Background: Wheezing is one of the leading respiratory symptoms in childhood, which with recurrent occurrence can lead to asthma at school age. Few studies, with conflicting findings, have investigated the contribution of stool bacteria in childhood wheezing illness. Additionally, most of the published studies were conducted in high-income countries using small sample sizes and targeted bacterial detection techniques. To address these limitations, we conducted this study to determine the association of the infant stool bacterial diversity and composition with wheezing development. Methods: We conducted a longitudinal case-control study nested within the Drakenstein Child Health Study, Western Cape, South Africa. We included 140 infants with wheeze (cases) and 140 age-matched controls. Passed faecal samples were collected from infants at birth, whilst aspirated or passed faecal samples were collected at six weeks, six months and 12 months. Deoxyribonucleic acid (DNA) was extracted from all the samples using the manual ZR Fecal DNA MiniPrep™ Kit. Sequencing of the hypervariable V4 region of the 16S rRNA gene was performed using the Illumina MiSeq technique. The resulting sequencing data was subjected to bioinformatic quality control checks and statistical analysis. Results: One of the main findings from the systematic review of wheezing data was the observed association between the development of atopic wheezing and four bacteria namely Faecalibacterium, Lachnospira, Veillonella and Rothia. This study included 280 children (n=123 female and n=157 male) recruited from TC Newman (38.5%) and Mbekweni (61.5%) areas. Stool samples were collected from 159 infants at birth, 114 infants at six weeks, 141 infants at six months and 98 infants at 12 months of age. Aspirate and passed stool samples were found to be similar and comparable in terms of the bacterial composition. At age six weeks, 16% (23/140) of the infants included had wheezed once. During the first year, the proportion of infants with recurrent wheezing (≥ 2 wheezing episodes) was 56% (78/140). Independent factors influencing the composition and diversity of the infants faecal bacteria were feeding practices, age of mothers and living conditions (all p0.05). In addition, we did not find significant difference between infants with and without wheeze for selected bacteria reported to be associated with wheezing. However, we observed an increased relative abundance of Proteobacteria at 12 months of age in infants who had previously wheezed. Additionally, Lactobacillales was in apparently significantly higher proportions in recurrent wheezers as compared to infants who wheezed once (p< 0.05). We did not observe distinct faecal bacterial profiles between infants with and without wheeze at any taxonomic level analysis (all p> 0.05). In addition, we did not find significant difference between infants with and without wheeze for selected bacteria reported to be associated with wheezing. However, we observed an increased relative abundance of Proteobacteria at 12 months of age in infants who had previously wheezed. Additionally, Lactobacillales was in apparently significantly higher proportions in recurrent wheezers as compared to infants who wheezed once (p< 0.05). Conclusion: This study shows that aspirated stool is a good alternative for passed stool in microbiome studies. Furthermore, it revealed the complex and dynamic nature of the faecal bacteria, as well as factors influencing the faecal bacterial profile during the first year of life. More studies in this cohort, including the use of metagenomic and metabolomic approaches are required to demonstrate the role played by bacterial types and their metabolites in the development of wheezing illness. Finally, because recurrent wheezing is one of the precursors of asthma, there is a need for an additional follow-up of these infants in order to investigate the contribution of the early faecal microbiome in the development of asthma at school age.

The response of Mycobacterium smegmatis to a cold shock was investigated in order to gain insight into the stress responses of members of the genus Mycobacterium. Mycobacterium smegmatis cultures were shocked from 37°C to 30°C, 25°C, 15°C, and 10°C and the effects on both growth (ATP concentration, culture turbidity, colony-forming units) and metabolism (incorporation of ¹⁴C-leucine and ³H-uracil) were investigated. The magnitude of the cold-shock response was found to be dependent upon the degree of the cold shock. A cold shock to 10°C had the greatest effect and resulted in a "lag period" of 24 hours in both the growth and metabolism of the culture. The synthesis of proteins was reduced 20-fold during this period, indicating at block in translation. The cold-shock response in Mycobacterium smegmatis was an adaptive response with growth eventually being resumed at the colder temperature, but at a reduced rate. Using the techniques of one-dimensional sodium-dodecyl-sulphate polyacrylamide gel electrophoresis and two-dimensional protein gel electrophoresis, ³⁵S-methiononine-labelled proteins that were synthesised during the cold shock were analysed. At least fourteen radio-labelled proteins were induced during the first 24-hour period and these demonstrated two distinct patterns of cold-shock induced expression: transient and continuous. Depending upon the pattern of expression and size, the cold-shock proteins were classified as "cold-induced proteins", "cold-shock proteins" or "cold-acclimation proteins". CipM, a 27kDa protein, was identified as the major cold-shock protein through one-dimensional protein electrophoresis. From N-terminal sequence data generated from a protein (CipM.1) within this band, a corresponding degenerate DNA probe was used to isolate cipM.1. This gene was cold-inducible, with mRNA levels transiently increasing 5-7 fold after a 37°C to 10°c cold-shock. Homologues of this cold-shock gene are found in the genomes of Mycobacterium tuberculosis and Mycobacterium leprae. The corresponding mycobacterial proteins showed homology at the N-terminus to the HU~ subunit of HU of Escherichia coli and possessed similar C-terminal praline, lysine and alanine degenerate repeats to the mycobacterial heparin-binding hemagglutinin. The response of several mycobacterial cold-shock gene homologues to a cold shock was also investigated, by northern-hybridisation and S1 nuclease analysis. The cspA homologue of Mycobacterium smegmatis demonstrated a 16-24 fold transient induction in mRNA levels following a 37°C to 10°C temperature-shift, while gyrA mRNA levels were maintained at a constant level throughout the cold shock. Although some similarities were demonstrated between the cold-shock response of Escherichia coli and Mycobacterium smegmatis, definite differences occur in the proteins that are involved in the adaptive stages of the response.

As an initial step towards characterisation of the molecular processes that define the phenotype of the mycobacterial stationary phase, the effect of growth phase of Mycobacterium smegmatis on total protein synthesis and on the heat shock response was investigated. De novo protein synthesis was monitored by labelling with 35 [S]methionine and the protein expression profiles analysed using one- and/two-dimensional polyacrylamide gel electrophoresis, autoradiography, and/or immunoblot analysis. The ATP content of the culture was found to be a more accurate indicator that cells were entering stationary phase than the number of colony forming units (CFU). A plateau in the ATP growth curve preceded several stationary phase-induced events : a transitory cessation in the increase in number of CFU ; a decrease in the rate of accumulation of the cell division protein FtsZ; inhibition of the synthesis of 58, 30.5, and 20 kDa exponential phase proteins; induction of the 48, 46, 32, 31, 25, and 20 kDa stationary phase (postexponential phase) proteins ; and the highest induction of the 95 kDa, 75 kDa (DnaK), 66 kDa ( GroEL ), and - 17 kDa (doublet) proteins in response to heat shock. Identification of the stationary phase-induced proteins should enable their roles in the multigenic processes that occur during transition into stationary phase to be determined. The amino acid sequence of one of the - 17 kDa heat shock proteins (with an apparent molecular weight of 16.8 kDa, named Hspl7-2) showed significant homology to open reading frame 28 of M tuberculosis cosmid MTCY01B2. This is the first time a functional characteristic has been assigned to this open reading frame, and it remains to be seen if Hspl 7-2 represents a new family of heat shock proteins. Synthesis and secretion of the antigen (Ag)-85 complex proteins was demonstrated for the first time in M smegmatis. Heat shock resulted in increased release of Ag85A and Ag85B but not of Ag85C in M smegmatis. No heat-induction of the Ag85 complex could be demonstrated in My cobacterium bovis BCG. Whereas heat shock resulted in increased release of the 19 kDa lipoprotein antigen in both M bovis BCG and M tuberculosis H37Rv, its presence in M smegmatis could not be demonstrated. This study presents an experimental approach which may prove useful in investigating the effect of various environmental stresses on the profile, and hence the function of secreted proteins.

Bibliography: p. 169-207.

This thesis aimed to elucidate the structure-function relationships determining the differential fidelities of the dnaE1- and dnaE2-encoded mycobacterial PolIIIα subunits under conditions of genotoxic stress. To this end, the role in DnaE1 intrinsic fidelity of highly conserved PHP domain residues was explored by site-directed replacement of targeted amino acids, resulting in a panel of Mycobacterium smegmatis mutants carrying selected dnaE1 alleles. A complementary approach investigated the contribution of the mycobacterial proofreading DnaQ subunit homolog to the maintenance of DnaE1-dependent replicative fidelity by generating a targeted dnaQ knockout mutant. The third component of this study focused on the inferred role of a highly-conserved N-terminal extension and C-terminal pentapeptide motif in the function of the alternative, error-prone DNA PolIIIα subunit, DnaE2.

The laboratory diagnosis of fungal infection is complicated, based on the microscopic detection, culture isolation, and detection of serological response. In recent years, there has been an increase in the utilization of molecular diagnostic techniques for the detection and accurate identification of fungal pathogens. This was a laboratory accuracy study evaluating the performance of a selected pan-fungal PCR using 70 previously identified reference fungal isolates. The DNA yield and purity of three different DNA extraction methods was assessed, using 6 representative fungal isolates. The ZR Fungal/Bacterial DNA MicroPrep™ produced a median concentration of 17.28 ng/μl,which was significantly higher (p value = 0.0079) than the MagNA pure LC DNA Isolation Kit III and QIAamp DNA Mini Kit, which produced median yields of 11.08ng/μl and 3.54 ng/μl, respectively. The selected pan-fungal PCR was optimized PCR and successfully performed on 62 of the 70 reference isolates. A selection of 56 amplicons were submitted for DNA sequence determination. Of all the sequences queried on the NCBI and Ribosomal Development Project (RDP) databases, 95/111(86%) were concordant with the results obtained from the reference laboratory. Study findings have shown that the selected pan-fungal PCR, coupled with DNA sequence analysis is an excellent diagnostic tool for the identification of medically relevant fungi. This assay, in combination with conventional culture, is useful for the rapid and accurate identification of fungal isolates. Future work will involve evaluating the utility of this assay for the detection and identification of medically relevant fungi in deep tissue biopsies.

DNA damage repair mechanisms are critical to the adaptive evolution of Mycobacterium tuberculosis as obligate human pathogen, including the emergence of drug-resistance during anti-tuberculosis (TB) chemotherapy. In experimental models, DnaE2-dependent translesion synthesis (TLS) and UvrB-dependent nucleotide excision repair (NER) have been identified as major mediators of DNA damage tolerance and repair, respectively. Given the inferred dominance of these pathways, this thesis aimed to elucidate otherwise cryptic repair mechanisms which might buffer loss of DnaE2 and UvrB in bacilli exposed to genotoxic stress. Using dnaE2 and uvrB deletion mutants of the model mycobacterium, M. smegmatis (MSM) mc2 155, we applied genome-wide transposon (Tn) mutagenesis to identify conditionally essential repair pathways under treatment with genotoxins of different mechanistic classes. To this end, the DNA crosslinking agent, mitomycin C (MMC), and the gyrase inhibitor and clinically relevant TB drug, moxifloxacin (MOX), were used. The goal was to reveal potential targets for co-drugs that might shorten treatment duration and reduce the risk of drug resistance by severely limiting the intrinsic capacity of MTB to tolerate lethal drugs for extended periods. Among others, our analysis identified GlgB, which is involved in glycogen biosynthesis, and mycothiol biosynthesis proteins, MSMEG_0933 and MSMEG_5261, as compensating the absence of UvrB during MMC treatment. Under MOX treatment, the absence of UvrB was compensated by the RecC/Single-strand Annealing pathway. In contrast, DnaE2 deficiency revealed the conditional essentiality of the PadR family transcriptional regulator, MSMEG_2868, under MMC exposure. Importantly, in all cases, results from the Tn screen were validated using CRISPR interference targeting the identified genes. Of particular interest, we observed that UvrB was essential to compensate loss of DnaE2, whereas the reciprocal was less definitive: while DnaE2 appeared dispensable in MMC-treated uvrB, Tn analyses suggested that dnaE2 might be essential in the untreated ∆uvrB mutant. This result, which is consistent with very recent results suggesting the co-ordination of NER and DnaE2 functions in Caulobacter crescentus, is intriguing in potentially revealing a previously unappreciated role for DnaE2 in mycobacterial NER function. Taken together, these results support the utility of Tn-based whole-genome screens in revealing unexpected genegene interaction networks, and provide additional impetus to explore ancillary, non-essential metabolic functions as alternative targets for novel combination therapies designed to cripple intrinsic mechanisms of mycobacterial resistance.

Includes bibliographical references (p. 140-168).

Includes bibliographical references (leaves 92-115 ).

Bibliography: leaves 172-193.