A genomics love story: Tying the knot between micro- and macroevolution in the Mycobacterium tuberculosis complex

Abstract

Tuberculosis (TB) is a leading cause of death globally. Understanding the population dynamics and evolution of TB’s causative agent Mycobacterium tuberculosis complex (MTBC) over short time-scales in-host (micro-evolution) and long time-scales between hosts (macro-evolution) is vital in understanding the efficacy of antibiotic treatment and in rational vaccine design. We investigate these evolutionary dynamics by analyzing more than 32,000 globally and genetically diverse MTBC genomes. In chapter 1 and 2, we make use of 614 paired longitudinally collected clinical MTBC isolates that underwent Whole-Genome Sequencing (WGS) representing 307 patients from eight studies to investigate genetic diversity in-host. We analyze isolates from the sputa of 200 patients to investigate MTBC diversity during the course of active TB disease after excluding 107 cases suspected of reinfection, mixed infection or contamination. We conclude that low abundance resistance variants can predict fixation at a later time point and identify significant in-host variation in antibiotic resistance genes, metabolic genes and genes known to modulate host innate immunity. In chapter 3, by analyzing 31,440 MTBC genomes, we set out to survey whether markers of resistance to certain antibiotics (isoniazid, amikacin, kanamycin, bedaquiline, clofazimine) may be affected by epistasis if they involve over-expression of a non-essential drug resistance gene. We surveyed our sample for instances in which the markers of resistance occur in a regulatory region and the (resistance-conferring) target has a loss-of-function (LoF) mutation that has emerged either before or after the introduction of antibiotics. We demonstrate several instances in which genotypic drug-susceptibility tests can result in major diagnostic errors when only the regulatory regions are probed and observe the first example of antibiotic reversion in the MTBC. In chapter 4, we leverage 31,428 MTBC genomes to characterize the genetic diversity present in the MTBC and mutational processes that give rise to this genetic variation. Further, we investigate regions of the genome that have been subject to positive selection by inferring the number of times each variant has arisen independently (parallel evolution) in our large dataset of sequenced isolates. We observe an extensive amount of parallel evolution for both single nucleotide and insertions/deletion variants. We find that phase variation is a major driver of adaption in MTBC.