Includes abstract. Includes bibliographical references.

Bibliography: leaves 133-156. / Cervical cancer is considered an important clinical problem in sub-Saharan Africa. Recent studies have suggested that epithelial tumors may be regulated by cyclooxygenase enzyme products. The purpose of this thesis was to determine the expression, localisation and possible functional role of cyclooxygenase enzymes in cervical carcinomas. The initial aim of the study was to determine whether cyclooxygenase-1 and cyclooxygenase-2 expession and prostglandin E₂ synthesis are up-regulated in cervical cancers. Real-time quantitative reverse-transcriptase polymerase chain reaction and Western blot analysis confirmed cyclooxygenase-1 and cyclooxygenase-2 ribonucleic acid and protein expression in all cases of squamous cell carcinoma and adenocarcinoma investigated. In contrast, minimal expression of cyclooxygenase-1 or cyclooxygenase-2 was detected in histologically normal cervix. Immunohistochemical analyses localised the site of cyclooxygenase-1 and cyclooxygenase-2 expression and prostaglandin E₂ synthesis to neoplastic epithelial cells of all squamous cell carcinomas and adenocarcinomas studied.

The mycobacterial DNA damage and repair pathways involved in the emergence of drug-resistance during host infection remain poorly understood, yet are critical to any efforts to develop novel "anti-evolution" drugs aimed at reducing the capacity of Mycobacterium tuberculosis to adapt genetically during tuberculosis (TB) treatment. The thesis presented here aimed to investigate the contribution of the DNA damage (SOS) response in adaptive mutagenesis, and focused on two specific components: the role of the specialist translesion synthesis DNA polymerase, DnaE2, in mutagenesis under stress and, secondly, the function of the mycobacterial homologue of a putative SOS response associated peptidase (SRAP) protein which has been identified in comparative genomics analyses of organisms possessing a DnaE2-type C family DNA polymerase. This work focused on the putative SRAP protein which was predicted to form part of the mycobacterial DNA damage response as a functional switch by binding DNA in an autoproteolytic dependent manner. To this end, SRAP deletion mutants were generated for both M. smegmatis (MSMEG_1891) and M. tuberculosis (Rv3226c). Despite the fact that SRAP was upregulated in both M. smegmatis and M. tuberculosis following genotoxic stress, no DNA damage phenotype was detected in any SRAP deletion mutant using a variety of DNA damaging agents. In parallel, an eGFP-tagged M. smegmatis SRAP allele was constructed to enable visualisation of SRAP upregulation and sub-cellular recruitment using fluorescent microscopy; however no eGFP expression could be visualised after MMC treatment. It was not clear whether this was due to faulty eGFP expression in the fusion protein, or to low-level induction of SRAP. In a biochemical approach to elucidate SRAP function, soluble M. smegmatis SRAP protein was expressed and purified using a N-terminal hexa-histidine tag. No proteolytic activity was detected in gelatine or casein zymography, perhaps indicating that SRAP has a very specific substrate. Moreover, while it was predicted that autocatalytic cleavage of the C-terminus was required for activation of SRAP, no such cleavage was detected using hexa-histidine tag staining, possibly pointing to a set of very specific conditions for activation. In combination, therefore, neither microbiological nor biochemical assays could elucidate a definitive role for SRAP in the mycobacterial DNA damage response. DnaE2 has been directly implicated in induced mutagenesis to rifampicin (Rif) resistance in Mycobacterium tuberculosis following exposure of bacilli to genotoxic stress. In previous work in our group, a vitamin B₁₂-sensitive ΔmetH strain was found to form "B₁₂-resistant" suppressor mutants at a frequency higher than could be explained by spontaneous mutagenesis alone. The first part of this thesis investigated the potential role of DnaE2 in the high-frequency emergence of B₁₂-resistance by mutating DnaE2 in the ΔmetH background. Whereas elimination of polymerase function in a DnaE2ᴬᴵᴬ mutant abrogated DNA damage-induced mutagenesis to Rif resistance, no change in B₁₂ sensitivity was detected in a ΔmetH dnaE2ᴬᴵᴬ double mutant. PCR sequencing of spontaneous B₁₂-resistant mutants revealed mutations in genes previously associated with the suppressor phenotype; moreover, there was no apparent difference in the nature of mutations observed in both parental and dnaE2ᴬᴵᴬ mutant strains. Instead, these results suggest that an alternative mechanism must exist to enable adaptive mutagenesis in methionine-starved mycobacteria.

Includes bibliographical references.

Background An effective HIV vaccine remains the main hope for controlling the HIV epidemic and is a global health priority. The genetic diversity of the virus across the globe is a major impediment to developing an effective vaccine. Whether a universal vaccine is possible still remain elusive. Therefore, there is need to fully characterise clusters of commonly targeted regions across the different HIV-1 clades. Centralized sequences have been suggested as vaccine immunogens and peptide reagents for assessing vaccine-induced responses, but their cross-reactivity has not been fully assessed in larger cohorts of subtype C-infection and in regions of differing HIV epidemics. In addition, the functional profile of HIV-specific T-cells recognizing variant epitopes has not been fully characterized. Whether cross-reactivity observed by IFN-γ production in an ELISpot assay can be observed at physiological concentrations of the peptides and for other functions of HIV-specific T-cells is an important question that remains to be answered. Methods The cross-reactivity of HIV-specific T-cells was assessed using clade-specific peptide reagents forming part of current candidate vaccine inserts based on the HIV-1 Gag protein from clades CDu422, CCH, A, B and D in 40 clade C-infected study participants using the IFN-γ ELISpot assay. To test the reactivity of group M consensus peptide reagents, 66 individuals, 44 of whom were ARV naïve, were assessed for HIV-specific T cell responses to group M Gag and Nef peptides. A selection of these individuals was screened for HIVspecific T-cell responses to clade CDu422 Gag peptides. Cross-reactivity of peptide variants was assessed at physiologically relevant peptide concentrations by functional avidity studies using peptide dilution IFN-γ ELISpot assays. Additionally, the cytokine profile, cytotoxic potential and proliferative capacity of cross-reactive peptide variants was characterised using multiparameter flow cytometry. Results The magnitude and breadth of HIV-specific T-cell responses were similar between the two clade-C peptide reagents in a clade C-infected population. However, the magnitude and breadth of responses to peptides based on clades A, B and D were significantly lower compared to the clade C peptides. Clusters of commonly targeted regions cross-reactive across the four clades investigated resided predominantly in conserved regions. Interestingly, there were Gag regions that were exclusively recognized in the different clades that had significantly lower entropy scores for the reactive variants than their nonreactive counterparts, suggesting that the variability in targeted regions could have been shaped by host immune pressure. For consensus group M peptides, the magnitude and breadth of Gag responses were significantly higher than that of Nef in clade C-infected individuals. In addition, consensus group M Gag peptides had significantly lower magnitude and breadth of HIV-specific T-cell responses compared to clade C peptide reagents, suggesting loss of responses by centralised reagents despite their central nature to group M viruses. On the contrary, the magnitude and breadth of responses to consensus group M Gag peptides were comparable to that of clade-mismatched peptides, namely clades A, B and D, suggesting that these reagents can be used interchangeably. Peptide dilution assays showed that amino acid mismatches have discordant effects on functional avidity and that some peptides are cross-reactive at physiological concentrations. Similarly, discordant effects (differences in functional avidity, cytokine and cytotoxic profiles and proliferative capacity) of amino acid mismatches on cytokine and cytotoxic potential profiles as well as proliferative capacity were observed. Conclusion People infected by a particular HIV clade can recognize HIV peptides based on other clades. However, the magnitude and breadth of responses are greater for the matched clades compared to mismatched clades, suggesting that there may be an advantage of using vaccines based on matched over unmatched clades. Group M based consensus sequences can be recognized in HIV-infected individuals, but with a lower frequency, magnitude and breadth of responses compared to clade-matched peptides, suggesting a limitation of these peptides both as reagents and vaccine immunogens. However, the frequency, magnitude and breadth T-cell reponses to consensus group M peptides were comparable to clademismatched responses, suggesting that these reagents may be used interchangeably. Furthermore, amino acid variations across corresponding viral regions have discordant effects on the functional avidity, cytokine profile, cytotoxic potential and proliferative capacity; implying that qualitative measures of cross-reactivity beyond IFN-γ frequencies need to be considered. These data may aid in the development of reagents for the assessment of vaccine-induced responses as well as in HIV vaccine immunogen design.

The emergence of antibiotic resistant strains of Mycobacterium tuberculosis, coupled with the time it takes to perform phenotypic drug susceptibility testing of this organism, makes the treatment of tuberculosis increasingly difficult. Several genotypic assays for the rapid detection of drug resistance in M. tuberculosis have been developed, but the sensitivity with which these assays identify resistance differs geographically. Additionally, the identification of phenotypically resistant isolates with no identifiable genotypic marker suggests that other factors, such as differential gene expression, may play a role in the development of drug resistance in M. tuberculosis. This investigation aims to both develop and evaluate rapid genotypic assays for the detection of resistance to both first- and second-line drugs in M. tuberculosis, and to investigate the role of alternative sigma factors in the progression to multidrug resistant M. tuberculosis.

Development of effective and affordable HIV vaccines is one of the best and cost-effective strategies for controlling the HIV epidemic and a top priority in endemic areas. Successful future candidate HIV vaccines are expected to elicit effective antibody and T cell-mediated responses. This is envisaged to be attained through induction of potent T cell-mediated immune responses to control viral replication in the tissues and disease progression as well as a durable antibody immune response which comprises of broadly neutralizing antibodies to block virus entry at the mucosal sites. Both types of immune responses are influenced by a T helper cell type 1 (Th1) immune response. In developing worlds such as Sub-Saharan Africa co-infections of HIV and schistosomiasis are common. Helminth infections such as schistosomiasis induce strong Th2 biased immune responses that have been reported to alter HBV, BCG, Tetanus toxoid and some candidate HIV vaccine-specific immune responses. Because Th1 and Th2 are almost mutually exclusive, it is suggested that, in the presence of chronic helminthic infections, Th1 responses elicited by HIV vaccine may be attenuated, hence, reduce vaccine efficiency. On the other hand, vaccination with an effective HIV vaccine might shift the immune bias towards a Th1 response, resulting in worsening of the helminth-associated pathology, thus, making the vaccine unsafe to the recipients. This study aimed at investigating if chronic helminth infection (Schistosoma mansoni: Sm) has a negative impact on the immunogenicity of HIV vaccine candidates (SAAV DNA-C2, SAAVI MVA-C, and Env gp140 protein) previously shown to induce cellular, mixed and antibody immune responses in mice. The objectives of this study were to (i) infect mice with live Schistosoma mansoni infection in order to induce a predominantly Th2 immune response in a mouse model; (ii) evaluate if helminth-induced Th2 immune biasing negatively affects responses to candidate HIV vaccines; (iii) evaluate the ability of ant ihelminthic chemotherapy in restoring normal responses to HIV vaccines in helminth infected mice; (iv) evaluated if HIV vaccine that predominantly induces strong cellular immune responses result in worsening of the helminth-associated pathology and (v) evaluate if helminth eggs in the absence of live helminth infection drives a Th2-dominant response that can affect HIV vaccine responses. The BALB/c mouse model has been used extensively at the University of Cape Town for studying the Smassociated immunology as well as for initial evaluation of candidate HIV vaccines. Female BALB/c mice were either chronically infected with Sm cercariae or inoculated with Sm eggs (SmE) before being subsequently vaccinated twice, 4 weeks apart with three vaccination regimes that elicit cellular (SAAVI DNA-C2 prime + MVA-C boost denoted: DNA+MVA), antibody (gp140 Env protein) and mixed cellular and antibody (SAAVI MVA-C prime + gp140 Env protein boost denoted: MVA+gp140) responses. Some groups of mice infected with live Sm were treated twice with praziquantel (PZQ) prior to vaccination. Spleens, blood and livers were harvested for analysis of vaccine-specific T cell, antibody responses and histological studies using ELISpot, ELISA, CBA, ICS staining, H&E/CAB staining and hydroxyproline assay. Our findings demonstrated that in a mouse model, chronic Sm-infection induces a predominantly Th2 immune biased response marked with elevated parasite-specific IL-4; IL-6 and IL-10 as well as elevated total IgG1 and IgM, while resulting in decreased Th1 markers. Furthermore, chronic infection significantly inhibited cellular responses to the MVA+gp140 vaccine regimen shown by IFN-γ and IL-2 ELISpot; CBA and ICS staining. Similarly, in DNA+MVA vaccinated mice, a significant reduction in vaccine-specific responses was observed in Sm-infected groups compared to uninfected vaccinated groups shown by IFN-γ and IL-2 ELISpot; CBA and ICS against HIV immunogens. antihelminthic treatment with PZQ resulted in the partial restoration of Th1-Th2 balance in the Sm-infected hosts, with the levels of vaccine-induced IFN-γ; TNF-α and IL-2 being partially restored despite the presence of elevated Th2 cytokines after treatment with PZQ. A significant overall decrease in Env gp140 specific IgG, IgG1, IgG2a and IgG2b antibody responses was observed in the Sm-infected mice vaccinated with gp140 or MVA+gp140 vaccine regimen compared to uninfected vaccinated controls. Surprisingly, antihelminthic treatment did not restore vaccineinduced antibody responses. Our histology data showed that DNA+MVA vaccinated mice develop increased liver pathology during chronic schistosomiasis compared to unvaccinated Sm-infected groups shown by larger livers; spleen and enlarged granuloma formation. However, no significant difference in collagen content (a marker of fibrosis) measured by hydroxyproline assay in both vaccinated and unvaccinated infected groups was observed. Our findings further demonstrated that in a mouse model, inoculation with SmE also induces a predominantly Th2 immune biased response marked with elevated parasite-specific IL-4; IL-6 and IL-10 while resulting in decreased Th1 markers. No significant impact on cellular responses evaluated by ELISpot and CBA were observed. However, gp140 specific IgG, IgG1 IgG2a and IgG2b antibody responses were significantly reduced in groups challenged with SmE. Overall, these findings show that chronic helminthiasis in the mouse model induces a strong Th2 biasing which is associated with attenuation of both T cell and antibody response to HIV vaccines. Elimination of helminths by chemotherapy may partially restore T cell responses, but not necessarily antibody responses. These findings further suggest that vaccinating helminth infected individuals with HIV vaccines that induce strong cellular responses may increase the pathology induced by the parasites, rendering the vaccine unsafe in helminth endemic areas. Furthermore, this study suggested that in the absence of an active chronic Sm-infection, SmE left trapped in the tissues following antihelminthic treatment, may continue to induce strong Th2 responses which are capable of downregulating vaccine-specific responses, especially the antibody-mediated responses. This study strongly recommends that development of HIV vaccines should also focus on designing vaccines that can overcome helminth-induced immunity.

BisPMB (E, Z)-1,8-(Bis-p-methoxyphenyl)-2,3,7-trithiaocta-4-ene 7-oxide) is a synthetic analogue of the garlic compound ajoene. It is 12 times more active at inhibiting the growth of oesophageal squamous cell carcinoma WHCO1 cells and displays selectivity for cancer cells over normal cells. BisPMB is therefore attractive as a potential cancer therapeutic. In this study, bisPMB was found to inhibit WHCO1 cancer cell proliferation in a time and concentration dependent manner with 24 hour IC50's between 6.7 - 8.1 μM against a range of oesophageal cancer cell lines including WHCO1, KYSE30 and WHCO6. The normal oesophageal epithelial cell line, HET1A was found to be five times less responsive to bisPMB. Furthermore, bisPMB was found to induce apoptosis and G2/M cell cycle arrest in WHCO1 cells. Gene expression data obtained from the microarray analysis showed that bisPMB primarily targets the unfolded protein response (UPR) in WHCO1 cells. We also found that bisPMB deregulated the ER stress genes involved in protein processing in the endoplasmic reticulum and also deregulated MAPK pathways in WHCO1 cells. At a protein level, bisPMB was found to induce an increase in protein ubiquitination and in the expression of ER stress and UPR genes ATF4, Grp78 and CHOP in WHCO1 cells. We also observed a decrease in ATF6 90 kDa protein and transient XBP-1 mRNA splicing. The activation of p38, JNK and ERK MAPK pathways in bisPMB treated WHCO1 cells was also observed. Furthermore siRNA mediated knock-down of CHOP abolished the anti-proliferative effect of bisPMB in WHCO1 cells. However, inhibition of JNK and p38 MAPK by chemical inhibitors, SP600125 and SB 203580 respectively, had no effect on bisPMB antiproliferative activity against WHCO1 cells. On the other hand, inhibition of ERK1/2 MAPK by U0126 enhanced the anti-proliferative effect of bisPMB in WHCO1 cells. These results support the hypothesis that ER stress and MAPK signalling pathways are essential for bisPMB induced cytotoxicity in oesophageal cancer cells.

Includes abstract. Includes bibliographical references (leaves 139-162). / Most HIV vaccines currently in development aim to protect people from infection or disease by eliciting strong anti-HIV cytotoxic T lymphocyte (CTL) responses. Evolved evasion mutations that undermine host immune responses pose a major challenge to the development of such vaccines. Understanding the mechanisms that selectively favour the emergence of CTL evasion mutations in vivo and the impact of these mutations on both disease progression and long-term HIV evolution will not only contribute to our understanding of HIV pathogenesis, but will also inform vaccine design strategies. This study aimed at investigating CTL escape mutations in HIV-1 Gag and Nef, during the acute and early phases of infection and the impact of these mutations on subsequent disease progression in a cohort of recently HIV-1 subtype C infected females. Of 36 women recruited into the study within 12 weeks of infection (median 6 weeks) and followed for six months, 32 were infected with single viruses. Two participants were infected with epidemiologically unlinked viruses (dual infection), and in a further two individuals the viruses were highly divergent suggestive of dual infection and/or recombination. These individuals were excluded from further analysis as it was difficult to predict CTL escape due to high degrees of diversity between sequences. In the remaining 32 study participants, there was a high frequency of CTL escape with putative escape mutations identified in 21 of 32 individuals (66%). Twelve of these 21 (33%) harboured viruses which developed escape mutations in Gag, and 17 (53%) developed escape mutations in Nef. In the conserved structural protein, p24, potential reversion mutations were more frequent than potential escape mutations. During the first six months of infection whereas potential reversion mutations occurred at low entropy sites, potential escape mutations occurred at high entropy sites. Although there was no detectable association between the timing of escape mutations and disease progression, there was an association between the degree of deviation of the p24 sequence from the subtype-C population consensus (a measure of escape mutation load) and CD4+ counts. Analysis of the earliest sampled viruses from HLA-B*57/B*5801 negative study participants for viral genetic markers associated with disease progression identified two iv polymorphisms, A146X (n = 9) and T242N (n =6), that were associated with improved viral control. The polymorphisms are well-known escape mutations in HLAB* 57/B*5801 restricted epitopes. This suggested transmission of these variants from individuals carrying these alleles. Further evidence that viruses carrying the T242N and/or A146X mutations had been previously passaged through B*57/B*5801 positive individuals came from the fact that the observed T242N mutations reverted to wild type during follow-up. There was no significant change in viral load and CD4+ counts upon reversion of the T242N mutations. In vitro replication assays using chimeric viruses containing gag sequences from one of participants showed that the virus harbouring the T242N mutation was fitter than that carrying the reversion mutation. These viruses harboured other T242N associated compensatory mutations suggesting that these compensatory mutations may themselves carry a fitness cost in the absence of the T242N mutation. This suggests that there possibly exist networks of B*57/B*5801 associated mutations and that reversion of some of these mutations in isolation does not necessarily restore viral fitness. Lastly, the kinetics of CTL escape in HLA-B*5801 positive participants (n = 6) and the impact of escape on disease progression was investigated. CTL escape within B *5801 positive individuals was found to predominantly occur within the TW10 in Gag (n = 4) and KAF9 in Nef (n = 6) epitopes. The emergence of the T242N mutation in TW10 was always preceded by mutations elsewhere in the epitope and was associated with the occurrence of previously described compensatory mutation upstream of the epitope. The targeting of TW10 and the emergence of T242N escape mutations were associated with higher CD4+ counts at 12 months postinfection in the B*5801 positive individuals (p = 0.0231 and p = 0.0282, respectively). Independent of host HLA genotypes, the presence of the A146X and T242X mutations was associated with higher CD4+ counts (p = 0.0495). This study provides some useful insights into HIV-1 subtype C pathogenesis. The notion that CTL escape mutations do not invariably result in less fit viruses is evidenced by the observation that escape was not obviously associated with disease progression in this cohort, while escape mutations in the Gag p24 region within B*5801 positive individuals v in particular, was associated with improved viral control. There is therefore evidently a complex interaction between escape and compensatory mutations and further work is required to identify the impact of compensatory mutations on viral fitness. Overall, this study provides further evidence that vaccines need to elicit responses that specifically target the functionally constrained regions of the HIV proteome.

Includes bibliographical references (leaves 109-130). / Although the incidence and mortality due to the oesophageal squamous cell carcinoma (OSCC) in Black South Africans is extremely high, very little is known about the aetiology and molecular biology of the disease. In order to make a contribution to the understanding to the causes of this disease we investigated the role of the polymorphisms in the genes coding for the cytochrome P450 (CYP1A1, CYP1A2, CYP1B1, CYP2E1), sulphotransferase 1A1 (SULT1A1), glutathione S-transferases (GSTT1. GSTM1 and GSTP1) alcohol dehydrogenase (ADH2 and ADH3) and aldehyde dehydrogenase (ALDH2) because the products of these genes are involved in the metabolism or biotransformation of harmful compounds.