Physiologic recovery of Mycobacterium tuberculosis from drug injury: A molecular study of post antibiotic effect in vitro and in vivo

Abstract

Post-antibiotic effect (PAE) describes the delay in bacterial growth that continues after antibiotics are cleared. The physiologic basis of PAE in Mycobacterium tuberculosis (Mtb) remains poorly understood. Here, we evaluated the long-standing hypothesis that PAE reflects the time required for bacteria to recover from drug-induced physiologic damage by comparing Mtb after varying durations of treatment with the four-drug isoniazid, rifampin, pyrazinamide, ethambutol combination in vitro and in BALB/c mice using two novel molecular readouts of bacterial health. In aerobic axenic culture and in the high-dose aerosol mouse model, quantification of Mtb rRNA synthesis via the RS ratio and Mtb transcriptional profiling via SEARCH-TB revealed that longer drug exposure was associated with greater injury and adaptation during treatment, as well as slower recovery after treatment, i.e., longer PAE. Recovery followed a conserved sequence, from resumption of rRNA synthesis, to broad transcriptional reprogramming, to eventual CFU change. In mice, however, physiologic recovery was markedly slower and less complete than in vitro, indicating longer PAE in the context of immunity. Our observation that PAE in Mtb depends on the duration of drug exposure and correlates with the degree of bacterial injury support the hypothesis that nonlethal physiologic damage contributes to PAE. Our observation that PAE of the standard TB regimen is longer in mice than in vitro indicates that immunity augments PAE for Mtb. Molecular evaluation of bacterial physiology provides a new basis for probing recovery from drug exposure and understanding PAE. Author summary: Tuberculosis (TB) is the leading cause of death from infection worldwide. Antibiotics used to treat Mycobacterium tuberculosis (Mtb) often inhibit bacterial growth well beyond the time that drugs are cleared and unmeasurable. It has long been hypothesized that this post-antibiotic effect (PAE) could reflect time required for bacteria to repair sub-lethal cellular damage. A barrier to understanding is that PAE has traditionally been studied by enumerating bacterial burden rather than evaluating change in bacterial health. Additionally, the PAE of TB drugs has been studied exclusively in vitro rather than in the context of immunity. Here, we used several novel molecular measures of bacterial health in in vitro and mouse experiments to evaluate Mtb recovery after short treatment with the global standard 4-drug combination treatment. We discovered three novel findings. First, longer treatment resulted in longer PAE. Longer treatment caused progressively greater injury and adaptation during treatment and slower physiologic recovery afterwards, suggesting that the worse the damage, the slower the recovery. Second, recovery followed a reproducible sequence, beginning with restoration of rRNA synthesis, followed by widespread transcriptional reprogramming, and finally increases in bacterial numbers. Third, although this recovery sequence was conserved between in vitro and mice, recovery in mice was markedly slower and incomplete, resulting in a substantially longer PAE. Together, these findings support the damage hypothesis as a component of PAE and highlight the importance of immunity in controlling Mtb after drug treatment.

Original publication: PLOS Pathogens (2026-04-22). Source. Source DOI: 10.1371/journal.ppat.1014131.