+) imaging studies reveal that fast synaptic transmission mediated by nicotinic receptors is increased in α-Syn KO mice. However, I found that α-Syn KO mice have a reduced incidence of spontaneous circular muscle contractility, suggesting that there are changes in the circuitry underlying motor patterns. Collectively, these findings suggest that there are alterations in the enteric neural circuitry of α-Syn KO mice and that α-Syn is important for cholinergic transmission. In Chapter 5, I used Ca2+ imaging and high-resolution microscopy to elucidate the mechanisms behind glutamatergic transmission within the ENS. Thus far there is conflicting evidence to suggest the involvement of ionotropic receptors and metabotropic glutamate receptors (mGluRs) in synaptic transmission. I show that many myenteric varicosities that contain vesicular glutamate transporter 2 (vGluT2) are non-cholinergic and express synaptic vesicle proteins synaptophysin using tools I developed in Chapter 2. Using 3D rendering I showed that calbindin (calb+) neurons receive more vGluT2 varicosities than nNOS+ neurons. Exogenous application of glutamate predominantly excites calb+ neurons in the myenteric plexus. Calb+ neurons also receive slow synaptic transmission mediated by endogenous release of glutamate excited by a train of electrical stimuli. Using ionotropic and group I metabotropic glutamate receptor (mGluR) antagonists, I found that group I mGluRs are involved in mediating slow synaptic transmission. This study demonstrates a role for glutamate in mediating excitability of myenteric calb+ neurons. Overall, I have developed powerful methodologies that will provide valuable tools to contribute to understanding mechanisms underlying excitatory transmission within the ENS. The molecular heterogeneity of cholinergic varicosities identified in this thesis, provides a foundation for elucidating ACh release at enteric synapses. I have also shown that post-synaptic density markers that identify excitatory synapses in the autonomic nervous system (ANS) are unsuitable for labelling excitatory synapses in the ENS. This indicates that mechanisms underlying excitatory transmission could differ between the ANS and ENS. I have highlighted the difficulty in establishing a marker for post-synaptic densities within the ENS, which is necessary for a detailed understanding of excitatory transmission. Moreover, I have shown that α-Syn is associated with cholinergic synapses and the deletion of the synaptic vesicle protein has consequential effects on cholinergic transmission and function, thus implicating α-Syn in gastrointestinal pathophysiology. I have also identified a role for group I mGluRs in mediating excitatory slow synaptic transmission, indicating that glutamate is an excitatory neurotransmitter within the ENS. These findings provide a foundation for future analyses of synaptic function in the ENS and point to key questions for further investigation of this understudied nervous system.

Respiratory syncytial virus (RSV), a leading cause of acute lower respiratory illness in infants, immunosuppressed adults and the elderly, is responsible for more deaths each year than influenza. Despite this, there are no freely available treatment options, making the development of safe and efficacious anti-RSV therapeutics a high priority. However, in order to achieve this, a deeper understanding of the RSV-host cell interaction is required. RSV infection has previously been found to induce global changes in the mitochondrial proteome and interfere with mitochondria-mediated antiviral signalling, but details of the RSV-host cell mitochondrial interaction are poorly understood. Therefore, the aim of this thesis is to explore the impact of RSV infection on host mitochondria and its role in viral pathogenesis in order to identify novel anti-RSV strategies. The results presented in this thesis reveal for the first time that RSV induces a staged, microtubule/dynein-dependent redistribution of mitochondria, concomitant with reduced mRNA levels of genes encoding mitochondrial proteins, compromised mitochondrial respiration, dissipated mitochondrial membrane potential (Δѱm), and increased generation of mitochondrial reactive oxygen species (ROS). It was also found that inhibiting mitochondrial redistribution or mitochondrial ROS production strongly suppressed RSV virus production, highlighting the RSV-mitochondrial interface as a potential antiviral target. Analysis of RSV proteins identified the matrix protein (M) is sufficient and necessary to induce mitochondrial perinuclear clustering, downregulation of mitochondrial genes, inhibition of mitochondrial respiration, loss of Δѱm, and accumulation of mitochondrial ROS in infection, while deletion and mutation studies identified its central nucleic acid-binding domain, and arginine/lysine residues 170/172 in particular, as essential for its remodelling ability in host cell mitochondria. Recombinant RSV carrying the arginine/lysine mutations in M was unable to elicit these effects on host mitochondria, and its replication in infected cells was severely impaired, underlining the importance of M-dependent effects on mitochondria to RSV infection. Importantly, clinically relevant human cell models of RSV infection were examined, highlighting the importance of RSV’s impact on host mitochondria to its infectious cycle, and its relevance to human disease. Further, inhibiting the accumulation of mitochondrial ROS in infected cells was confirmed as a viable anti-RSV approach in these systems, and work was extended to include a mouse model that showed significantly reduced RSV-related pathology as a result of treatment with a mitochondrial ROS scavenger. In summary, the studies presented in this thesis shed new light on the impact of RSV infection on host cell mitochondria by establishing the unique ability of RSV, facilitated by the M protein, to co-opt the host cell mitochondria to enhance virus production. In addition, the importance of RSV’s impact on host mitochondria for pathogenesis is explored in multiple disease models, highlighting it as a potential target for the development of anti-RSV treatments. Significantly, the studies reveal the inhibition of mitochondrial ROS levels for the first time as a viable approach to counteract RSV infection.

Central to the capacity of many viruses to cause disease are their ability to subvert the host-cell’s innate antiviral immune defence, the interferon (IFN) response, through expression of IFN antagonist proteins. For rabies virus (RABV) and other members of the genus Lyssavirus, antagonism of IFN signalling is mediated by the products of their P gene, the phosphoproteins (P1 – P5). These isoforms target the vital IFN-activated signalling transcription factor signal transducers and activators of transcription 1 (STAT1) through diverse mechanisms related to their distinct intracellular localisations and association with different cellular proteins/structures. In particular, the P3 isoform interacts with the cellular microtubule (MT) cytoskeleton and induces STAT1-MT colocalisation as part of its strategy to inhibit STAT1 signalling. The contributions of this mechanism to viral pathogenesis in vivo, and the molecular details of the P3-MT interaction, however, remain unresolved. To address this, we used confocal imaging, single-molecule localisation microscopy, immune signalling assays and proteomics to perform quantitative comparative analyses of the intracellular phenotypes and IFN antagonist function of P3 proteins from a pathogenic RABV strain, Nishigahara (Ni), and a non-pathogenic, Ni-derivative strain, Ni-CE. We found that the MT-association of Ni-CE-P3 was significantly impaired compared to Ni-P3, and this correlated with a defective capacity to antagonise STAT1 signalling. Furthermore, we identified a single Ni-CE-P mutation, N226-H, that alone was sufficient to partially inhibit these processes and attenuate the capacity for RABV to cause lethal disease in mice. These data indicate that the P3-MT interface contributes significantly to RABV pathogenesis. Further analysis of the molecular mechanisms underlying P3-MT association identified several residues of P3 protein that may contribute MT-binding, including a possible regulatory mechanism involving phosphorylation of a previously identified protein kinase C target site, S210. We also present evidence that P3 protein induces bundling of MT filaments, which was also inhibited by the Ni-CE-P mutations. As this was similarly observed in cells infected with RABV in a P gene-dependent manner, this may indicate that MT-bundling may not only provide an indicator of P3-MT interaction in transfected cells, but might also be an important element of P3 protein activity in infected cells. Finally, examination of the P3 proteins expressed by diverse lyssaviruses revealed that they exhibit highly heterologous intracellular localisations, but retain a conserved targeting of STAT1. Despite this diversity, our data does suggest roles for MT-association in the STAT1 antagonist function of many lyssavirus P3 proteins. Together, the data presented in this thesis has advanced our understanding of the contributions of the P3-MT interface to lyssavirus immune evasion.

Human immunodeficiency virus (HIV) infection remains a major global health issue. Antiretroviral drugs improve life expectancy and significantly reduce the rate of viral transmission; however, we are far from finding a cure for HIV. The major barrier to finding a cure is the persistence of the replication-competent yet transcriptionally silent latent reservoir. Current latency reversal agents (LRA) lack efficacy to eliminate all the latent proviruses from the reservoir. The response to the same LRAs is varied in latently infected cells ex vivo or in vitro. We hypothesised that HIV could generate different populations of latently infected cells that differ in HIV integration sites and response to reactivation by LRAs. We used a Nef-competent EGFP reporter virus to generate infection and to determine the latently infected cells in chemokine-treated CD4+ T cells in vitro. We first demonstrated that EGFP expression is dependent on viral integration and can be used to determine productively expressed and latently induced infected cells in our culture system. Infection and latency were established in both resting untreated and CCL19-treated CD4+ T cells in vitro. Addition of integrase inhibitor, raltegravir, at time of infection reduced the levels of EGFP expression in both T cell conditions, providing evidence that in our culture system EGFP expression is dependent on viral integration. There was a 4-fold reduction in EGFP expression in the CCL19-treated compared to the matched resting untreated cells. The reduction in the EGFP expression following addition of integrase inhibitor strongly suggested that incubating CD4+ T cells with CCL19 favors viral integration in vitro. We subsequently showed that the addition of IL-7 significantly increases the levels of latency in the chemokine-treated CD4+ T cells. Thus, we clearly showed that both resting and chemokine-treated CD4+ T cells are permissive to direct infection with HIV in vitro. However, the effect of CCL19 in the induction of latency is more pronounced with the addition of IL-7.   We further asked whether the establishment of latency affects the response to reactivation by LRAs or T cell receptor (TCR) signalling. We used resting CD4+ T cells to establish infection in the pre-activation pathway and used activated T cells as a model for the establishment of infection in the post-activation pathway. Co-culturing EGFP- cells with allogeneic monocytes alone or in combination with an antibody against CD3 (aCD3); we showed a significant increase in EGFP expression from latently infected cells in the pre-activation latency model. Response to allogeneic monocytes in combination with signals derived from aCD3 significantly correlated with T cell proliferation and there was a minimal spontaneous EGFP expression from latently infected cells in this culture. In contrast, allogeneic monocytes alone or in combination with aCD3 reduced the EGFP expression from latently infected cells in the post-activation latency model. There was no correlation between T cell proliferation and viral expression. The level of spontaneous EGFP expression from latently infected cells was high, and the inhibition of EGFP expression by monocytes was dependent on the direct contact between monocytes and T cells. We further showed that the interaction between T cells and monocytes at time of infection induced spontaneous expression, providing evidence that monocyte-T cell interaction at an early time post infection maintains latency in activated T cells. By direct comparison of pre- and post-activation latency in vitro we, therefore, demonstrated that effective strategies to reverse latency would depend on how latency is established. We further profiled the HIV integration sites in pre- and post-activation latency models and showed a significant enrichment of the sites in genic, exon and intron; in sense direction in the introns of pre-activation latency compared to the post-activation models, suggesting preferential integration of proviral DNA in these locations.   By indexing genes with integration sites with gene expression available for these genes in GEO dataset using RNA-Seq analysis, we found a set of genes that are not expressed during activation of T cells in response to TCR stimulation. This observation was found across all T cell subsets in the GEO datasets and suggests there is a common mechanism in T cells that allows for viral entry and integration in non-expressing genes. Our study has clearly shown that how latency is established is a critical factor affecting how latency is maintained or reversed in response to LRAs. Understanding the relationship between chromatin status of the genes that are never expressed during activation of T cells and establishment of infection or latency is of interest for designing strategies to induce the expression from latency or to permanently silence the virus.

Metal dyshomeostasis is an important neurodegenerative event that can affect the structure and function of proteins, such as tau. Tau binds metals including copper, iron and zinc, and may undergo conformational changes that promote its aggregation, leading to neurodegeneration. Tau is mainly an axonal microtubule stabilising protein. However, tau is also located to dendrites where it mediates the transport of proteins to the postsynaptic terminal and may modulate the neuronal susceptibility to excitotoxicity. Furthermore, tau is present in bodily fluids, which can be used for detection of pathological biomarkers. The finding of tau in extracellular fluids and dendrites has directed its current research focus to the study of tau mechanisms of secretion and spreading. The aim of this project is to investigate the role of metals in the modulation of tau secretion, and to analyse if platelet tau and metal levels could serve as neurodegenerative blood biomarkers. To determine if metals could modulate tau secretion, primary cortical cell cultures from tau-knockout and wild-type mice were treated with copper, iron, zinc, glutamate and clioquinol, an 8-hydroxyquinoline derivative with ionophore activity. The results showed that copper, iron and clioquinol decreased tau secretion from cells whilst zinc and glutamate increased it. Copper and iron decreased APP secretion from cells whilst glutamate and clioquinol increased it. Additionally, tau-knockout cells showed decreased APP secretion compared to wild-type controls and metals did not alter their APP secretion. To further investigate if this occurs also in vivo, interstitial fluid (ISF) was collected using microdialysis on wild-type mice fed either with copper, iron or zinc. Zinc and iron increased tau ISF levels whilst copper decreased it. The analysis of baseline ISF tau levels collected over time from untreated animals suggested that tau could be involved in circadian rhythm. To evaluate if Alzheimer’s disease (AD) platelets present tau abnormalities and metal dyshomeostasis, platelet samples from 55 healthy and 45 AD subjects from the Australian Imaging, Biomarkers and Lifestyle flagship study of ageing (AIBL) were analysed by inductively coupled plasma-mass spectrometry (ICP-MS) and Western blotting. The results showed that copper is decreased in AD platelets. Additionally, healthy controls who are ApoEε4 carriers had the highest platelet copper content. Iron levels were decreased in AD platelets. Platelet zinc levels positively correlated with verbal fluency. ROC curves showed that copper and iron were poor biomarker candidates. There were no changes in total and GSK3β pS9 levels in AD platelets. To characterise platelet tau species, mouse brain and platelets from human, wild-type and tau-knockout mice were analysed by Western blotting and immunofluorescence. The results showed that tau immunoreactivity might be an antibody artefact. To further investigate this, tandem-MS was used to identify tau in a human platelet sample, using brain samples from wild-type and transgenic tau overexpressing mouse as positive controls. Tau was identified in mouse brain samples but not in human platelets. Additionally, a qRT-PCR was performed to platelets from humans, wild-type and tau-knockout mice to determine tau mRNA levels. The results showed that MAPT was expressed in wild-type and tau-knockout mice tissues but was not detected in either human or mouse platelets. The conclusions drawn from this thesis are that: (1) extracellular zinc may modulate tau secretion; (2) APP secretion may be tau dependent; (3) in AD platelets, there was no detectable level of tau and there was no change in GSK3β levels and activity; (4) platelets from AD patients exhibit metal dyshomeostasis, resulting in decreased copper and iron content. This change in platelet copper and iron concentrations alone do not provide sufficient sensitivity and specificity as AD biomarkers. Further studies are required to identify a panel of platelet biomarkers, together with changes in copper and iron that will provide high sensitivity and specificity as biomarkers for AD.

Background: Pericardial adipose deposition occurs in ageing and obesity, and independently contributes to the development of atrial fibrillation. The mechanisms underlying this association are not yet understood. Investigations to date have focused on physical conduction block posed by infiltrating adipose and the secretion of pro-inflammatory/pro-fibrotic paracrine factors into the atria. Though not yet investigated in the pericardial adipose, white adipose depots are established sites of oestrogen synthesis. Considering the reported actions of oestrogens on the heart, it is hypothesised that pericardial adipose may represent an important source of local oestrogen synthesis, exerting paracrine actions on the myocardium. Research questions: 1. Do myocardial and pericardial adipose tissues express aromatase, and do locally-derived oestrogens affect the vulnerability to atrial arrhythmia? (Chapter 2) 2. Does disruption of aromatase activity in aged and obese mice influence basal cardiac electrophysiology and the susceptibility to atrial arrhythmia? (Chapter 3) 3. Can liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS) sensitive methodology be used to quantify androgens and oestrogens in human and mouse myocardium and pericardial adipose tissues? (Chapter 4) Methods: Aromatase expression in human and rodent myocardium and pericardial adipose was measured by Western immunoblotting. Arrhythmia vulnerability was assessed in isolated hearts from male C57BL/6 mice (‘young’, ‘aged’ or ‘aged’ + high fat diet). Hearts were perfused with a hypokalaemic solution (2 mmol [K+]) and subjected to programmed electrical stimulation to provoke arrhythmias. In addition, hearts were perfused with acute perfusion with 17β-oestradiol (or vehicle) and arrhythmic provocation repeated. Aromatase knockout and wild type mice (male and female) were fed a control or high fat diet for 40 weeks. Mice were subjected to electrocardiographic and echocardiographic assessment prior to isolated heart atrial arrhythmia provocation experiments. Human and mouse myocardium and adipose tissues were homogenised, derivatised with dansyl chloride and subjected to LC-MS/MS for sex steroid quantification. Mass spectrometric technique was developed using the aromatase knockout as a positive control for androgens and a negative control for oestrogens. Results: 1. Aromatase is expressed in human/rodent myocardium and pericardial adipose, conferring the capacity for local androgen to oestrogen synthesis. Pericardial adipose capacity to synthesise oestrogens increased by 30-50x in aged hearts, which were significantly more vulnerable to atrial arrhythmias. (Chapter 2) 2. The aromatase knockout model of oestrogen depletion and androgen excess revealed a sex-differential phenotype in the susceptibility to atrial arrhythmia. Left atrial action potential duration was prolonged and arrhythmia vulnerability greater in female aromatase knockout mice compared to all other groups. The combined influence of extensive pericardial adipose deposition and a highly androgenic/oestrogen-depleted environment was unique to the female aromatase knockout mice and may have been decisive in driving the exacerbated vulnerability to atrial arrhythmias. (Chapter 3) 3. LC-MS/MS methodologies were optimised for the detection and quantification of sex steroids in human/mouse myocardium and adipose. Successful quantification of testosterone and progesterone was achievable, but concentrations of oestrogens in tissues were below the technical limits of detection. (Chapter 4) Conclusions: This thesis identifies that pericardial adipose expresses aromatase and indicates a probable capacity for oestrogen synthesis, hence supporting the presence of a local cardiac androgen-oestrogen system. Pericardial adipose derived oestrogens (and androgens) are recognised as probable paracrine mediators capable of altering atrial arrhythmic vulnerability. In addition, the data support the clinically observed correlation between pericardial adipose accumulation and atrial fibrillation. Mass spectrometric methodology is capable of quantifying tissue testosterone and progesterone concentrations, but tissue oestrogens are below the limits of detection. Taken together, this thesis advances the mechanistic understanding of the link between pericardial adipose accumulation and greater atrial arrhythmia vulnerability.

Malaria is the disease caused by infection of red blood cells with the protozoan parasite, Plasmodium. In 2016 alone, 216 million people suffered from malaria, leading to 445 000 deaths. Artemisinin-based combination therapies are the first-line treatment currently recommended by the World Health Organisation, for uncomplicated Plasmodium falciparum malaria. Youyou Tu, the scientist who led the team that discovered artemisinin, was awarded a share in The Nobel Prize for Medicine in 2015, in recognition of the importance of this discovery. Yet the mechanism of action of this life saving drug remains largely unknown. Artemisinin and its derivatives (ARTs) are sesquiterpene lactones that contain a 1,2,4- trioxane core and endoperoxide bridge. ARTs are widely accepted to be pro-drugs that are activated inside the cell by iron-catalysed reductive scission of the endoperoxide bridge. The resultant radical species is thought to rapidly react with accessible nucleophiles, including free thiols of cysteine residues in nearby proteins. Several studies have demonstrated that ARTs form adducts with hundreds of different parasite proteins in different compartments of the cell. However, the critical event leading to parasite death has not yet been elucidated. We hypothesised that ART-induced parasite killing is triggered by a lethal accumulation of cellular and protein damage. Here we show, using the novel cell-permeable thiol probe, tetraphenylethene maleimide (TPE-MI), that treatment of P. falciparum cultures with the clinically relevant ART derivative dihydroartemisinin (DHA), causes an increase in the level of unfolded proteins in the cell, indicating protein damage. We further show that DHA activates the Unfolded Protein Response (UPR), a well-conserved eukaryotic signalling pathway triggered by accumulation of unfolded proteins in the endoplasmic reticulum (ER). The UPR works to restore by protein homeostasis by arresting protein translation, thereby halting the influx of newly synthesised unfolded protein into the ER and thus preventing further increases in unfolded protein and protecting the cell from unfolded/misfolded protein damage. The UPR of protozoan parasites primarily involves stalling of protein translation via eIF (eukaryotic initiation factor)-2a phosphorylation. The P. falciparum kinase responsible Abstract ii for regulation of the UPR has not yet been identified. Here we identify, using genetic knockdown and small molecule inhibition, that Protein Kinase 4 (PK4) phosphorylates eIF2a in response to DHA and 1, 4-dithiothreitol (DTT), a reducing agent well known to induce ER stress and activation of the UPR. The ER also works to restore protein homeostasis by disposing of terminally misfolded proteins via the ubiquitin-proteasome pathway, a process termed ER-associated degradation. We show that DHA treatment actually results in a reduced capacity of the proteasome to degrade protein, thereby resulting in a build-up of polyubiquitylated proteins. Furthermore, we find that co-treatment of parasites with DHA and chemical inhibitors of polyubiquitylation, prevents polyubiquitylated proteins from accumulating, reduces the level of ER stress and strongly antagonises DHA-induced killing. These findings lead us to conclude that accumulation of polyubiquitylated proteins is the critical event that underlies DHA-induced parasite killing. Furthermore, we show that ART resistant parasites appear to accumulate a lower level of polyubiquitylated proteins following ART treatment. Thus, a defect in protein ubiquitylation may underlie ART resistance. Altogether, we propose a mechanism for ART action, whereby ARTs kill parasites via a two-pronged approach: inducing protein misfolding/unfolding and preventing protein degradation.

Neurodegenerative diseases such as Alzheimer’s disease and prion disease are closely related with specific gene and protein dysfunction. Prion diseases, also known as transmissible spongiform encephalopathies (TSEs), are characterized by the structural transformation of the cellular prion protein (PrPC) to the disease associated isoform (PrPSc). One hallmark of Alzheimer’s disease is the accumulation of beta amyloid (Aβ) plaques in the brain, resulting from the pathological cleavage of the amyloid precursor protein (APP) by β-secretase (BACE1) and the γ-secretase complex. MicroRNAs (miRNAs) are a class of small non-coding RNAs that regulate gene or protein expression by targeting mRNAs and triggering either translational repression or mRNA degradation. Distinct expression levels of miRNAs, including miR-29b and miR-146a, have been detected in various biological fluids and tissues from prion disease and Alzheimer’s disease patients, as well as in cell and animal models. These miRNAs could be potential diagnostic biomarkers of these diseases, suggesting that investigating the miRNA functional roles and miRNA-target regulation pathways will improve our understanding of the disease regulation networks. The first aspect of the thesis utilized CRISPR/Cas9 gene editing to knockdown miR-29b and miR-146a respectively in a number of cell lines, and cell clones with stable miRNA knockdown were generated. Off-target analysis of the cell clones revealed the high specificity of CRISPR/Cas9 editing of miRNAs. Common and distinct pathways and novel targets of miR-29b were also identified in two cell lines using transcriptome profiling, and potential miR-29b targets in Alzheimer’s and prion diseases were revealed. In the second aspect of the thesis, miR-29b was shown to positively regulate prion protein levels in both miR-29b stable knockdown cell clones and miR-29b overexpressed cells. This regulation is not mediated through miR-29b target SP1 or potential target PPP2CA, which can interact with prion protein or was implicated in prion pathogenesis. miR-29b could further affect PrPSc generation through regulating prion protein levels and potentially affect prion progression. miR-29b was also revealed to regulate APP and BACE1, the two key proteins in Alzheimer’s disease, in in vitro models. Lastly, the dual roles of miR-146a in regulating prion protein and inflammatory pathways were revealed in prion disease. miR-146a can upregulate prion protein levels in both overexpressed and stably downregulated cell models, as well as in miR-146a transgenic mice generated using CRISPR/Cas9 gene editing. miR-146a overexpression also resulted in the decreased formation of PrPSc in prion cell models. The miR-29b/miR-146a-PrP-PrPSc pathways possibly share a similar mechanism involving the interaction of prion protein with Argonaute protein – the key component of miRNA induced silencing complex (miRISC). Prion protein was demonstrated to be a direct target of miR-146a. miR-146a can also target inflammatory regulator TRAF6 in both prion infected cell models and in miR-146a transgenic mice. The findings from this thesis have important implications for the comprehensive understanding of prion disease pathogenesis. The miR-29b/miR-146a-PrP-PrPSc pathways and miR-146a mediated inflammatory pathway are added to the regulation network of prion disease. miRNAs represent novel regulators in prion diseases and other neurodegenerative disorders and hold promise to be future therapeutics to cure prion disease.