Abstract
Abstract For over 50 years, drug combinations have been the gold standard in the therapy of tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb). Clinically, synergistic combinations of drugs are favoured over monotherapy as they allow the use of lower doses and also suppress the development of drug resistance. However, multiple drug-resistant TB (MDR-TB) is on the rise, and is threatening the effective control of the disease globally. Prior to this thesis, drug synergy studies in mycobacteria have largely been determined empirically with little or no mechanistic understanding of the mechanisms of synergy. Also, although it has been assumed that mutation rates for drug combinations are multiples of the mutation rates for single drugs, this assumption has not been thoroughly tested. In this thesis, drug synergy was demonstrated in Mycobacterium smegmatis (Msm), a model TB organism, using Isoniazid (INH) and Rifampicin (RIF), the two most effective bactericidal drugs used in TB therapy. A novel indirect approach of cell inhibition was employed, whereby the single drugs on their own were utilized at sub-lethal concentrations that alone were only weakly inhibiting, but in combination resulted in synergy. The hypothesis investigated in this study was that synergy was a consequence of the interaction between antibiotics inducing a cryptic perturbation state (AICPs), such that each single antibiotic produced a synergistic effect in combination. To investigate changes induced in the AICPs, RNA-sequencing (RNA-seq) was performed on RNA extracted from cells in the AICPs. Perturbed genes were mapped onto Kyoto Encyclopaedia for Genes and Genomes (KEGG) pathway maps, to identify affected metabolic pathways and identify differences between perturbations due to INH and RIF. Differential Producibility Analysis (DPA) was also used to highlight metabolite differences in states induced by each antibiotic. Mutation rates were also measured to INH and RIF singly and in combination using the fluctuation assay. Mapping of the differentially expressed genes from RNA-seq onto KEGG pathway analysis identified perturbation of a significant number of genes encoding proteins involved in sulphur metabolism and ribosome synthesis pathways, exclusively in the AICPs to INH, but not in the AICPs to RIF. Perturbations to genes encoding ABC transporters were identified in the AICPs to both antibiotics, but the specific transporters involved differed to the two antibiotics. DPA revealed stronger metabolite signals in the AICPs in response to INH, affecting putrescine, iron, malonate, and β-alanine that affected several areas of metabolism. However, only one metabolite involved in potassium ion transport was significantly perturbed by DPA analysis in the AICPs induced by RIF. The action of these metabolites and genes acting in parallel pathways provides a novel explanation for drug synergy in mycobacteria. Mutation rates to resistance to INH and RIF were estimated separately and, also, mutations to resistance to both drugs present in combination. Surprisingly, mutations to resistance to both drugs were found at a rate four orders of magnitude higher than expected from the mutation rate to single drugs. This result needs to be confirmed by further studies but, if confirmed, would suggest that drug combinations may, paradoxically, enhance the mutation rate to two or more drugs. These results have provided novel insights into the mechanism of drug synergy in mycobacteria.