Infections due to Streptococcus pneumoniae (the pneumococcus) remain a substantial source of morbidity and mortality in both developing and developed countries despite a century of research and the development of effective therapeutic interventions (such as antibiotic therapy and vaccination). The ability of the pneumococcus to evade multiple classes of antibiotic through several genetically determined resistance mechanisms and its evasion of capsular polysaccharide based vaccines through serotype replacement and capsular switching, all reflect the extensive diversity and plasticity of the genome of this naturally transformable organism which can readily alter its genome in response to its environment and the pressures placed upon it in order to survive. The purpose of this thesis is to investigate this diversity from a genome sequence perspective and to relate these observations to pneumococcal molecular epidemiology in a region of high biodiversity, the pathogenesis of certain disease manifestations and assess for a possible bacterial genetic basis for the pneumococcal phenotypes of, “carriage” and, “invasion.” In order to do this, microarray comparative genomic hybridization (CGH) has been utilized to compare DNA from a variety of pneumococcal isolates chosen from 10 diverse serotypes and Multilocus Sequence Types and from clinically relevant serotypes and sequence types (particularly serotypes 3, 4 and 14 and sequence types ST9, ST246 and ST180)) against a reference, sequenced pneumococcal genome from an extensively investigated serotype 4 isolate – TIGR4. Microarray comparison of the transcriptional profiles of several isolates has also been undertaken to compare gene expression from isolates of serotype 1 (ST227 and ST306) and serotype 3 (ST180) related to particular disease states and exposure of a multi-resistant pneumococcus to an antimicrobial (clarithromycin) commonly used to treat pneumococcal pneumonia.

Microspheres have become an important material in the biomedical and environmental sciences, particularly for use in the detection of pathogenic microorganisms and toxins as well as for use as carriers in drug delivery. In this study, their use in microbial detection, with particular emphasis on immunochromatographic assays, was investigated. Two main types of microspheres were studied: colloidal gold and polymeric. The original plan to combine the use of colloidal gold and fluorescentdye- labelled PMMA microspheres as a signal generator, in order to enhance detection signals and improve detection limit, was abandoned when preliminary detection experiments showed that the use of colloidal gold was not so beneficial after all, taking into account the amount of light lost to absorption by the gold particles. Therefore, it was decided to use Rhodamine B-labelled PMMA microspheres. PMMA particles, both unlabelled and internally labelled with Rhodamine B, were synthesised by emulsion polymerisation and yielded monodisperse particles of around 200 nm and 300 nm, respectively. An attempt to co-polymerise MMA with HEMA to form 200 nm-sized monodisperse P(MMA-HEMA) microspheres in order to create functional –OH groups on the microsphere surface to be used in chemical covalent coupling with monoclonal antibodies resulted in aggregated microspheres and non-uniform particles, and were therefore not used for covalent coupling. Attachment of monoclonal antibodies onto the surface of Rhodamine-B labelled PMMA microspheres by passive adsorption also resulted in aggregated particles. Diffusion and detection experiments were carried out on the Rhodamine-B labelled PMMA microspheres. Diffusion of PMMA microspheres along a nitrocellulose strip was found to be 42% slower than the diffusion of colloidal gold along a nitrocellulose strip. Using a spectrophotometer, detection experiments were performed on dilutions of an original stock solution of 1% w/v PMMA in distilled water. The detection limit was found to be of the order of 10-3. This study has investigated the materials that constitute an immunochromatographic assay and has illustrated some of the complications associated with the synthesis of copolymer microspheres and immobilisation of antibodies onto their surfaces. Further work on how to improve on the methods discussed in this study have also been recommended.

Streptococcus pneumoniae, also known as the pneumococcus, is an important human pathogen, with high burdens of disease and mortality worldwide. There are over 90 serotypes of this pathogen, demonstrating the vast amounts of diversity present. Currently, there are two pneumococcal vaccines, both targeting the polysaccharide capsule. However, one vaccine is ineffective in the paediatric population, whilst the other only targets a minority of disease-causing serotypes, and has increased disease caused by serotypes not present in the vaccine. One solution is a new pneumococcal vaccine targeting a protein virulence factor possessed by all pneumococci, which would afford cross-serotype protection. As a result, it is important to assess the diversity of pneumococcal virulence factors in order to determine their potential as vaccine candidates, as excess diversity present may prevent full serotype-independent protection of a vaccine. Furthermore, diversity studies offer important insight on pneumococcal biology, epidemiology and pathogenesis. The diversity in the toxin pneumolysin (Ply) was greater than previously thought, with 14 protein alleles discovered. However, diversity remained significantly lower than surface-exposed virulence factors, indicating this toxin may be more suitable as a vaccine candidate. Furthermore, all 14 alleles were recognised by polyclonal antibodies, showing the potential cross-serotype protection of a vaccine targeting this toxin. A novel non-haemolytic Ply allele was associated with clones recently expanding in the pneumococcal population, as well as serotypes associated with outbreaks of pneumococcal disease. The non-haemolytic toxin may therefore play a role in driving clonal expansion in certain genetic backgrounds, or be involved in establishing outbreaks of pneumococcal disease. The diversity in the neuraminidase A (NanA) enzyme was significantly higher than in Ply, with many point mutations, mosaic blocks and insertions regions present in 18 divergent alleles. This level of diversity should not be prohibitive to the use of this protein as a vaccine candidate, as polyclonal antibodies recognised 4 NanA alleles of significant diversity, indicating the possibility of cross-serotype protection. The role of NanA in pathogenesis of pneumococcal haemolytic uraemic syndrome (p-HUS) was investigated, although there was no correlation between p-HUS and NanA allele or activity. The novel discovery that pneumococcal NanA was inhibited by viral neuraminidase inhibitors of influenza allowed insight into the synergistic relationship between these two deadly pathogens, and showed for the first time that treatment with these drugs acts on both the primary and secondary pathogen. One of these inhibitors, Oseltamivir, was found to have potential in treating secondary pneumococcal pneumonia, which may help decrease the significant burden of this disease, as well as reducing the over-reliance on antibiotics for treating pneumococcal diseases. Homologues of Ply and NanA were identified and characterised in the related species Streptococcus mitis and Streptococcus pseudopneumoniae, giving insight into the evolutionary relationships between these species. Furthermore, the presence of these homologues in related species gives rise to the possibility of pneumococci acquiring altered genes through horizontal gene transfer. The selective pressure of a vaccine targeting these proteins may give evolutionary advantage to these pneumococci, resulting in evasion of a vaccine. Microarray studies have been used to assess pneumococcal diversity at a genome-wide level. Gene expression studies identified candidate genes which may play a role in p-HUS pathogenesis. Further studies into this area will improve the diagnosis and treatment of this disease. Furthermore, a large number of established pneumococcal virulence factors, many of which are vaccine candidates, were found to have homologues in closely related commensal species. These results must be taken into consideration for future protein-based pneumococcal vaccine design.