Most-probable number based minimum duration of killing assay for determining the spectrum of rifampicin susceptibility in clinical M tuberculosis isolates

Accurate antibiotic susceptibility testing is essential for successful tuberculosis treatment. Recent studies have highlighted the limitations of minimum inhibitory concentrations (MIC) based phenotypic susceptibility methods in detecting other aspects of antibiotic susceptibilities in bacteria. Duration and peak of antibiotic exposure, at or above the MIC required for killing the bacterial population, has emerged as another important factor for determining the antibiotic susceptibility. This is broadly defined as antibiotic tolerance. Antibiotic tolerance can further facilitate the emergence of antibiotic resistance. Currently there are limited methods to quantify antibiotic tolerance among clinical M. tuberculosis isolates. In this study, we develop a most-probable number (MPN) based minimum duration of killing (MDK) assay to quantify the spectrum of M. tuberculosis rifampicin susceptibility within subpopulations, based on time duration of rifampicin exposure required for killing the bacterial population. MDK90-99 and MDK99.99 defined as the minimum time duration of antibiotic exposure at or above MIC required for killing 90-99% and 99.99% of the initial (pre-treatment) bacterial population respectively. Results from the rifampicin MDK assay applied to 28 laboratory and clinical M. tuberculosis isolates showed that there is variation in rifampicin susceptibility among isolates. Rifampicin MDK99/99.99 time for isolates varied from less than 2 to 10 days. MDK was correlated with larger sub-populations of M. tuberculosis from clinical isolates that were rifampicin tolerant. Our study demonstrates the utility of MDK assays to measure the variation in antibiotic tolerance among clinical M. tuberculosis isolates and further expands clinically important aspects of antibiotic susceptibility testing