The evolution of drug resistance in the different lineages of the Mycobacterium tuberculosis complex

Abstract

The eradication of drug resistance (DR) in Mycobacterium tuberculosis (Mtb) has been proven to be challenging. Several DR mechanisms that are not fully understood have been suggested to play significant roles in the acquisition, maintenance, and spread of DR-Mtb strains. A DR mechanism of interest in this study was epistasis; an important evolutionary process suggested to play a role in the evolution of DR in Mtb. This mechanism of interest results in co-adapted alleles in the Mtb population meaning that the acquisition of new mutations are not independent of each other. Our study aimed at modelling evolutionary trajectories of DR in the human adapted lineages of the M. tuberculosis complex. To accomplish this an extensive search of sequence data in public databases as well as metadata was done. The data then went through quality analyses that yielded a dataset of 9388 sequences that represented the diversity of Mtb isolates from different lineages. Additionally, statistical analysis identified significant non-random associations between linked common and lineage specific DR mutations. This analysis enabled the identification of epistatic interactions among pre-requisite/stepping stone mutations, intermediate mutations, DR mutations, and compensatory mutations, all of which were suggested to play a role in the evolution of DR in Mtb. Lastly, evolutionary trajectories towards DR were modelled for the overall Mtb population and the different lineages Identifying linked common associations resulted in the identification of 64 universal markers, in functional and non-functional genes, of DR to more than three antibiotics. Functional mutations included a DR mutation and other mutations involved in key biological processes such as lipid metabolism, and redox metabolism suggested to have a strong association with multiple drug resistance. Non-functional genes were largely comprised of highly mutable pe and ppe genes. Associations among these markers were then determined to investigate the likelihood of the occurrence of a “mutator phenotype” necessary for the acquisition of a multiple drug resistance phenotype. Indeed when these associations were modelled they demonstrated the occurrence of a “mutator” phenotype as a result of PE_PGRS genes before the acquisition of multiple drug resistance. Further analysis was done to determine lineage specific non-random associations between polymorphisms in the various lineages. Epistatic interactions among polymorphisms formed multiple evolutionary pathways that resulted in various evolutionary trajectories. Several evolutionary paths within the various lineages were established and showed the importance of the various categories of mutations and their role in DR acquisition and compensation. Several pre-requisite mutations in genes of known function and hypotheticals were suggested to be important in the acquisition of high level resistance. Intermediate mutations were also suggested to play a role in the acquisition of several evolutionary trajectories. The study also identified several pairs of DR mutations that formed positive epistatic interactions promoting further resistance to additional drugs. Compensatory mutations in several genes were suggested to be important in the amelioration of fitness costs as a consequence of DR acquisition. Ultimately these results showed the importance of epistasis in the evolution of DR in Mtb. This study then concluded by emphasising the importance of understanding evolutionary trajectories towards DR in Mtb because this would lead to the improvement of diagnostics, treatment outcomes, and the effective use of antibiotics. The study also highlighted the importance of further investigation into inter-lineage diversity and its role in the evolution of DR. Furthermore, much progress has been made in the fight against the tuberculosis disease, however, significant gaps remain in our understanding of the pathogenesis of this bacterium.