Guanosine tetraphosphate (ppGpp) is a bacterial signaling molecule involved in activating the stringent response, a cellular reaction to environmental stress that downregulates cell division and metabolism processes to conserve nutrients. The stringent response is implicated in some instances of antibiotic persistence, so broadening the current understanding of ppGpp signaling is useful. This thesis seeks to generate a ppGpp biosensor that will bind ppGpp and emit fluorescent light in its presence, which will allow for improved research into the pathways and functions of the signaling molecule. To generate a novel ppGpp biosensor, I converted a biosensor previously used to detect cyclic di-GMP (a different signaling molecule) to contain a binding site transformed to now bind specifically with ppGpp. The genetic sequence for the cyclic di-GMP binding site was replaced with the ppGpp hydrolase domain which has a specific affinity for ppGpp; however, hydrolase activity would provide unwanted breakdown of the ppGpp, so it is mutated further to neutralize hydrolase activity. The desired outcome of this thesis results in a biosensor with a binding site that has a specific and sufficient binding affinity for the ppGpp molecule. Using this, we can determine how ppGpp levels are regulated in bacteria under conditions of stress, and how this signaling molecule is related to the survival of bacteria in response to antibiotic treatment.

Tee PhD thesis presents the study of the response of Mycobacterium tuberculosis, the causative agent of tuberculosis, upon prolonged exposure to lethal concentrations of the first line anti-tuberculosis drug, rifampicin. The study shows that prolonged exposure to lethal concentration of rifampicin causes cell death initially by several log orders of cells, followed by a persistence phase, from where rifampicin resisters emerge carrying mutation in the RRDR locus of the rpoB gene. This phenomenon was found to occur even when the drug concentration is well above MBC levels. Luria-Delbruck experiment, in a modified format, showed that the resisters emerged as fresh mutants, and not due to the growth of pre-existing natural resisters to rifampicin. The per sister cells, which showed high levels of hydroxyl radical generation, were found to have thickened outer layer, unlike the mid-log phase cells, which restricts the permeability of a fluorescent-conjugate of rifampicin, 5-FAM-rifampicin, 10-fold less in per sister cells, as compared to mid-log phase cells. The thickened outer layer has high negative surface charge and is hydrophilic in nature. It is proposed that the hydrophilic-natured thickened outer layer might have restricted the permeability of lethal concentrations of hydrophobic-natured rifampicin. This, in turn, might have ensured the presence of sub-lethal concentration rifampicin inside the per sisters, which in turn might have generated the hydroxyl radical that caused mutagenesis to generate rifampicin resisters. The Chapter 1 is the Introduction to the thesis presented in 4 parts – Part 1.1, 1.2, 1.3 and 1.4, introducing briefly the history of antibiotics, antibiotic resistance, antibiotic persistence and a brief history of tuberculosis respectively. It is concluded with a rationale behind the present study. The Chapter 2 presents the entire Materials and Methods used in the experiments described in the thesis. The Chapter 3 presents the data on the response of M. tuberculosis to rifampicin upon extended exposure. Rifampicin exposure of susceptible M. tuberculosis H37Ra cells showed a decrease in their CFU/ml followed by a persistence phase wherein the CFU/ml remained constant. However, prolonged exposure of rifampicin even at higher concentrations showed regrowth in the culture, which was found to be due to the emergence of rifampicin-resistant bacteria. Screening of rifampicin-resistant mutants showed point mutations in the rifampicin resistance determining region (RRDR) of the rpoB gene in all the mutants. In parallel, using trans formants of M. tuberculosis expressing unstable GFP under the respective native ribosomal RNA promoter, the metabolic status of the per sister cells was determined. When actively growing highly fluorescing cells were exposed to lethal concentration of rifampicin, their metabolism diminished, as illustrated by the decrease in their fluorescence during persistence phase, followed by the emergence of a sub-population of bacteria which were again metabolically active. In order to verify whether the rifampicin resisters are freshly formed mutants or have come from the naturally existing resisters, Luria-Delbruck fluctuation test was performed in a modified manner. The number of rifampicin resisters that emerged from the persistent phase was found to vary amongst different cultures from different days and different times of exposure, showing fluctuation. However, the addition of theorem before persistence phase almost abolished the generation of the rifampicin-resistant bacilli, indicating the role of hydroxyl radical in the emergence of rifampicin resisters. The generation of hydroxyl radical in mycobacteria exposed to rifampicin was confirmed using electron para-magnetic resonance spectrometry (EPR), with the spin trap agent, 5,5- Dimethyl-1-pyrroline N-oxide (DMPO) specific for hydroxyl radical. An increase in the formation DMPO-OH adduct in the persistence phase cells was observed, in comparison to mid-log phase cells. Exposure to theorem significantly diminished the adduct formation. The persistence phase cells also showed significantly high levels of signal specific to the hydroxyl-specific fluorescent dye, hydroxyphenyl fluorescein (HPF), as compared to the mid-log phase cells. In addition we have determined the oxidative stress in the bacilli upon rifampicin exposure using a redox biosensor (Mrx1-roGFP) which also showed high oxidative stress in the persistent phase. These observations confirmed the presence of high levels of oxidative stress and hydroxyl radical in the rifampicin persistent cells, in comparison with mid-log phase cells. Whole genome sequencing of the four independently isolated rifampicin resistant M. tuberculosis showed genome wide mutations having less common mutations with respect to the wild type genome, indicative of the occurrence of random mutagenesis. In addition mutation frequencies were comparable between the samples with respect to the wild type sample. About 69% of the mutations were A-C or T-G, followed by A-T or T-A, which is known to be due to oxidative stress in the cells. Variations in the colony morphology were also observed on the persistent phase cells, indicating the occurrence of mutagenesis in the bacterial genome during rifampicin treatment. The Chapter 4 is on the morphological and ultrastructural studies on rifampicin-exposed M. tuberculosis cells. Transmission electron micrographs of rifampicin per sisters showed significant thickening of the outermost capsular layer (OL), as compared to the mid-log phase cells. This observation was verified by staining the cells with ruthenium red, which specifically stains anionic polysaccharide of the OL. Zeta potential (ZP) measurement of the surface charge of the persistent cells showed high negative ZP, as compared to the mid-log phase cells. The negative ZP value was found to gradually increase during the course of rifampicin treatment and to decrease during the regrowth phase. Hexadecane assay showed larger proportion of per sister cells being retained in the aqueous phase, as compared to the mid-log phase cells. This indicated the higher hydrophilicity of the per sister cells, which was in agreement with the higher surface negative charge of the cells. The permeability of rifampicin per sister cells to 5-FAM-rifampicin (rifampicin conjugated to 5-carboxy fluorescein, which is as hydrophobic as rifampicin) was found to be 10-fold less than that of the mid-log phase cells. However, removal of the thick OL by bead beating (BB) with 4 mm glass beads significantly improved the permeability of per sister cells to 5-FAM-rifampicin. On the contrary, no difference in the 5-FAM-rifampicin uptake was observed in mid-log phase cells, with or without BB. These observations implied that the thick hydrophilic-natured OL with high negative surface charge may be playing a significant role in limiting the permeability of hydrophobic-natured rifampicin entry into the persisted cells. This in turn may ensure the presence of sub-lethal concentration of rifampicin inside the persisted cells that are exposed to lethal concentration of the antibiotic. Exposure of bacteria to sub-lethal concentration of antibiotics has been reported to generate oxidative stress in the bacteria, leading to mutagenesis. A model has been proposed based on these observations in which the persistent mycobacteria are protected from lethal concentrations of the rifampicin by the thick OL which in turn ensures sub-lethal intracellular antibiotic concentration, leading to the generation of hydroxyl radical mediated mutagenesis and thereby emergence of rifampicin resisters. This thesis is concluded with the list of salient findings, publications and references.