Thesis (Ph. D.)--University of Iowa, 2008. / Thesis supervisors: Herbert W. Hethcote, Tong Li. Includes bibliographical references (leaves 105-110).

Thesis (Ph. D.)--University of Texas at Austin, 2007. / Vita. Includes bibliographical references.

Thesis (Ph. D.)--University of Washington, 1998. / Vita. Includes bibliographical references (leaves [135]-143).

A research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in partial fulfillment of the requirements for the degree of Master of Science in Epidemiology in the field of Epidemiology & Biostatistics January 2014 / Studies have shown an increased risk of mortality amongst adults with invasive pneumococcal disease (IPD) with certain serotypes but there are no data from South Africa. We aimed to determine the association between serotypes and in-hospital mortality among adults aged 15 years and older with IPD in South Africa. Methods IPD cases were identified through the GERMS-SA national laboratory-based surveillance programme. Patient data from 25 enhanced surveillance sites from 2003-2008 (pre-pneumococcal conjugate vaccine introduction) with available data on serotype and in-hospital outcome were used. We assessed the association between the 20 most common serotypes and mortality among patients ≥15 years of age using univariate and multivariable logistic regression models. Results From January 2003 through December 2008, there were 3953 cases of IPD amongst adults older than 15 years of age meeting the study inclusion criteria. Amongst the 20 commonest serotypes, the incidence of serotypes 4, 19A, 23F and 18C increased significantly, and serotypes 1, 25 and 5 decreased significantly from 2003 to 2008. Serotype 1 was the commonest serotype overall (16%, 651/3953), followed by serotype 19A (11%, 443/3953) and serotype 4 (7%, 289/3953). The case-fatality ratio (CFR) was 55% (641/1166) for meningitis and 23% (576/2484) for bacteremia (p<0.001). Serotype 19F had the highest CFR of 48% (100/207), followed by 39% (99/252) for serotype 23F and 38% (246/651) for serotype 1. On multivariable analysis, factors independently associated with mortality were disease caused by serotypes 1 (OR 1.93, 95% CI 1.05–3.53) and 19F (OR 2.89, 95% CI 1.38–6.06) compared to serotype 4; increasing age (25-44 years, OR 1.75, 95% CI 1.03–2.95; 45-64 years, OR 3.56, 95% CI 2.00–6.35; ≥65 years, OR 5.17, 95% CI 1.89–14.14; compared to 15-24 years); living in provinces with intermediate (OR 1.65, 95% CI 1.16–2.35) or high poverty rates (OR 1.72, 95% CI 1.02–2.92) compared to provinces with low poverty rates; having meningitis (OR 4.07, 95% CI 2.98–5.55) compared to bacteremia; prior antibiotic treatment in the last two months (OR 3.93, 95% CI 2.50–6.20); inappropriate antibiotic treatment (OR 2.37, 95% CI 1.74–3.22) and positive HIV status (OR 1.69, 95% CI 1.04–2.75). Conclusion Serotypes associated with increased mortality are included in the 10-and-13-valent pneumococcal conjugate vaccine and may be expected to become less common in adults as a result of indirect effects following routine immunization in infants. HIV-infected adults experience increased mortality and the more widespread availability of antiretroviral therapy is likely to substantially improve the quality of life of HIV-infected individuals in terms of physical and mental health and decrease the incidence of IPD and therefore mortality.

The immune response to infection with Mycobacterium tuberculosis (Mtb) involves complex interactions between macrophages, T-cells, cytokines and accessory molecules. Mycobacteria evade the host’s immune response by interfering with cell mediated immune systems. Granulomas are central to the host’s defenses against Mtb. These responses may be modified by immune alterations especially in patients co-infected with Human Immunodeficiency Virus (HIV). This study investigated the immunohistochemical profile of CD4+, CD8+, CD68+, Th-17 (also known as Interleukin-17 cells) and Forkhead box (FOXP3) cells, accessory molecule expression (HLA Class I and II) and selected cytokines (Interleukin 2, 4, 6 and Interferon-) of various cell types within mycobacterial granulomas, in lymph nodes from ten HIV negative and ten HIV positive patients. Tissue from a foreign body granuloma in skin was utilised for comparison. This study illustrated retention of CD4+ lymphocyte numbers within granulomas from HIV negative (-) patients but documented a reversal in the ratio of CD4+ to CD8+cells in granulomas from HIV positive (+) patients. Similar IL-17 cell counts were noted in mycobacterial granulomas from both HIV (-) and HIV (+) patients. CD68 was identified in all macrophages and HLA Class II stained 100% of cells. Mycobacterial granulomas from HIV (-) patients showed marginally lower numbers of HLA Class I cells when compared to those from HIV (+) patients. The percentage of FOXP3 positive cells differed significantly between mycobacterial granulomas from HIV (-) and HIV (+) patients. This study highlights the complex interplay between different cell types and cytokines secreted into the microenvironment that ultimately results in containment of the organism or disease progression. Tuberculosis mono-infection causes variation in the expression of cell markers such as FOXP3 with accompanying noteworthy changes in cytokine production in areas of granuloma formation. The alterations noted in TB and HIV co-infection are even greater and point toward evolution of micro-organism synergism with host demise.

MMed (Otorhinolaryngology), Faculty of Health Sciences, University of the Witwatersrand / The submandibular space is part of the deep neck fascial spaces. Infection within these spaces can cause significant mortality and morbidity. The infections are usually opportunistic, being commensal micro-organisms from the oral cavity and oro-pharynx. The study is a review of the microbiology of the submandibular space infections seen at the tertiary academic referral hospitals, University of the Witwatersrand. The period of the study is from the 1st January 2006 to 31st December 2006. 93 patients were reviewed of which 52 had aspirates taken. The predominant micro-organisms were the Gram positive anaerobic cocci. Their resistance and sensitivity pattern was also studied. A literature search for this category of infection was also conducted. The most commonly isolated micro-organisms were the gram positive cocci.

A dissertation submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree Master of Science. Johannesburg 2015 / The controversy surrounding the findings that copy number variation (CNV), of the CCL3 encoding genes, influences HIV-1 infection and disease progression has been in part attributed to the variable results obtained from methods used for copy number evaluation. Like CCL3, the genes encoding the CC chemokine CCL4, also a natural ligand of the CCR5 receptor, are found to occur in population-specific multiple copy number and have been shown to play a protective role against HIV-1. In this study, we evaluated the standard method of quantitative Real-Time PCR (qPCR) and Droplet Digital PCR (ddPCR) for CCL4L gene copy number determination. The CCL4 encoding genes are CCL4, occurring in two copies per diploid genome (pdg), and the non-allelic CCL4L genes, comprised of CCL4L1 and CCL4L2, which are both found in multiple copies pdg. Copy number of CCL4L, CCL4L1 and CCL4L2 was determined in a cohort of HIV-1-uninfected individuals from the South African Black (n=23) and Caucasian (n=32) population groups using qPCR and ddPCR, with the addition of another 30 black individuals to the ddPCR cohort. A stronger correlation between the number of CCL4L copies and the sum of CCL4L1 and CCL4L2 copies generated by ddPCR (r=0.99, p<0.0001) compared to qPCR (r=0.87, p<0.0001) was observed. Real-Time qPCR exhibited greater inaccuracy at higher copy numbers which is particularly relevant to our cohort of Black individuals who have a higher range of CCL4L copies (3-6) compared to Caucasians (0-4) and a higher population median (4 and 2, respectively). Medians and ranges of CCL4L1 (Black: 2, 0-4, Caucasian: 0, 0-2) and CCL4L2 (Black: 2, 1-5, Caucasian: 2, 0-3) were also higher in the Black population. Droplet Digital PCR was shown to be a far superior method to qPCR for assessment of CCL4 gene copy number variation, the accuracy of which is essential for studies of the contribution of variable gene copy number to phenotypic outcomes of host infection and disease course. We further used the CCL4L copy number data to examine variation of these genes with reference to measured stimulated CCL4 production by the same individuals. Although significant differences in the copy number range medians and patterns of distribution of the genes CCL4L1, CCL4L2 and combined CCL4L were observed between the two populations, on a whole, the two populations do not differ significantly with respect to CCL4 production. Caucasian females however had a higher level of protein production per copy of CCL4L than Black females. When stratifying production based on population specific copy number median, Black individuals showed a decreased level of protein production at a CCL4L copy number below the median, although this was not maintained when the CCL4L1 and CCL4L2 genes were analysed individually. CCL4 copy number and production data was compared to data generated for CCL3 in the same cohort. CCL4 chemokine levels were significantly higher in both the Black and Caucasian populations and Black individuals had a higher number of gene copies of the CCL3L genes giving rise to functional CCL3 protein than the CCL4L genes producing functional CCL4 protein. These results highlight genetic differences between divergent populations, differences in distribution of CCL4 and CCL3 encoding genes and protein production, and suggest an intricate regulation of the CCL4L encoding genes. In order to investigate the role of CCL4 in HIV-1 control, we next assessed variation in the numbers of CCL4L copies in relation to CCL4 production in a cohort of 14 long-term nonprogressors (LTNPs). While no associations between copy number and CCL4 production were observed, the LTNP cohort had significantly lower levels of CCL4 production than the HIV-1-uninfected cohort, and this was maintained when Black and Caucasian individuals were examined individually. When the LTNP cohort individuals were divided based on viral loads, individuals with viral loads <400 RNA copies/ml had significantly lower CCL4 production than those with viral loads >400 RNA copies/ml. This finding suggests a role for the amount of CCL4 produced in the reduced pace of HIV-1 progression observed in LTNP individuals. Since genetic variation other than copy number can also influence CCL4 protein production, we then proceeded with a thorough genetic characterization of the CCL4 gene of uninfected individuals and investigated any possible relationships with select variants and CCL4 production. Sequencing the complete gene and flanking regions of 23 Black and 32 Caucasian revealed several intra as well as extragenic SNPs, with one newly identified SNP at position -1063. The Black population exhibited a higher degree of variability compared to the Caucasian population. We described four haplotypes in the Caucasian population, three haplotypes in the Black population, and a single haplotype shared between both population groups. Of the four indels that were identified, a three bp deletion (rs3216921) was the only indel present in the Caucasian population and was not identified in the Black population. Of all the indels, this indel had the highest allelic frequency (14%). Comparisons of haplotypes and prevalent SNPs with protein production in both population groups did not show any significant differences. Caucasian individuals harbouring the 3bp deletion however, had significantly higher levels of CCL4 production (p=0.024). These results form a good base from which to further investigate the impact of select genetic variants on CCL4 production and possibly HIV-1 control. This study has succeeded in optimising a ddPCR assay for the copy number determination of the CCL4 genes and has interrogated the relationship between CCL4L copy number and CCL4 production in both HIV-1-uninfected individuals as well as a subset of LTNPs. The results suggest a complex, intricate regulation of CCL4 that appears to play a role in HIV-1 control. In conclusion, this study forms the basis for future work to build on and to further explore the role of this CCR5 ligand in HIV-1 disease.

Thesis (Ph.D.(Physiology))--University of the Witwatersrand, Faculty of Health Sciences, 2012. / The acute phase response is implemented by infected hosts in response to exposure to pathogens, including bacteria and viruses. Acute phase responses comprise physiological and behavioural changes, such as fever and a range of “sickness behaviours”, including lethargy and anorexia as well as impairment in learning and memory. Similar to other sickness behaviours, the effect of infection on learning and memory processes has been attributed to the release of pro-inflammatory cytokines, including interleukin-1β (IL-1β) and interleukin-6 (IL-6). However, the exact role of IL-1β and IL-6 in mediating infection-induced cognitive impairment is not clear. Unlike fever, anorexia and lethargy, which may benefit an infected host, the physiological benefit of cognitive impairment during illness is doubtful. To initiate an acute phase response experimentally, moieties of typical bacteria (Gramnegative and Gram-positive) and viruses frequently are employed. Moieties from the atypical Mycoplasmas seldom have been used. Consequently, there is a dearth of information on the physiological mechanisms that underlie acute phase responses following Mycoplasma infection, despite the prevalence of the disease in the general population. Mycoplasma pneumoniae frequently causes community-acquired pneumonia, which may have serious extra-pulmonary complications, including cognitive deficits. Therefore, I investigated fever and sickness behaviours as well as cytokine responses in simulated, atypical Mycoplasma infection. I implemented an animal model of simulated Mycoplasma infection and characterised fever and sickness behaviours, including lethargy and anorexia as well as impairment in learning and memory during acute and recurrent acute simulated infection. I also characterized the response in the periphery and in the brain of individual pro-inflammatory cytokines, IL-1β and IL-6, to administration of fibroblast-stimulating lipopeptide-1 (FSL-1), which simulates Mycoplasma infection. Using rats, I recorded fever and lethargy with biotelemetry and assessed effects of simulated Mycoplasma infection on learning and memory using a Morris Water Maze. In addition, I examined the histology of tissue from the hippocampus, a key brain area involved in spatial learning and memory, to assess residual effects of simulated Mycoplasma infection on learning and memory. I showed that bolus administration of a pyrogenic moiety from Mycoplasma, fibroblaststimulating lipopeptide-1 (FSL-1), dose-dependently induced fever, lethargy, anorexia and body mass stunting in rats. However, FSL-1 administration did not induce concomitant impairment in spatial learning and memory. Importantly, at the time of testing in the Maze, I found the concentrations of IL-1β to be up-regulated in both the hypothalamus and the hippocampus, while the concentrations of IL-6 were unaffected. I also showed that recurrent acute injections of FSL-1, at 10 d intervals, induced recurrent fevers, lethargy and anorexia without the development of pyrogenic tolerance to any of the sickness responses measured. However, there was no residual body mass stunting in rats and also no growth retardation, despite the recurrent simulated infection. Equally importantly, there were neither lasting detrimental effects on spatial learning and memory nor any residual histological damage to the hippocampus of rats. My findings in simulated Mycoplasma infection are important, firstly because Mycoplasma infection is prevalent in both developing and developed countries and frequently causes outbreaks, and secondly because Mycoplasma infection affects children and adolescents of school-going age. My findings also are encouraging: although lasting detrimental effects, including impairment in learning and memory as well as body mass stunting may occur in other infections, these appear not to be inevitable outcomes in Mycoplasma infection.

"The C. albicans community is currently laying the foundation of understanding how this human pathogen causes infection. C. albicans infections represent a major medical and economic burden for today’s society with an estimated 400,000 blood stream infections worldwide and direct costs exceeding 1$ billion dollar a year in the U.S. alone. Although finding the biological causes of this disease seemed to be beyond our reach in the past, various aspects of the infection have been recently unveiled including its pathology, immunology, histology, and epidemiology. Here we explored the genetic components of this disease by studying the complex host-pathogen dynamics through a series of in vivo, ex vivo and in vitro experiments. By using a pathogen unbiased reverse genetic approach and a host gene candidate strategy we uncovered some of the genes and pathways that are important for pathogenicity and immunity. In particular we explored the complex host-pathogen dynamics using a C. albicans - C. elegans model system and identified four novel putative virulence factors. We focused on Zcf15, a C. albicans transcription factor that has been poorly characterized in the literature and that plays an important role in the pathogen’s ability to resist host generated reactive oxygen species (ROS). By leveraging the power of RNASeq and ChIP-Seq we identified Zcf15 transcriptional targets and DNA binding sites. These studies suggest that Zcf15 plays a critical role in carbon metabolism and that it exerts its ability to protect the pathogen from ROS by controlling the expression of thiol- peroxidases and other detoxifying enzymes. We also showed here that in C. elegans, the host’s ability to counteract the infection relies on the MAPK pathway, evidence that mirrors what has been found by others in mammals and that emphasizes the usefulness of studying C. albicans infections in smaller genetically traceable organisms like C. elegans. The nematode model is also shown here to be a powerful tool not only to study the genetic bases that drive infection and immunity but also to identify new compounds that can be used for therapeutic intervention. This model was instrumental in identifying filastatin, a small molecule that was subsequently found by our collaborators to be capable of reducing virulence in mammals. The antifungal properties of filastatin are currently undertaking further preclinical testing. Overall this thesis shed light on the complex mechanisms of C. albicans pathogenicity and host immunity and identified novel virulence determinants that can be used by the larger community for further biological studies or even drug development. "

Thèse d'exercice : Médecine. Anesthésie et réanimation chirurgicale : Université de Nantes : 2005. / Bibliogr. f. 44-50 [61 réf.].