Search

The aim of this study was to identify and characterise type II toxin-antitoxin (TA) systems in Burkholderia pseudomallei, the causative agent of the human disease melioidosis. 8 putative TA systems were identified within the genome of B. pseudomallei K96243. 5 of these were located witihn genome islands. Of the candidate toxins, BPSL0175 (RelE1) or BPSS1060 (RelE2) caused growth to cease when expressed in Escherichia coli, whereas expression of BPSS0390 (HicA) or BPSS1584 (HipA) (in an E. coli ΔhipBA background) caused a reduction in the number of culturable bacteria. HicA also caused growth arrest in B. pseudomallei K96243 ΔhicAB. These toxin induced phenotypes were enhanced by an <3kDa extracellular factor that accumulated in the spent medium during growth. Expression of the cognate antitoxins could restore growth and culturability of cells. Expression of hicA in E. coli gave an increased number of persister cells in response to ciprofloxacin or ceftazidime. Site directed mutagenesis studies identified two key residues within the HicA toxin that were essential for both the reduced culturability and increased persistence phenotypes. Deletion of hicAB from B. pseudomallei K96243 did not affect persister cell or survival frequencies compared to the wild type following treatment with a variety of stress conditions. Deletion of the ΔhipBA locus from B. pseudomallei K96243 also had no affect on bacterial persistence or survival under the conditions tested.

The emergence of multidrug-resistant bacteria has led to an ever-growing antibiotic resistance crisis. To combat this growing crisis, new strategies for controlling bacterial cell growth must be developed. An underdeveloped target for new antimicrobial therapeutics are toxin-antitoxin (TA) systems. TA systems are widely dispersed genetic operons in prokaryotes which consist of a non-secreted protein (toxin) which targets essential metabolic enzymes causing cell death. The toxin’s cellular toxicity is neutralized by its cognate antitoxin (RNA or protein). In the case of type-II TA systems, the toxin protein is neutralized by directly binding to a protein antitoxin forming a toxin-antitoxin complex. The work presented within this dissertation aims to address several gaps in understanding how the protein-protein interactions of type-II TA systems are formed and maintained, with long term goals of targeting these protein-protein interactions to develop new means to control bacterial cell growth. This work focuses on the ParDE1 TA system from Pseudomonas aeruginosa, composed of the ParE toxin protein which targets DNA Gyrase.

This work combines traditional microbiology with bioinformatic and synthetic biology approaches to study antibiotic tolerance. Antibiotic tolerance is a widespread phenomenon that facilitates antibiotic resistance and decreases the effectiveness of antibiotic treatment. Tolerance is distinct from antibiotic resistance, because tolerance is short term survival and typically results from phenotypic variations rather than genetic variation. The molecular mechanisms underlying tolerance are varied and debated in the literature. I have explored two intracellular processes related to tolerance, toxin-antitoxin (TA) systems (Chapter 2) and proteases (Chapter 4). Specifically, I focus on the ratio of antitoxin-to-toxin in type II TA systems, because type II TA systems must be regulated in such a way that antitoxins are more prevalent than their toxins. Our analysis of RNA-sequencing and ribosome profiling data demonstrates that most type II TA systems in E. coli are regulated at the translational level, while others rely on various combinations of transcriptional and post-transcriptional regulation. Before publishing this article, researchers often cited transcriptional regulation as the primary method of regulating TA systems. Studying antibiotic tolerance and other subpopulations necessitates the ability to study single-cell dynamics in the context of the whole population. To facilitate single-cell analysis, we have developed single-cell tracking software that leverages machine learning to identify cells. The software then tracks the cell based on this classification and returns data on cell size, location, division and fluorescence. The software provides the means of quantifying cell behavior before and after antibiotic treatment. One such system we would like to apply this software to is our work on proteolytic queueing and antibiotic tolerance. Proteases are responsible for protein degradation and, as such, regulate many cellular functions. To better identify the role proteases play in persistence, we used proteolytic queueing to interfere with proteolytic activity. We found that interfering with degradation at the protease ClpXP increases antibiotic tolerance ~80 and ~60 fold in an E. coli population treated with ampicillin and ciprofloxacin, respectively. I used stochastic modeling to support our results, and we have experimentally determined that altering the expression of the synthetic system affects the level of tolerance in the population. I am currently using next-generation sequencing to identify the systems being affected by the queue.

A key impediment to the control of tuberculosis is the ability of the causative agent, Mycobacterium tuberculosis (MTB), to adopt a persistent, non-replicating state that renders the bacteria resistant to both host and antibiotic killing. The expression of toxin-antitoxin (TA) systems in other bacteria regulate the generation of persister subpopulations and act as stress response elements that curtail bacterial replication under adverse conditions. We have comprehensively identified putative TA systems in M. tuberculosis and identified 30 that are functional when expressed in M. smegmatis. Why M. tuberculosis has so many TA systems and what role they play in the unique biology of the pathogen is unknown. Comparative genomic analysis revealed that the vast majority of these systems are conserved in the M. tuberculosis complex (MTBC), but largely absent from other mycobacteria, including close relatives of M. tuberculosis. Expression profiling demonstrated that four systems are specifically activated during stresses likely encountered in vivo, including hypoxia and phagocytosis by macrophages. Given that hypoxia is an in vitro condition that induces the persistent state in M.tuberculosis, we sought to further characterize the two TA systems induced during hypoxia, Rv2009-2010 and Rv1955-1956. Intriguingly, both of these systems are part of a horizontally-acquired genomic island that contains other hypoxia-regulated genes. A knock-out mutant of Rv2009-2010 is able to out-compete wild type M. tuberculosis during competitive growth assays under hypoxic conditions. Additionally, this mutant appears to have an altered transcription profile during hypoxia, as compared to wild type bacteria. Further examination of this region of the genome revealed that an adjacent gene, Rv2008c, is misregulated in cells lacking the Rv2009-2010 operon and may contribute to some of the phenotypic changes observed in the mutant.In contrast to canonical two-gene TA systems, the Rv1955-Rv1956 locus encodes four genes. Using multiple protein interaction techniques, we have shown that one of these accessory proteins, Rv1957, interacts directly with that antitoxin and the TA complex. We show that Rv1955-1957 is transcribed by at least two promoters that are differentially regulated by stress. Additionally, we provide evidence that Rv1957 is required for full repression of the TA system operon. The complex architecture and regulation suggest that Rv1955-1957 via a novel mechanism to promote persistence in a major human pathogen.