The evolution of fluoroquinolone-resistance in In Vitro and in natural populations of mycobacterium tuberculosis

Despite the advent of antimicrobials, tuberculosis (TB) caused the greatest amounts of deaths due to an infectious disease in 2018, accounting for 1.2 million deaths alone and an additional 0.3 million deaths due to TB-HIV co-infections. Current TB treatment regimens may impose substantial economic and logistical burdens for patients and health care systems, as even treatment against drug susceptible strains of Mycobacterium tuberculosis (Mtb), the aetological agent of TB, requires daily doses of antimicrobials for 6 to 9 months. Further complications arise when patients are infected with multidrug-resistant strains of Mtb, which increases treatment duration to 9 to 24 months. Treatment success rates are also reduced from approximately 85% for drug-susceptible cases down to almost 50% for multidrug-resistant cases of TB. Therefore, substantial efforts by the medical and research communities are currently underway to develop new treatment regimens that are both more efficacious and can reduce treatment duration against both drug-susceptible and multidrug-resistant strains of Mtb.
Fluoroquinolones (FQs) form a vital component in established and experimental TB treatment regimens. Older generations of FQs have been used to treat multidrug-resistant forms of TB, while newer and more potent FQs are being tested in experimental regimens that aim to reduce TB treatment duration. Extensive biochemical work has shown that FQs target DNA gyrase, the sole type II topoisomerase
in Mtb. Molecular epidemiological studies demonstrate that clinically-relevant FQ-resistance (FQ-R) mutations are restricted to a limited set of chromosomal mutations in the genes encoding DNA gyrase: gyrA and gyrB. However, little work has been done on exploring the evolution of FQ-R in populations of Mtb. Investigating how different Mtb populations evolve under FQs pressure, and how FQ-R mutations affect the continued evolution of Mtb populations, may aid in maintaining the potency and potential use of FQs.
Treatment regimens for TB generally use standardized, empirical dosing, including when using FQs. Previous work has shown that different Mtb genetic variants can associate with different frequencies of drug-resistance (DR), even when using standardized treatment regimen. Bacterial genetics have also been shown to modulate the phenotypes that DR mutations confer. Whether the genetic variation present in natural populations of Mtb would also modulate the frequency of FQ-R emergence, or the phenotypes that FQ-R mutations confer, is currently not known. It is also unclear how FQ-R mutations themselves affect the continued evolution of Mtb populations.
In this Thesis, we explored how FQ-R evolves in Mtb. Specifically, we used extensive in vitro experiments coupled with analysis of publicly available Mtb genomic sequences isolated from clinical strains to test whether the genetic variation in Mtb modulated the frequency and phenotypes of FQ-R mutations. We then used a mathematical modeling framework and in silico simulations to test the relative contributions of bacterial factors hypothesized to be relevant in DR evolution in determining the frequency of FQ-R. Lastly, we used further in vitro assays andMtb genomic sequences from clinical strains to test the impact of FQ-R mutations on the continued evolution of Mtb populations.
This Thesis consists of 6 main chapters. The first chapter provides a general introduction into TB and the rationale for this Thesis, while the second chapter states the main Aims and Objectives. Three chapters then present the results of our research work, with one chapter dedicated per stated Objective. The last chapter provides a synopsis of our main findings, states general limitations, highlights the contribution of this Thesis to the Mtb and antimicrobial resistance research communities, and highlights
potential future directions that can build upon this Thesis work.
In Chapter 1, we introduce the global burden of TB and treatment regimens for TB. We then highlight the problem of antimicrobial resistance,and the potential use of more potent FQs in reducing TB treatment times. Lastly, we introduce evolutionary concepts in DR evolution, highlight the role of bacterial genetics in DR evolution, and state the rationale for this Thesis.
In Chapter 2, we state the Main Aim and Objectives of this Thesis.
In Chapter 3, we use the Luria-Delbrück fluctuation analysis, further in vitro assays, and genomic sequences analysis to test whether the genetic variation present in natural populations of Mtb modulates the frequency and phenotypes of FQ-R mutations. We show that the Mtb genetic background can lead to differences in FQ-R that spans two orders of magnitude. Furthermore, we find that the
Mtb genetic background modulates the phenotypes conferred by FQ-R mutations in vitro, and the observed types and relative frequencies of FQ-R mutations both in vitro and in the clinic.
In Chapter 4, we adapt a mathematical modeling framework developed by Ford et al., 2013 to simulate the frequency of FQ-R in order to test the relative contributions of different bacterial factors in FQ-R evolution. Our results suggest that not all relevant bacterial factors have been accounted for in the model, and that a new model of DR evolution is required for Mtb.
In Chapter 5, we again use the Luria-Delbrück fluctuation analysis coupled with analysis of Mtb genomes to test whether FQ-R mutations can affect the continued evolution of Mtb populations. We observe that FQ-R mutations can increase the frequency of acquiring further DR mutations in Mtb in vitro. However, genomic analysis demonstrate that FQ-R mutations do not necessarily associate with increased genetic diversity in natural populations of Mtb.
In Chapter 6, we highlight the key findings of this Thesis. We then state the limitations, discuss the implications of our results, and propose future directions in the study of FQ-R evolution in Mtb.