Characterization and Elimination of Stochastically Generated Persister Subpopulation in Mycobacteria

Abstract

Mycobacterium tuberculosis (MTB) persists in the host for long periods, even during antibiotic treatment. Eliminating persister cells, which underlie this phenomenon and are the primary reason for treatment failure, is essential for shortening TB treatment regimen. Here, we report a novel methodology, Per-Sort that takes advantage of a physiological characteristic (translational dormancy) of persisters to isolate them without antibiotic enrichment. Using Per-Sort, we have discovered that translationally dormant cells pre-exist as a fraction of a percent of an isogenic mycobacterial population, suggesting they are generated stochastically as a bet hedging strategy. We show that this pre-existing subpopulation is tolerant to antibiotics, low in oxidative metabolism, and expand in number upon nutrient starvation. Finally, through transcriptional profiling at single-cell resolution, we have determined that the pre-existing persisters are a heterogeneous mix of vapC30, mazF, and relA/spoT overexpressing cells that are eliminated and sensitized to antibiotic killing through induction of respiration.

Keywords: Mycobacterium, pre‐existing persisters, antibiotic tolerance, nutrient starvation