Brachyspira hyodysenteriae is an anaerobic intestinal spirochaete and the aetiological agent of swine dysentery (SD). Throughout the UK and Europe, pathogenic and potential non-pathogenic isolates of B. hyodysenteriae have been recovered from pig herds, creating major obstacles for the detection and control of this economically important pathogen. Therefore, the main aim of this research was to compare one representative of virulent (P8544) and one representative of avirulent (P7455) strains of B. hyodysenteriae using genomic and proteomics approaches with a view to identify distinctive genes or proteins. The B. hyodysenteriae draft genomes of P8544 and P7455 consisted of a circular 3.0 Mb chromosome and a 31,469-34,822 bp circular plasmid that is also present in the only published B. hyodysenteriae genome, WA1. A considerable number of genes (~27-35) were identified in both the virulent and avirulent strains that shared high sequence homology with genes found in other spirochaetes, such as B. murdochii and B. intermedia, as well as in other species of bacteria; these may have been acquired via horizontal gene transfer. Comparative genomics of the two pathogenic genomes P8544 and WA1 versus the non-pathogenic genome P7455 revealed that the gene encoding for the methyltransferase type 11 (Bhyoa7455_20) was identified as being unique to the P7455 plasmid sequence and was successfully PCR amplified in a greater number of avirulent than virulent strains. However, as this was only just statistically significant (P=0.049), screening of a much larger strain set would clearly be required to support this gene as a suitable genetic marker to distinguish virulent and avirulent B. hyodysenteriae strains. Bacterial acquisition of iron in-vivo is crucial for successful colonisation and persistence in the host. A further aim of this study was to compare the growth phenotype of B. hyodysenteriae isolates P8544 and P7455 grown under iron-limiting conditions; such as would be found in-vivo in the large intestine of the host. Analysis of P8544 and P7455 growth rate in iron-sequestered media (containing 0.1 mM of the iron-chelator dipyridyl) demonstrated that both these isolates could replicate in this media although with an extended lag-phase of approximately 32-34 hrs; growth rate was on par with the iron-replete conditions. qRT-PCR analysis of eight putative iron-acquisition genes under iron-sequestered and iron-replete conditions revealed a difference in transcription for a number of ABC-transporter genes in P8544 and P7455, however, none of these were classified as statistically significant. Non-quantitative shotgun proteomic based approach was used to analyse outer-membrane protein (OMPs) expression of P8544 and P7455 under low-iron and iron-replete growth conditions and revealed alteration in the OM expression profiles between the isolates and conditions using KEGG analysis. The majority of expressed proteins under iron-replete conditions were categorized in membrane transport (11%) and carbohydrate metabolism (7%). Under iron-restriction the OM profile changed most obviously in a decreased percentage of proteins particularly assigned in the categories energy metabolism and membrane transport. The percentage of proteins assigned no predicted function increased by 19% under iron-limited conditions highlighting the fact that biological functions of the majority of these expressed proteins in such an environment remains to be determined. Two-dimensional gel-electrophoresis (2-DGE) of whole cell fraction indicated that the alkyl-hydrogen peroxide reductase protein (AhpC) in P7455 and the non-haem iron-containing ferritin (Bhyov8544_1528) in P8544 were significantly (P<0.05; 1.5-fold) more expressed under iron-restricted conditions than under iron-replete conditions. These data confirmed the importance of iron to virulent and avirulent B. hyodysenteriae. The so far identified significantly expressed proteins may serve as a potential biomarker for global diagnostic purposes for B. hyodysenteriae infections rather than a tool for differentiation for virulent and avirulent isolates. However, further work is required to prove if these candidates are expressed in-vivo and conserved in a wider panel of field isolates. In conclusion, this research has contributed to the scientific knowledge regarding B. hyodysenteriae stress responses induced by iron-starvation and has provided further insight into the genetic and proteomic make up of this spirochaete. This work should also aid future investigations concerning the biology and pathogenicity of this important and grossly understudied swine pathogen.

No description available.

Escherichia coli contributes to the commensal microbiota of most mammals by producing vitamins and aiding digestion. However, several strains of E. coli have evolved as pathogens and have become highly adapted for specific niches through the acquisition of pathogenicity islands. E. coli O157:H7 is a commensal bacterium of cattle but if transferred to humans, usually by contaminated food products, it can act as a pathogen. In humans E. coli O157:H7 causes diarrhoea, haemorrhagic colitis and in some cases haemolytic uremic syndrome which can result in death. Clinical treatment of this pathogen is difficult since the antibiotics currently available have been shown to worsen the clinical outcome of infection. Therefore, the discovery of novel anti-bacterial therapies is of high importance. A novel approach to limit pathogenesis is to target the factors used by bacteria to cause disease - their key virulence factors. In theory, such approaches should only compromise the ability of the pathogen to cause disease, rather than its ability to survive, thereby reducing the selective pressure for the development of resistance mechanisms. The type three secretion system (LEE T3SS) is a key virulence factor for E. coli O157:H7 as it facilitates tight attachment to host cells and the secretion of effector proteins. The importance of this virulence factor for the disease process makes it an attractive target for anti-microbial therapies. The salicylidene acylhydrazides (SA) are a class of compounds that inhibit the expression of the T3SS of several Gram-negative pathogens. When this study was started, the mode of action of these compounds was completely unknown. An affinity pull-down assay to identify the binding proteins of the SA compounds was conducted. Identification of the target proteins of the compounds was the first step in determining their mechanism for decreasing the expression of the LEE T3SS in E. coli O157. The pull-down identified nineteen putative targets, none of which had previously been linked to the regulation of the LEE T3SS. The aim of this thesis was to systematically investigate the putative targets using a combination of approaches: phenotypic studies of deletion mutants and structural and functional characterisation of the target proteins. From the nineteen putative targets seven were selected for further investigation namely Tpx, WbrA, FolX, FkpA, FklB, SurA and AdhE. Several of these proteins were shown to bind to the SA compounds however deletion of only one of the target proteins resulted in a decrease in LEE T3S. This target, AdhE, offers an exciting new lead in the search for novel targets for antibacterial therapies.

Kinetoplastid protozoans have unusual cell cycles. Three unitary organelles need to be replicated and segregated to the daughter cells at each cell division; the nucleus, kinetoplast and basal body, associated with the flagellum The replication and segregation of these organelles requires to be co-ordinated. Inthis study the timing of nuclear and kinetoplast cell cycle events was examined in two species of kinetoplastids, Trypanosoma brucei and Leishmania mexicana, which are major pathogens of humans, to ascertain the degree of co-ordination between the two organelles in the cell cycle. Particular attention was paid to three main features; the duration of S-phase for each organelle, the relative timing of mitosis and kinetoplast division and the lengths of the post-mitotic/division cytokinesis periods. Two life cycle stages were examined for each species i.e. procyclic and bloodstream forms of T. brucei and promastigote and amastigote forms of L. mexicana. The objectives of this study were to establish firstly whether a similar pattern of events occurs for both stages within each species examined and secondly, whether or not T. brucei and L. mexicana share common features to their cell cycles. Studies of cell cycle events were conducted using immunofluorescent labelling to detect S-phase in both organelles, enabled by use of an anti-bromo-deoxyuridine (BrdU) antibody as an S-phase marker. Division of organelles was identified by staining the cells with DAPI, a DNA intercalating dye. Initial studies confirmed that the cell cycles of both species were consistent with the general pattern of events of the eukaryotic nuclear cycle. Also, as has been described previously in kinetoplastids, the kinetoplast DNA (containing mitochondrial DNA) has a pattern of discretely separated phases of replication and division analagous to the nuclear cycle. Analysis was made of the relative timing of S-phase, mitosis/division and cytokinesis for each life cycle stage, producing data which were statistically testable. For both life cycle stages of T. brucei a similar pattern of events was observed. Three populations of BrdU labelled cells were identified; cells which were BrdU labelled in the nucleus only, labelled in the kinetoplast only and labelled for both organelles. These observations indicate that there was a non-co-ordinate start and finish to S-phase, with an overlap in the timing of the S-phase periods. Kinetoplast division was initiated and completed before the start of mitosis in the nucleus and an extended cytokinesis period for each organelle was identified, although of longer duration in the kinetoplast. For both life cycle stages of L. mexicana, three populations of cells labelled with BrdU were observed as in T. brucei, again indicating that there was a non-co-ordinate start and finish to S-phase, with an overlap in the timing of both periods. The sequence of division for the organelles for both life cycle stages differed to that observed in T brucei. The nucleus divided before the kinetoplast and there was a much shorter cytokinesis period for both organelles in comparison to T brucei. These differences in the pattern of events may reflect either a difference in the control of cell cycle timing in each species or in the morphology of each cell type. Trypanosomes have to re-arrange their organelles, such that they occupy sites on either side of the division furrow, hence the extended cytokinesis phase. In L. mexicana, however, this requirement for organelle re-positioning is reduced or absent and therefore the cytokinesis periods are much shorter. Antigenic variation is the key strategy which allows African trypanosomes to evade the effects of the host's immune response. The switching from expression of one variant surface glycoprotein (VSG) to that of another has been indirectly linked to the cell cycle, as bloodstream slender forms divide and undergo antigenic switching, whereas bloodstream stumpy forms are non-dividing and do not appear to switch. An attempt to examine directly the proposed link between antigenic switching and the cell cycle was made using cell cycle markers (anti-BrdU antibody as an S-phase marker and DAPI staining to determine the configurations of the organelles), as well as VAT specific antibodies against VSGs expressed in the cloned lines studied. Unfortunately, trypanosomes expressing two VSG coats simultaneously and therefore undergoing antigenic switch, could not be detected in any of the cloned lines examined. This unresolved difficulty was attributed to a deficiency in the detection system employed. Differentiation in kinetoplastids is thought to be linked to the cell cycle, as progression through the life cycle involves transitions from proliferative to nonproliferative forms. The differentiation of Leishmania major promastigote forms, from dividing non-infective stages to non-dividing infective meta cyclic forms, involves major molecular and morphological changes. Using three metacyclic-specific markers (non-agglutination by peanut agglutinin, a monoclonal antibody (3F12) against metacyclic-specific epitopes of the major surface molecule lipophosphoglycan (LPG) and a rabbit anti-serum (ab336) against a metacyclic-specific surface protein, the Gene B protein) metacyclic production in in vitro culture was examined. Observation of population growth CUIVes suggests that differentiation is in part intrinsically programmed within the parasite but may also be inducible by environmental changes. Comparison of these data with mathematical models indicated that the promastigote population was likely to be heterogeneous, containing sub-populations that replicate and differentiate at different rates. Two major assumptions of the heterogeneous model are that meta cyclic forms are both non-dividing and incapable of dedifferentiation. These two possibilities were tested experimentally and indicated that meta cyclic forms are indeed non-dividing, but are also capable of de-differentiation, albeit at a very low rate. Examination of meta cyclic production at the cellular level also indicated that the event which causes commitment to differentiation occurs at least one cell division before symmetrical production of two daughter meta cyclic forms. It has been found that in the presence of a chronic infection the growth of a secondary infection is significantly inhibited (Turner et al., 1996). The available data suggests that there appears to be an overall down-regulation in the growth of the entire mixed population. The aim of this component of the project was to select for 'growth' mutant trypanosomes i.e. mutants which overcome growth inhibition. Bloodstream trypanosomes which had been mutagenised in vitro with ethyl methane sulphonate (EMS) were inoculated into mice in the presence of a pre-existing chronic infection. Populations of mutagenised trypanosomes which grew significantly in comparison to controls where only low rates of growth occurred, were selected by optical cloning. This rationale led to the generation of growth mutant clones which stably expressed the altered growth phenotype. These studies may permit the development of methods for the investigation of the regulation of growth and virulence in these parasites. In conclusion, the analysis of the relative timing of cell cycle events in T. brucei and L. mexicana has highlighted both common and novel features to the cell cycles of these kinetoplastids. The study of meta cyclic production in L. major has also revealed the intrinsic and extrinsic nature of regulation of differentiation in these parasites, as well as the commitment to differentiate at the cellular level. Furthermore, the generation of 'growth' mutant trypanosomes provides an additional tool for the future study of growth regulation in trypanosome infections.

Plasmodium chabaudi has been shown to undergo antigenic variation during the course of infection in mice. The importance of this model is the similarity and applicability of its features to infection of humans in P. falciparum. This thesis presents work performed using P. chabaudi to study various aspects of antigenic variation in asexual erythrocytic malaria parasites. The course of infection of P. chabaudi in N1H mice shows an initial acute parasitaemia which clears to subpatency. This is usually followed, after a period of days, by a second, and occasionally a third, recrudescent parasitaemia of lesser magnitude and duration. A cloned parent parasite population and cloned parasite populations derived from a recrudescence of the parent were tested in an indirect fluorescent antibody test on live, schizont-infected RBC (live IFAT) using a panel of hyperimmune sera raised against these populations and against one of the recrudescent clones after mosquito transmission. This test can detect antigens on the surface of parasitised RBC. The results of this analysis indicated that all the recrudescent clones were antigenically different from the parent and some were different from each other. In total, including the parent, six variant antigen types (VATs) were identified. Some of these also appeared to vary in immunogenicity. The effects of mosquito transmission on expression of variant antibodies was also examined using the panel of hyperimmune sera in the live IFAT. Mosquito transmission of two antigenically distinct recrudescent clone populations resulted in a change in antigenicity of both types to an apparently similar VAT, which had the same apparent identity as that of the original, post mosquito transmission but pre-cloning, parent population.

The increasing spread of antimicrobial resistant bacteria and the decline in the development of novel antibiotics have incited exploration of other avenues for antimicrobial therapy, such as the use of antibiotics that enhance the host’s defenses to infection. This study explores the influence of antibiotics on the innate immune responses to bacteria. The aims were to investigate antibiotic effects on bacterial viability, innate immune cells in response to bacteria and interactions between bacteria and the host. Five exemplar antibiotics at maximum serum concentration (C\(_m\)\(_a\)\(_x\)) and minimum inhibitory concentrations (MIC) were tested. \(Salmonella\) Typhimurium SL1344 was chosen as the model pathogen. Following incubation of SL1344 with C\(_m\)\(_a\)\(_x\) concentrations of ceftriaxone bacterial viability was undetectable. When SL1344 was incubated with ceftriaxone and ciprofloxacin treated neutrophils, there was reduced bacterial viability. Ciprofloxacin pre-treated neutrophils had reduced ability to phagocytose bacteria, while oxidative burst was increased following exposure to ceftriaxone. Adhesion of SL1344 to J774 macrophages pre-exposed to both concentrations of ciprofloxacin and ceftriaxone was increased, but only C\(_m\)\(_a\)\(_x\) of azithromycin and streptomycin. Expression of IL-1β and TNFα mRNA was greater in SL1344 infected macrophages pre-exposed to ciprofloxacin or ceftriaxone, than in macrophages exposed to antibiotics alone or SL1344 alone. In conclusion, it was found that clinical relevant concentrations of certain antibiotics enhance the response of immune cells and their interaction with bacteria, by increasing phagocytosis and killing in neutrophils, increasing bacterial adhesion to macrophages and increased cytokine production. These immunomodulatory potentials of antibiotics can be harnessed and exploited for broader therapeutic use.

Thiomarinol is a novel hybrid antibiotic produced by \(Pseudoalteromonas\) \(rava\) sp. nov. SANK 73390. It is structurally similar to a clinically significant antibiotic mupirocin, but includes an additional pyrrothine moiety joined to the mupirocin-like marinolic acid via an amide bond. Thiomarinol has been shown to be more potent against a wider range of microorganisms. This potency was hypothesised to be due to either an increase in inhibition of its target enzyme, isoleucyl-tRNA synthetase (IleS) or an increase in antibiotic uptake and/or inefficient efflux by bacterial cells or a combination of both. This thesis describes experiments that investigate the basis for this increase in potency.

Mammalian cell entry (MCE) domains are so called due to the reported ability of an Escherichia coli strain harbouring the mce1 gene from Mycobacterium tuberculosis to invade mammalian cells. Bioinformatic analyses presented here demonstrated that proteins containing a single MCE domain are widespread in bacteria and that proteins containing multiple MCE domains are specific to and have evolved within Proteobacteria. Gene neighbourhood analyses revealed that MCE domain containing proteins are components of transporters and that multi MCE domain containing proteins constitute a novel type of transporter. E. coli was shown to harbour three MCE proteins: the single MCE domain protein MlaD and two multi-domain proteins PqiB and YebT. All three proteins were shown to locate to the inner membrane and bind phospholipids. Phenotypic studies revealed that their functions overlap but are distinct. Infection studies with Salmonella showed that the proteins are important for systemic infection but are not required for mammalian cell entry. Phospholipid growth experiments with Salmonella demonstrated that they are important for phospholipid uptake. These findings suggest that MCE domain containing proteins in Proteobacteria are not directly involved in mammalian cell entry and instead play a role in other aspects of mammalian infection related to phospholipid trafficking.

The intestinal immune system plays an essential role in maintaining the delicate balance between mounting protective responses against invading pathogens and sustaining tolerance towards self-antigens and the endogenous microbiota. Disturbing this balance leads to intestinal inflammation, such as is seen in inflammatory bowel disease (IBD). IBD is characterised by alterations in the mucosa-associated microbiota, such as the increase in adherent and invasive Escherichia coli (AIEC) species in Crohn’s disease (CD) patients. Concomitantly, the inflamed mucosa exhibits an elevated rate of apoptosis in IBD, a phenomenon that also accompanies infection with a range of enteric bacterial pathogens. While phagocytosis of apoptotic cells by dendritic cells (DC) is required for self-tolerance in the healthy intestine, there is evidence to suggest that apoptotic cell uptake during infection activates protective T cell responses. In order to investigate the link between the recognition of apoptotic cells and intestinal inflammation, we used a range of different in vivo enteric bacterial infection models. Previously published work had implicated the AIEC strain NRG857c in the induction of chronic intestinal inflammation in vivo. We did not, however, find that NRG857c caused any signs of chronic colitis in mice, either by histological examination, or in-depth analysis of both innate and adaptive immune responses in the lamina propria and mesenteric lymph nodes (MLN). Due to the important role of DC in acquiring apoptotic cell antigen and priming protective T cell responses, we next characterised the expression of apoptotic cell receptors, specifically TIM4, on DC populations in steady state mucosal tissues. We demonstrated that TIM4 expression was enriched on CD11b- CD103+ DC, which have previously been shown to cross-present apoptotic cell-derived antigen. However, upon migration in mesenteric lymph, all intestinal DC populations upregulated TIM4, and migratory CD11b+ CD103+ had the highest frequency of TIM4+ cells in the MLN. However, blocking TIM4 did not affect DC migration in vivo. We also found that infection with C. rodentium elevated the percentage of TIM4+ DC in a population-specific manner, but that TIM4 was not essential for the induction of protective T cell responses during infection with either C. rodentium or S. Typhimurium. We therefore provide a detailed analysis of the intestinal immune response to bacterial infection, focussing specifically on the role of the apoptotic cell receptor TIM4 on intestinal DC populations.

Periodontal disease (PD) is a multifactorial disease of the oral cavity affecting the majority of the population. Although not a direct cause of mortality, PD is a health concern because it affects the majority of the population and has a negative impact on oral health, ability to chew, appearance, quality of life, dental care costs and can lead to tooth loss. Dental plaque is a microbial biofilm, which is necessary but not sufficient for the development of periodontitis. The interactions between the biofilm and the host cells, both local tissue and immune cells, can lead to tissue destruction and ultimately tooth loss. Clinical management of periodontitis involves mechanical removal of plaque from the tooth surface. Treatment is time consuming, in some patients only partially successful and recurrence is common. Therefore, understanding how the host interacts with microbial biofilms in both health and PD will help improve treatments and identify novel targets for therapeutic and preventative strategies. The hypothesis of this thesis is that the bacterial composition of oral biofilms may modulate host cell responses which contribute to the pathogenicity of PD. The overarching aim of this research was to develop an in vitro co-culture model system to study how biofilm composition can influence the host immune response. The studies document the development of health-associated, intermediate and disease-associated biofilms with host tissue and immune cells, and the use of these models to test antimicrobial and anti-inflammatory compounds as potential treatments for PD. The biofilms developed were assessed for survival in cell culture conditions and batch reproducibility by PCR and morphology visualised using SEM. The health-associated biofilm included Streptococcus mitis, S. intermedius and S. oralis (3-species); the intermediate biofilm additionally included Veillonella dispar, Actinomyces naeslundii, Fusobacterium nucleatum and F. nucleatum spp. Vincentii (7-species); and the disease-associated biofilm included further addition of Porphyromonas gingivalis, Prevotella intermedia, and Aggregatibacter actinomycetemcomitans (10-species). These biofilms were co-cultured with an oral epithelial cell line and primary gingival epithelial cells, as well as neutrophils and a myeloid cell line. Host cell viability was assessed by AlamarBlue®/LDH and changes in mRNA and protein expression of chemokines and cytokines were assessed by quantitative PCR and ELISA/Luminex®, respectively. Cellular responses were further evaluated by microscopy and flow cytometry. Generally, co-culture of health associated biofilms with host cells resulted in minimal impact on cell viability and generally low inflammatory gene expression and protein release, with some genes including CXCL5 and CCL1 being downregulated compared to the cells only control. Intermediate biofilms caused some cell death and a marked upregulation of inflammatory genes and protein release, including a 302.7 fold increase of epithelial cell IL-8 gene expression compared to the cells only control. These intermediate biofilms elicited significant upregulation of CD40 and CD69 expression on the monocyte cell line compared with untreated controls. Co-culture of the 10 species disease associated biofilms with host cells resulted in significant host cell death of both epithelial cells and monocytes. The 10 species biofilm caused significantly increased pro-inflammatory gene expression, but only low levels of protein could be detected in the supernatants. Similar trends in upregulation of inflammatory gene expression but low levels of protein release was observed in co-culture with differentiated pro-monocytes, whereas upregulation of inflammatory gene expression and protein release in neutrophil co-cultures was observed. The effect of antimicrobial and anti-inflammatory compounds, resveratrol and chlorhexidine, was evaluated using this model system. Prior treatment of epithelial cells with resveratrol and biofilm with chlorhexidine significantly reduced IL-8 release from epithelial cells in co-culture with biofilms for 4 and 24 hours. In conclusion, this research has developed and validated 3 complex multi-species biofilms to study host: biofilm interactions in vitro. Furthermore, using these models in co-culture with multiple host cell types, clear differences in the host response to different biofilms were observed. The variations in inflammatory response of host cells and oral biofilms observed in this study help further understanding of the complex host: biofilm interactions within the oral cavity which contribute to PD. This model demonstrated its potential as a platform to test novel actives, highlighting its use a tool to study how actives can influence host: biofilm interactions within the oral cavity. Future use of this model will aid in greater understanding of host: biofilm interactions. Such findings are applicable to oral health and beyond, and may help to identify novel therapeutic targets for the treatment of PD and other biofilm associated diseases.