Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb), was the second leading infectious cause of death after COVID-19 in 2022, and is a major contributor to antimicrobial resistance worldwide [1]. Pyrazinamide (PZA), a first-line drug for TB treatment, is effective in killing Mtb. Inclusion of PZA in treatment regimens for drug-susceptible patients significantly shortens therapy duration, from 12 to 6 months [2]. PZA is a prodrug that must be converted to its bioactive form, pyrazinoic acid, by the pyrazinamidase PncA, and is active under acidic pH [3,4]. PncA mutations can cause PZA resistance in Mtb. The specificity of PncA sequencing in predicting PZA resistance exceeds 90% in China [5]. PZA effectively penetrates the cellular and necrotic compartments of TB lesions, and markedly decreases bacterial burden [6,7]. In a prior study, after 12 weeks’ treatment of Mtb infected mice with a combination of isoniazid (INH) and high dose PZA, no bacteria were detected in the organs. Moreover, when guinea pigs were inoculated with homogenates from the spleens of these mice, they showed negative cutaneous reactions to tuberculin. However, after a 90-day treatment-free interval, Mtb was cultured from the spleens of 12 of the 30 infected mice, and the isolated bacteria were sensitive to both PZA and INH. Thus, although the combination of PZA and INH had high efficacy against Mtb, it was unable to kill all bacteria in the organs of the infected mice [8,9]. We hypothesized that the remaining bacteria might not have drug resistance-associated genetic mutations, and might have become drug-tolerant persisters.
In this study, we established an in vitro assay based on the BD BACTEC™ MGIT™ 960 system to identify persister formation according to PZA exposure. The PZA MIC of the H37Rv reference strain used in this study in pH 5.9 medium was 50 μg/ml, as determined with the Alamar Blue method [10]. When H37Rv was grown to an OD of 0.4–1.0 in 7H9 medium, 5×106 or 5×104 CFU H37Rv was inoculated in MGIT 960 tubes with different concentrations of PZA (0, 50 μg/ml, 100 μg/ml, 250 μg/ml, 500 μg/ml, or 1000 μg/ml) in pH 5.9 medium, then cultured for 72 h or 48 h. After exposure, the samples were washed with PBS and inoculated into new MGIT tubes to determine the time to detection (TTD) in days (Figure 1A). Although we used high concentrations of PZA to kill Mtb, even after treatment with 20× the MIC, some bacteria remained and were able to regrow in drug-free medium. Moreover, the TTD was prolonged with increasing PZA concentration (Figure 1B, C). When the PZA concentration reached 250 μg/ml, the TTD became stable. No PncA mutants were found in the regrown strains, according to PncA gene sequencing. To further identify PZA persister formation in vitro, we treated H37Rv with 100 μg/ml, 250 μg/ml, or 500 μg/ml PZA concentrations in MGIT 960 tubes for 24 h, 48 h, or 72 h. The bacteria in the tubes were centrifuged, washed, and spread onto drug-free 7H10 plates (Figure 1A). After treatment with high concentrations of PZA for 24 h, 48 h, or 72 h, more than 90% of the Mtb were killed, and a minority of bacteria remained alive and were able to be cultured on drug-free 7H10 plates. Bacterial growth did not increase with incubation time, thus indicating that PZA resistance was not induced by PZA treatment in this model (Figure 1D). Furthermore, 14 clones were isolated from the plates, among which no PncA mutants were identified, thereby suggesting that no genetic resistance occurred in these clones after PZA exposure. These results further demonstrated that PZA persisters can form after PZA exposure.
Mycobacterium tuberculosis H37Rv forms persisters after PZA exposure. (A) Schematic of the TTD assay and CFU assay for identifying Mtb persisters after PZA exposure by using the MGIT960 system. (B, C) H37Rv can form culturable persisters under high concentrations of PZA. The TTD was determined with the MGIT960 system. (D) H37Rv was exposed to high concentrations of PZA in the MGIT960 system, and persisters were identified with CFU assays on plates.
In this study, we established an assay using the BACTEC MGIT960 system to identify PZA persisters; this method could also be applied to research on other antibiotics. This report is the first to identify persister formation after PZA exposure in vitro. The findings may explain the TB relapse occurring in patients after treatment. Future in vitro and in vivo studies of related genes and mechanisms responsible for PZA persister formation are warranted, to aid in screening new drugs and decreasing TB treatment times.