A bactericidal tuberculosis drug regimen driven by inhibition of the terminal oxidases by pretomanid
Nurlilah Ab Rahman, Samsher Singh, Thomas Wiggins, May Delos Santos, Garrett C Moraski, Marvin J Miller, Michael Berney, Kevin Pethe
TL;DR
Pretomanid, a tuberculosis drug, works by partially inhibiting key energy-producing enzymes in bacteria, and combining it with other drugs improves its effectiveness and reduces resistance.
Contribution
Pretomanid's dual inhibition of terminal oxidases is revealed, enabling a synergistic drug combination for tuberculosis.
Findings
Pretomanid partially inhibits both cytochrome bcc:aa3 and bd oxidases in M. tuberculosis.
Combining pretomanid with telacebec and a cytochrome bd inhibitor achieves rapid bactericidal activity and resistance suppression.
The triple drug combination effectively targets both replicating and non-replicating M. tuberculosis in vitro.
Abstract
Pretomanid is a unique anti-tuberculosis agent that inhibits both cell-wall synthesis and bioenergetics in Mycobacterium tuberculosis. While targeting the cell wall triggers a rapid bactericidal effect on replicating mycobacteria, the release of nitric oxide is linked to bactericidal potency against antibiotic-tolerant, non-replicating subpopulations through interference with the electron transport chain. Nonetheless, the specific molecular target(s) of the drug remain unknown. Through the utilization of genetic and chemical biology approaches, we present evidence that pretomanid inhibits both the cytochrome bcc:aa3 and bd oxidase respiratory branches. This property leads to a pronounced synergy with telacebec (Q203), a clinical-stage drug targeting the cytochrome bcc:aa3, while concurrently curtailing the emergence of resistance to pretomanid. Furthermore, the incorporation of the…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9- —http://dx.doi.org/10.13039/501100001381PMO | National Research Foundation Singapore (NRF)
- —Ineos Oxford Institute for Antimicrobial Research (IOI)
- —http://dx.doi.org/10.13039/100000002HHS | National Institutes of Health (NIH)
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsTuberculosis Research and Epidemiology · Neutrophil, Myeloperoxidase and Oxidative Mechanisms · Metal-Catalyzed Oxygenation Mechanisms
The paper explainedProblemTuberculosis remains a leading infectious cause of death worldwide, and shorter, more effective regimens are needed. Pretomanid has been approved as part of a fixed-dose regimen, but how it acts on the oxidative phosphorylation pathway is not fully understood. A clearer understanding of this mechanism is critical to design rational drug combinations that improve efficacy and limit resistance.ResultsWe show that pretomanid partially inhibits both the cytochrome bcc:aa3 and bd oxidases in Mycobacterium tuberculosis. Combination with telacebec (Q203) enhances bactericidal activity, suppresses resistance, and achieves potent efficacy in mice. Addition of the cytochrome bd inhibitor ND-011992 produces a triple regimen that rapidly sterilizes replicating and non-replicating bacilli in vitro.ImpactThis work identifies inhibition of the terminal oxidases as a major contributor to the activity of pretomanid. The findings support the development of regimens combining pretomanid with inhibitors of the terminal oxidases as a promising strategy to shorten TB treatment and reduce emergence of resistance.
Introduction
While tuberculosis (TB) remains one of the most significant global health challenges of our time (WHO, 2023), substantial progress has been achieved in the development of drugs targeting multi-drug resistant (MDR) forms of the disease. Over the past decade, three drugs have gained approval for clinical use: bedaquiline (Andries et al, 2005) in 2012 (Gras, 2013) was swiftly succeeded by delamanid (Ryan and Lo, 2014) and, more recently, pretomanid (Deb and Biswas, 2021). Furthermore, several additional drug candidates have advanced to clinical development stages. Among those, telacebec (Q203), a drug candidate targeting the respiratory terminal oxidase cytochrome bcc:aa3 (cyt-bcc:aa3) (Beites et al, 2019; Pethe et al, 2013), has demonstrated potency in a phase 2a clinical study (de Jager et al, 2020; Pethe et al, 2013). The potency of Q203 is particularly promising when combined with the cytochrome bd oxidase (Cyt-bd) inhibitor ND-011992 (Lee et al, 2021), underscoring the significance of developing rational drug regimens rather than stand-alone agents. This perspective is best exemplified by the approval of pretomanid as part of a fixed-dose regimen with linezolid and bedaquiline (Burki, 2019). Therefore, comprehending the chemical–chemical interactions among approved and late-stage clinical drug candidates is crucial for combatting MDR tuberculosis. Pretomanid is an intriguing multi-target prodrug that concurrently inhibits cell-wall synthesis (Stover et al, 2000) and energy metabolism (Singh et al, 2008). This dual mode of action is particularly noteworthy, given recent studies reporting that inhibitors of energy metabolism, such as Q203 or bedaquiline, impede the bactericidal efficacy of cell-wall targeting agents (Lee et al, 2019; Zeng et al, 2019). Although the exact mechanism of this interference remains elusive, these observations suggest that the bactericidal potency of pretomanid, driven by its impact on cell-wall synthesis, could be counteracted by its influence on energy metabolism. Prompted by this hypothesis, we studied the mechanism by which pretomanid inhibits respiration and whether bacteriostatic energy metabolism inhibitors, like Q203, would antagonize or synergize with it. Our study unveils that pretomanid triggers a rapid increase in ATP levels at low concentrations, followed by a decrease in ATP levels at higher concentrations in replicating mycobacteria, consistent with the simultaneous inhibition of mycolic acid synthesis (Stover et al, 2000) and the nitric oxide-mediated disruption of the oxidative phosphorylation pathway (Singh et al, 2008). Mode of action studies point to a simultaneous inhibition of both aerobic respiratory branches by pretomanid. Most importantly, Q203 enhances the bactericidal effectiveness of pretomanid in vitro and in vivo, while concurrently suppressing the emergence of resistance to pretomanid. A triple combination pretomanid/Q203/ND-011992 was extremely potent at eradicating replicating and non-replicating M. tuberculosis. These findings suggest that pretomanid in combination with terminal oxidase inhibitors could form the foundation of a potent drug combination for MDR TB treatment.
Results
Pretomanid has an erratic effect on intracellular ATP in M. tuberculosis
The nitroimidazole pretomanid exhibits a dual mode of action: inhibiting mycolic acid synthesis and interfering with energy metabolism, two mechanisms eliciting contrasting effects on measured ATP levels in M. tuberculosis (Lee et al, 2019; Zeng et al, 2019). To assess the impact of pretomanid on ATP levels, we employed the classical BacTiter Glo procedure (Rao et al, 2008) in replicating M. tuberculosis. Our observations revealed a rapid surge in ATP levels upon drug exposure peaking at 5 μM of pretomanid followed by a decline to untreated levels from 5 to 80 μM (Fig. 1A). In comparison, isoniazid induced a dose-dependent ATP signal increase, while bedaquiline led to ATP level depletion (Fig. 1A), in line with previous reports (Kalia et al, 2017; Lee et al, 2019). This dose-response profile was interpreted by the dual mode of action of the drug that impacts cell-wall synthesis triggering an increase in ATP signal at concentrations from 0.078 to 5 μM, which is compensated by nitric oxide-mediated oxidative phosphorylation inhibition (Singh et al, 2008) at concentrations above 5 μM. Since cell-wall synthesis becomes dispensable under non-replicating conditions, as evidenced by the loss of potency of isoniazid (Betts et al, 2002; Gengenbacher et al, 2010), we hypothesized that the ATP burst triggered by pretomanid in replicating bacilli would be abrogated under non-replicating conditions, leaving only its impact on the oxidative phosphorylation pathway evident. To establish this, we maintained M. tuberculosis H37Rv for 14 days in Dulbecco’s Phosphate Buffer Saline (D-PBS) before treatment with pretomanid or control antibiotics. As anticipated, isoniazid did not have any effect on ATP level in non-replicating mycobacteria, while bedaquiline decreased ATP levels (Fig. 1B), as reported before (Gengenbacher et al, 2010). Consistent with our hypothesis, pretomanid failed to trigger an ATP burst under non-replicating conditions but rather prompted a dose-dependent depletion (Fig. 1B), aligning with direct inhibition of the electron transport chain. Therefore, employing a non-replicating condition offers an attractive model system to dissect the effect of pretomanid on bioenergetics exclusively, devoid of interference from its impact on cell-wall synthesis.Figure 1. Differential effect of pretomanid on ATP levels in replicating and non-replicating M. tuberculosis H37Rv.(A) M. tuberculosis H37Rv was treated with a dose-range of pretomanid (PMD, blue circle), isoniazid (INH, red square), or bedaquiline (BDQ, black circles) for 20 h before intracellular ATP levels quantification. Dotted lines: DMSO. Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. (B) Nutrient-starved non-replicating M. tuberculosis H37Rv was treated with Pretomanid (0.6, 2.5, and 10 μM) or the control drugs BDQ (3 µM) and INH (10 μM) for 24 h before intracellular ATP levels quantification. Experiments were performed in triplicate and n = 2 biological replicates. (C) Structure of the drugs used in this study. Source data are available online for this figure.
Pretomanid inhibits respiration in non-replicating M. tuberculosis
Nitric oxide (NO) is a known inhibitor of terminal oxidases (Antunes et al, 2004; Cooper, 2002; Fang, 1997; Giuffre et al, 2012). It competes with oxygen to bind to heme centers, thereby blocking electron flow and oxygen consumption. Therefore, we hypothesized that pretomanid may inhibit oxygen consumption in M. tuberculosis. Oxygen consumption rates were measured in non-replicating M. tuberculosis on the Seahorse XFe96 platform. Q203 triggered an immediate increase in OCR in H37Rv parental strain (Fig. 2A), as reported before (Lamprecht et al, 2016; Lee et al, 2021). Conversely, pretomanid inhibited OCR in a dose-dependent manner, reducing respiration to approximately 50% of the untreated control (Fig. 2A). Pretomanid inhibited OCR equally in the M. tuberculosis H37RvΔcydAB strain (Fig. 2B) and hampered the Q203-induced increase in respiration (Fig. 2C), indicating that the drug inhibits partially the function of both the cyt -bcc:aa3 and cyt-bd oxidase respiratory branches. To confirm that NO inhibits both terminal oxidases, we studied the effect of the NO-donor NOC-9 on the respiration of purified inverted-membrane vesicles (IMVs). Due to biosafety concerns, we used IMVs from Mycobacterium smegmatis. The OCR of purified membranes from the parental, ΔqcrCAB and ΔcydABDC strains was measured using standard Clarke electrodes. Results showed that NO inhibited OCR equally well in IMVs from the three strains (Fig. EV1), confirming an inhibition of both the cyt-bcc:aa3 and cyt-bd respiratory branches.Figure 2. Pretomanid inhibits respiration in non-replicating M. tuberculosis.(A) Nutrient-starved, non-replicating M. tuberculosis H37Rv was incubated in Seahorse XFe96 plates and treated with DMSO 1% (untreated control, black circles), 100 nM Q203 (green circles), 1.25 μM pretomanid (blue triangles) or 5 μM pretomanid (red squares). Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. (B) M. tuberculosis H37RvΔcydAB was incubated in Seahorse XFe96 plates and treated as described in (A). Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. (C) M. tuberculosis H37Rv was incubated in Seahorse XFe96 plates and treated with 100 nM Q203 (green circles), 100 nM Q203 + 1.25 μM PMD (blue triangles), 100 nM Q203 + 5 μM PMD (red squares), or DMSO 1% (untreated control, black circles). Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. **P value < 0.01, treated vs. untreated (DMSO) conditions. Statistical analyses were performed using an unpaired two-tailed Student’s t test. For P values, see Appendix Table S1. Source data are available online for this figure.
Inhibition of the terminal oxidases potentiate the bactericidal potency of pretomanid in non-replicating M. tuberculosis
Given the inhibitory impact of pretomanid on the OCR, we postulated that pharmacological inhibition of the terminal oxidases will enhance the bactericidal efficacy of the drug. The combination of pretomanid and Q203 resulted in a more extensive depletion of ATP levels than either drug achieved individually in non-replicating M. tuberculosis (Fig. 3A), highlighting a robust combined inhibition of oxidative phosphorylation. Time-kill experiments unveiled that pretomanid exhibited a bactericidal activity against non-replicating M. tuberculosis but could not eradicate the culture (Fig. 3B). Q203 alone was inactive, in accordance with a prior report (Kalia et al, 2017). Pretomanid in combination with Q203 displayed a heightened bactericidal potency, triggering a complete eradication of the bacterial culture (Fig. 3B), demonstrating that inhibition of the cyt-bcc:aa3 potentiates the bactericidal potency of pretomanid. Q203 or PMD alone, as well as their combination (Q203 + PMD), did not exhibit a carryover effect (Appendix Fig. S1). Furthermore, the bactericidal potency of pretomanid was enhanced against the M. tuberculosis H37RvΔcydAB strain (Fig. 3C). The finding that the bactericidal efficacy of pretomanid is significantly augmented by inhibition of either of the aerobic respiratory branches prompted an assessment of the combined impact of a simultaneous inhibition of both terminal oxidases on the potency of pretomanid. Firstly, the potency of the pretomanid-Q203 combination was re-evaluated in the H37RvΔcydAB strain (Fig. 3D). Remarkably, the pretomanid-Q203 combination exhibited a rapid bactericidal potency, achieving over 2 and 4 orders of magnitude reduction in bacterial load within 5 and 9 days, respectively, and cleared the culture within 20 days or sooner (Fig. 3D). Furthermore, addition of the Cyt-bd inhibitor ND-011992 (Lee et al, 2021) to pretomanid + Q203 resulted in a potent triple bactericidal drug combination in M. tuberculosis H37Rv (Fig. 3E), underscoring how simultaneous inhibition of the two terminal oxidases intensifies sensitivity to pretomanid. In line with the results obtained in the H37RvΔcydAB strain, a noticeable synergy between pretomanid and ND-011992 was also observed (Fig. 3E).Figure 3. Inhibition of the terminal oxidases sensitize non-replicating M. tuberculosis to pretomanid.(A) Nutrient-starved, non-replicating M. tuberculosis H37Rv was treated with 100 nM Q203, pretomanid at 2.5 or 5.0 μM (PMD_2.5 and PMD_5, respectively) with (red) and without (blue) 100 nM Q203 for 20 h before intracellular ATP levels quantification. Bedaquiline (BDQ) at 3 μM was used as a positive control (dotted line). Data are expressed as the mean ± S.D. of technical replicates for each condition as shown in source data, n = 2 biological replicates. **P < 0.05 (unpaired two-tailed Student’s t test). (B) Nutrient-starved non-replicating M. tuberculosis H37Rv was treated with DMSO (black circle), Q203 (100 nM, green square), BDQ (1 μM, purple triangle), pretomanid (1.2 μM, blue circle) and pretomanid in combination with 100 nM Q203 (red square). Viability was determined over time by CFU determination on agar plates. Dotted line: Limit of detection; 20 CFU/mL. Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates.**P < 0.005 (Student’s t test). (C) M. tuberculosis H37Rv and H37RvΔcydAB strains were treated with DMSO (black circles: H37RV, blue triangles: H37RvΔcydAB) or 1.2 µM pretomanid (red squares: H37Rv, green circles: ΔcydAB). Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. *P < 0.005 (unpaired two-tailed Student’s t test). (D) The H37RvΔcydAB strain was treated with DMSO (black circle), Q203 (100 nM, green square), pretomanid (1.2 μM, blue circle) and pretomanid + 100 nM Q203 (red square). Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. (E) Nutrient-starved non-replicating M. tuberculosis H37Rv was treated with DMSO (black circle), Q203 (100 nM, black square), ND-011992 (1 μM, black inverted triangle), pretomanid (1.2 μM, green circle), 1.2 μM pretomanid + 100 nM Q203 (blue circle), or 1.2 μM pretomanid + 100 nM Q203 + 1 μM ND-011992 (red circle). ***P *< 0.01 (Student’s t test, PMD + Q203 + ND-011992 vs. PMD + Q203). Dotted line: Limit of detection; 20 CFU/mL. Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. For P values, see Appendix Table S1. Source data are available online for this figure.
Q203 potentiates the bactericidal potency of pretomanid and limits the emergence of resistance in replicating M. tuberculosis
The interaction between pretomanid and Q203 was subsequently assessed in replicating M. tuberculosis H37Rv. Pretomanid alone rapidly killed replicating bacteria, leading to a reduction in bacterial numbers by several orders of magnitude within 2–3 days (Fig. 4A), an expected outcome for a bactericidal drug targeting cell-wall synthesis. However, starting from day 3, the bacterial count exhibited exponential growth, eventually reaching a bacterial density similar to that of the untreated control by day 10 post-treatment (Fig. 4A). Plating pretomanid-treated bacilli on nutrient-agar plates, with or without 10 μM of the drug, revealed a rapid expansion of pretomanid-resistant colonies (Fig. 4B). Isolation and characterization of eight escape mutants revealed high resistance to pretomanid (Table EV1), which was correlated with single-nucleotide polymorphism in fgd1 or fbiB (Table EV1). This underscores that the standalone bactericidal efficacy of pretomanid is rapidly alleviated by the emergence and propagation of resistant bacilli in vitro. Noteworthy, Q203 displayed a slight antagonistic effect on day 5 following co-treatment with pretomanid (Fig. 4A). This observation was attributed to the interference between the energy metabolism inhibitor Q203 and the cell-wall-targeting bactericidal action of pretomanid, resembling the interaction previously described for isoniazid (Lee et al, 2019; Shetty and Dick, 2018; Zeng et al, 2019). However, a notable shift towards synergistic interaction emerged from day 5 onward. Notably, the drug combination of pretomanid + Q203 eradicated the culture within 10-15 days of treatment (Fig. 4A), concurrently preventing the emergence of pretomanid-resistant bacterial strains (Fig. 4B). Fluctuation analyses confirmed the reduced emergence to both Q203 and pretomanid resistance during co-treatment (Table EV2). Moreover, the combination of Q203 + pretomanid exhibited enhanced efficacy against M. tuberculosis H37RvΔcydAB (Fig. 4C) and was further amplified by ND-011992 (Fig. 4D), mirroring the observations made under nutrient-starved conditions.Figure 4Q203 potentiates the bactericidal potency of pretomanid and limits resistance emergence in replicating M. tuberculosis.(A) Replicating M. tuberculosis H37Rv was treated with DMSO (black circle), Q203 (100 nM, green square), pretomanid (PMD, 10 μM, blue circle) or 10 μM pretomanid + 100 nM Q203 (red square). Limit of detection; 20 CFU/mL. Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. (B) Replicating M. tuberculosis H37Rv exposed to 10 μM pretomanid (blue circles) or 10 μM pretomanid + 100 nM Q203 (red squares) was plated on pretomanid-containing agar plates to quantify the emergence of pretomanid resistance. Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. (C) Replicating M. tuberculosis H37RvΔcydAB was treated as described in (A). Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. (D) Replicating M. tuberculosis H37Rv was treated with 1% DMSO (black bar), 1 μM ND-011992 (grey bar), 100 nM Q203 + 10 μM pretomanid (red bar) or a combination of 100 nM Q203 + 10 μM pretomanid + 1 μM ND-011992 (purple bar). Statistical analyses were performed using an unpaired two-tailed Student’s t test. Data are expressed as the mean ± S.D. of triplicates for each condition, n = 2 biological replicates. For P values, see Appendix Table S1. Source data are available online for this figure.
The combination pretomanid-Q203 is potent in vivo
Next, we tested whether the positive in vitro interaction between Q203 and pretomanid could be replicated in vivo. BALB/c mice infected with M. tuberculosis H37Rv for 2 weeks (acute model) were treated for 3 weeks with either pretomanid (20 mg/kg), Q203 (10 mg/kg), or a combination of both at the same concentration. Bacterial burden in the lungs was assessed by counting colony-forming units (CFUs) on agar plates. The combination of pretomanid and Q203 demonstrated significantly greater potency compared to pretomanid alone, albeit to a modest extent (Fig. 5A), which could reflect the relatively high level of cyt-bd expression in M. tuberculosis H37Rv (Arora et al, 2014). To assess the added benefit of cyt-bd oxidase inhibition on the pretomanid–Q203 combination, we used the H37RvΔcydAB strain as a surrogate, since no optimized cyt-bd inhibitor with suitable pharmacokinetic properties is currently available for in vivo testing. Consistent with previous findings, Q203 exhibited much higher potency against the ΔcydAB strain (Kalia et al, 2017), whereas pretomanid maintained comparable efficacy against both strains (Fig. 5B). Aligning with in vitro observations, the pretomanid-Q203 regimen proved particularly potent against the ΔcydAB strain, reducing bacterial load by over one order of magnitude compared to Q203 alone (Fig. 5B). Given the high potency of the dual-drug combination after only three weeks of treatment, we next tested how quickly the dual-drug combination could reduce bacterial burdens to below the limit of detection in lungs and spleens of mice infected with H37RvΔcydAB. To do so, we increased the pretomanid dose to 75 mg/kg, a dose within the range commonly used for in vivo efficacy studies (Mudde et al, 2022) and evaluated bacterial burden in the lungs and spleen after three and 6 weeks of treatment in an established model of infection. The pretomanid-Q203 combination produced a clear additional reduction compared with either drug alone in the lungs after 3 weeks of treatment (Fig. EV2A). At this time point however, no benefit of the combination over single agents was observed in the spleens (Fig. EV2B). After 6 weeks of treatment, Q203 alone reduced lung CFU to below the LOD, consistent with its high potency reported previously (Kalia et al, 2017). Pretomanid alone was potent but unable to reduce bacterial burden below LOD, whereas the pretomanid-Q203 combination was as potent as Q203 alone (Fig. 5C). In the spleen, both pretomanid and Q203 lowered CFU substantially, but only the pretomanid-Q203 combination achieved CFU counts below the LOD (Fig. 5D).Figure 5. The pretomanid-Q203 combination is potent in vivo.BALB/c mice were infected intranasally with ≈5.10^3^ CFU of M. tuberculosis H37Rv (A) or M. tuberculosis H37RvΔcydAB (B). After 2 weeks of infection (acute model), mice were treated 5 times per week with 20 mg/kg pretomanid (PMD, red triangles), 10 mg/kg Q203 (green inverted triangles), a combination of both (purple diamonds), or by the vehicle control (blue squares). Bacterial burden in the lungs was determined by CFU counts on agar plates. BT: bacterial burden Before Treatment. (C, D) BALB/c mice were infected intranasally with ≈5.10^3^ CFU of M. tuberculosis H37RvΔcydAB. After 4 weeks of infection (established model), mice were treated five times per week with 75 mg/kg PMD (red triangles), 10 mg/kg Q203 (green inverted triangles), a combination of both (purple diamonds), or by the vehicle control (blue squares). Bacterial burden in the lungs (C) and spleens (D) was determined by CFU determination on agar plates. Statistical analyses were performed using an unpaired two-tailed Student’s t test. For P values, see Appendix Table S1. Source data are available online for this figure.
Discussion
Following the approval of delamanid in April 2014 (Ryan and Lo, 2014), a second nitroimidazole drug, pretomanid, received US FDA approval in Aug 2019 as part of a fixed-dose combination with bedaquiline and linezolid (Burki, 2019). Although the specific benefits of pretomanid within this combination are still under evaluation, its approval has prompted interest in exploring potential synergies with additional drugs. Our project was motivated by findings showing that agents targeting energy metabolism can compromise the bactericidal effects of cell-wall targeting antibiotics, such as isoniazid (Lee et al, 2019; Zeng et al, 2019). Given the dual mode of action of pretomanid on cell-wall synthesis and energy metabolism, we investigated whether Q203, an inhibitor of the cyt-bcc:aa3 respiratory branch that antagonizes isoniazid in vitro, would enhance or diminish the bactericidal potency of pretomanid. We found that Q203 and pretomanid synergistically improved killing of both replicating and non-replicating M. tuberculosis, suggesting inhibition of the cyt-bcc:aa3 by Q203 increases bacterial sensitivity to nitric oxide-mediated electron transport chain disruption. Since this interaction was reminiscent of the synergy between Q203 and the cyt-bd inhibitor ND-011992 (Lee et al, 2021), we initially assumed that pretomanid inhibited cyt-bd via nitric-oxide release. However, further mode of action studies indicated that pretomanid inhibits partially both aerobic respiratory branches. This conclusion was reached by the findings that pretomanid inhibited the Oxygen Consumption Rate equally well in the M. tuberculosis H37Rv parental and ΔcydAB strains, indicating that the drug inhibits the cyt-bcc:aa3 respiratory branch. Furthermore, the observation that pretomanid also inhibits the OCR increase triggered by Q203 also suggests partial inhibitory effect on the cyt-bd respiratory branch. Although we did not directly quantify pretomanid-derived nitric oxide under our experimental conditions, the pattern of partial inhibition across both respiratory branches remains most consistent with a nitric oxide-mediated mechanism acting at the level of the heme-containing terminal oxidases. This finding, along with the observed synergy in pairwise combinations of pretomanid/Q203 and pretomanid/ND-011992, supports the idea that pretomanid affects both cyt-bcc:aa3 and cyt-bd branches, although nitric oxide may also interfere with additional electron transport components. One of the most striking findings of this study is the heightened bactericidal efficacy of the pretomanid/Q203 combination against the H37RvΔcydAB, which led to the validation of the rapid bactericidal potency of the triple drug combination pretomanid-Q203-ND-011992 against replicating and antibiotic-tolerant, non-replicating M. tuberculosis. Surprisingly, we observed that the emergence of resistance to pretomanid is remarkably rapid in replicating M. tuberculosis. In a classical kill kinetic study, pretomanid initially displayed strong bactericidal activity, yet selection of resistant clone was observed after few days and ultimately dominates the population after 7 to 10 days of treatment. This swift emergence of resistance can be attributed to the dispensability of genes linked to pretomanid bioactivation (ddn and fgd1) or of the associated F_420_ biosynthetic pathway under in vitro conditions (Haver et al, 2015). In the case of the first-line prodrug isoniazid, resistance emergence mediated by KatG loss of function mutations, rapidly selected in vitro, is rarely observed in animal models or humans, owing to the essentiality of the catalase/peroxidase during infection (Laurent et al, 2020; Li et al, 1998). While the essentiality of the genes involved in pretomanid activation is not yet well-established during infection, the fitness cost associated with the loss of function of the F_420_ biosynthetic pathway, ddn, and fgd1 genes warrants investigation, as it could significantly impact the clinical usefulness of pretomanid. Remarkably, Q203 alone and in combination with ND-011992 prevented the emergence of pretomanid-resistant clones. This observation underscores that the triple drug combination not only exhibits rapid bactericidal potency but also possesses a reduced tendency for the emergence of resistance, a crucial characteristic for an effective tuberculosis drug combination. In mice, the combination of pretomanid and Q203 showed significant potency over either agent alone, reducing bacterial loads in M. tuberculosis-infected mice and proving especially effective against the H37RvΔcydAB strain. Interestingly, a recent report suggest that this combination might work even better against clinical isolates than against H37Rv (Komm et al, 2024), which may be linked to a lower level of cyt-bd expression relatives to laboratory-adapted strains. The increased potency against the ΔcydAB strain highlights the therapeutic value of optimized cyt-bd inhibitors with improved pharmacokinetic profiles. Such inhibitors would enable a more thorough validation of the therapeutic potential of cyt-bd inhibition within this combination, paving the way for a rapidly sterilizing drug regimen backbone. In conclusion, our findings advocate for further exploration of pretomanid in combination with terminal oxidase inhibitors to achieve rapid bactericidal action against both replicating and antibiotic-tolerant M. tuberculosis, with the potential to shorten treatment time, a critical need in TB management.
Methods
Reagents and tools tableReagent/resourceReference or sourceIdentifier or catalog number Experimental models Mycobacterium tuberculosis H37RvATCCATCC 27294M. tuberculosis H37RvΔcydABKalia et al, 2017N/AM. tuberculosis H37RvΔcydABcompKalia et al, 2017N/AM. smegmatis mc^2^155ATCC700084M. smegmatis mc^2^155ΔcydABDCBayly et al, 2021N/AM. smegmatis mc^2^155ΔqcrCABBayly et al, 2021N/ABALB/c mice (Mus musculus) InVivos Pte Ltd N/A Recombinant DNA N/A Antibodies N/A Oligonucleotides and other sequence-based reagents
PCR primer
Forward (5′-3′)
Reverse (5′-3)
fbiB GCCGCTGCTGATGACCGATCGGGAGGTTGATGGTTGG Fgd1 GTGGCCGCGAGCGAGGTGAACGCCCGAACCGTCAACAACACTGG Chemicals, enzymes and other reagents Bedaquiline (BDQ)GVK BioCustom synthesisBovine Serum Albumin (BSA)Capricorn ScientificBSA-1SCorning® Cell-Tak(TM) cell and tissue adhesiveCorningCLS354240D-glucose (dextrose)1st baseBIO-1101DNase IThermoScientific™EN0521DPBS, no calcium, no magnesiumThermoFisher Scientific14190144D-α-Tocopherol polyethylene glycol 1000 succinate (TPGS)Sigma-Aldrich57668gentleMACS™ C tubesMiltenyi Biotec130-093-237HEPESSigma-Aldrich54457Hygromycin BNacalai Tesque09287-84IsoniazidSigma-AldrichI3377KClSigma-AldrichP9541KOHSigma-Aldrich221473Middlebrook 7H10BD Difco262710Middlebrook 7H9BD Difco271310Middlebrook OADC enrichmentBD Difco212351MOPSSigma-AldrichM1254NADH Grade II, disodium saltSigma-Aldrich10128023001ND-011992In-house synthesisN/ANOC-9Santa Cruz Biotechnology, Incsc-202250OADCBecton Dickinson212351Pierce™ Protease Inhibitor Tablets, EDTA-freeThermo Fisher ScientificPIA32965Pretomanid (PMD)ApeXBio#A1736Sodium chlorideSigma-AldrichS3014Telacebec (Q203)GVK BioCustom synthesisTween® 80Sigma-AldrichP8074 Software Graphpad Prism 6Graphpad Software, LLCRRID: SCR_002798Oxytrace+Hansatech Instruments Ltd https://www.hansatech-instruments.com Wave Desktop 2.6.0.31Agilent Technologies https://www.agilent.c0m
Other Agilent Biotek Cytation 3Agilent TechnologiesRRID:SCR_013575BacTiter-GloPromegaG8231GEA PandaPLUS 2000GEAgentleMACS™ DissociatorMiltenyi Biotec130-093-235Oxygraph+ SystemHansatech InstrumentsPierce™ BCA Protein Assay KitsThermo Scientific™23225Seahorse XFe96 AnalyzerAgilent TechnologiesSeahorse XFe96/XF Pro FluxPakAgilent Technologies103792-100
Bacterial strains and culture conditions
Mycobacterium tuberculosis H37Rv and its derivative strains, H37RvΔcydAB and H37RvΔcydABcomp (Kalia et al, 2017), were cultivated at 37 °C in Middlebrook 7H9 broth supplemented with 0.5% glycerol, 0.05% Tween-80, and ADS growth enrichment for liquid cultures. Solid cultures were grown on Middlebrook 7H10 agar supplemented with Middlebrook OADC enrichment. For experiments conducted under replicating conditions, bacterial strains were harvested during early exponential phase, washed twice with glycerol-free Middlebrook 7H9, and adjusted to specific cell densities. The cultures were then adjusted to specific cell densities required for subsequent analyses. In experiments designed to mimic nutrient starvation conditions, bacterial strains were harvested at early exponential phase, washed twice with Dulbecco’s Phosphate Buffered Saline (D-PBS) supplemented with 0.025% Tween-80. The cultures were then adjusted to specific cell densities and incubated for a period of 7–14 days at 37 °C to induce a non-replicating state. All experiments involving M. tuberculosis were conducted within a certified biosafety level 3 laboratory.
Chemical compounds and synthesis
Telacebec (Q203) and Bedaquiline (BDQ) were custom synthesized at Aragen Life Sciences (formerly GVK BIO). Pretomanid (PMD) was purchased from ApexBio Technology. Isoniazid (INH) and other chemical reagents were procured from Sigma-Aldrich or ThermoFisher Scientific, unless otherwise specified. The compound ND-011992 was synthesized in-house.
Minimum inhibitory concentration (MIC) determination
The MIC determination assay was conducted following an established procedure (Pethe et al, 2013; Pethe et al, 2010). In brief, an inoculum with an optical density at 600 nm (OD_600_) of 0.005 was prepared and dispensed into 96-well flat-bottom plates at 200 μL per well. Subsequently, 2 µL of serially diluted drugs were added to each well. The plates were then incubated at 37 °C for five days to allow bacterial growth and drug interaction. Optical density measurements were acquired using a BioTek Cytation 3 Cell Imaging Multi-Mode Reader. The MIC_50_ values, representing the drug concentrations at which 50% inhibition of bacterial growth occurred in a full dose-response curve, were determined and analyzed. GraphPad Prism 7 software was employed for data analysis, visualization, and MIC_50_ determination.
ATP levels quantification
Intracellular ATP levels were measured using the BacTiter-Glo™ Microbial Cell Viability Assay (Promega, USA). An inoculum at an OD_600_ of 0.05 was prepared and dispensed into 96-well white plates at 100 μL/well. Serial-diluted drugs were then dispensed into each well at the desired concentrations with the DMSO concentration maintained at 1% across all wells. The plates were incubated at 37 °C for 15 h, after which 25 µL of the BacTiter-Glo reagent was dispensed into each well. Following a 12-minute incubation, the luminescence was recorded on the BioTek Cytation 3 Cell Imaging Multiple-mode reader. The assay was performed similarly for non-replicating cultures but at an OD_600_ of 0.15.
M. tuberculosis viability assays
An inoculum at an OD_600_ of 0.005 was prepared and dispensed into 24-well flat-bottom plates. Drugs were added to each well, and the plates were incubated at 37 °C for 14–21 days. Bacterial viability was determined by enumeration of colony-forming units (CFU) on agar plates. For non-replicating cultures, the assay was performed similarly but using an inoculum at an OD_600_ of 0.02. Data analysis was conducted using the GraphPad Prism 7 software.
Oxygen consumption assay
The basal oxygen consumption rates (OCR) were assessed using the Seahorse XFe96 Analyzer (Agilent). Nutrient-starved bacteria were maintained for seven days with gentle re-suspension every 2–3 days to ensure adequate oxygenation of the culture before recording OCR. The Agilent Seahorse cell culture microtiter plates were coated with 22.5 μg/mL Cell Tak (Corning®), and cells were dispensed into each well at 100 μL per well, followed by centrifugation to enhance cell adherence to the wells. Test compounds were injected into the respective injection ports, and oxygen consumption was recorded at 3-minute intervals post-mixing. Data obtained from the Seahorse XFe96 Analyzer were analyzed using the Wave Desktop 2.6.0.31 and GraphPad Prism 7 software.
Evaluation of drug efficacy in mice
Work with animals was approved by the Nanyang Technological University Institutional Animal Care and Use Committee (approval #A22091). Female 6–8 weeks old BALB/c mice were infected via the intranasal route with 4.10^3^ bacilli suspended in 30 µl PBS-Tween-80. Treatment was initiated 2 weeks post-infection. Drugs or drug combinations were formulated in 1% DMSO-20% D-α-Tocopherol polyethylene glycol 1000 succinate (TPGS). Q203 and pretomanid were administered at a dose of 10 and 20 mg/kg body weight, respectively, five times a week via oral gavage (PO). Lungs were harvested 3 weeks after treatment initiation and homogenized using the Miltenyi MACS using gentleMACS™ C tubes. Lung homogenates were immediately serial-diluted and plated on 7H10 plates. Agar plates were incubated at 37 °C for 3–4 weeks before CFU enumeration. Five mice per group and per condition were used.
Preparation of inverted membrane vesicles (IMVs)
The method developed by Hotra et al (Hotra et al, 2016) was used for IMVs preparation. M. smegmatis was grown to exponential phase (OD_600_ ∼0.3–0.6) in Erlenmeyer Flask at slow shaking of 60 rpm. A total of 5 grams biomass was harvested by centrifugation, followed by resuspension in 20 mL membrane preparation buffer (50 mM MOPS, 2 mM MgCl_2_, pH 7.5) containing EDTA-free protease inhibitor cocktail (Pierce, USA) and 1.2 mg/mL lysozyme. Suspension was stirred at room temperature for 45 min and supplemented with 300 µl of 1 M MgCl_2_ solution and 50 µl DNase I. Stirring was continued for 15 min. Cell lysis was achieved by 3 passages through prechilled high pressure homogenizer (GEA PandaPLUS 2000) at a pressure of 1400–1500 pressure. The suspension was centrifuged at 4200 rpm for 20 min to remove cell debris, followed by additional centrifugation step at 15,200 rpm for 40 min to further clean the suspension. The supernatant was finally ultracentrifuged at 37,100 rpm for 1 h. Pelleted IMVs were gently scraped from the tubes and resuspended in the membrane preparation buffer supplemented with 15% glycerol, aliquoted, snap frozen using liquid nitrogen and stored in −80 °C for further use. The protein concentration of the vesicles was determined by the bicinchoninic acid assay (BCA; Pierce, Rockford, IL, USA).
Oxygen consumption assay in IMVs
Oxygen consumption in M. smegmatis WT (parental strain) and strains deficient in cyt-bd and cyt-bcc:aa3 (Bayly et al, 2021) was measured using Clark type oxygen electrode (Oxygraph, Hansatech). Briefly, IMVs were diluted to 0.18 mg/mL total protein concentration in IMV OCR buffer (10 mM HEPES/KOH at pH 7.5, 100 mM KCl, 5 mM MgCl_2_) in a total volume of 0.5 mL at ambient temperature in an airtight chamber. IMVs were energised with 1 mM NADH. The nitric oxide donor NOC-9 (MAHMA NONOate) at 125 µM was added when oxygen levels were depleted by nearly half (compared to oxygen level prior to NADH addition). Oxygen consumption rate was measured as the best fit slope of oxygen signal between the rate intervals using OxyTrace+ software.
Fluctuation assay
Fluctuation assay was performed as described previously (Ford et al, 2013). Log-phase culture of M. tuberculosis H37Rv was diluted to ~10^2^–10^3^ cells/mL and separated into 20 inkwell bottles containing 10 mL of culture each. Cultures were left to grow for 18 days until OD_600_ reaches 0.8–1.0. The 20 cultures were concentrated by centrifugation to 0.5 mL. Each concentrated culture was then plated after appropriate dilution on 7H10 agar plates supplemented with test drug or drug combination or used for bacterial enumeration. Agar plates were then incubated for at least 21 days before colony counting. Q203 was tested at 500 nM and pretomanid at 1 µM PMD.
Graphics
The synopsis illustration was created with BioRender.com https://BioRender.com/1co2kuz.
Statistical analysis
Statistical analyses were performed using GraphPad Prism. For normally distributed data, comparisons were performed using a two-tailed unpaired t test or multiple t tests. Data are expressed as the means ± standard deviation (S.D.).
Supplementary information
Table EV1 Table EV2 Appendix Peer Review File Source data Fig. 1 Source data Fig. 2 Source data Fig. 3 Source data Fig. 4 Source data Fig. 5 Expanded View Figures
The reference list from the paper itself. Each links out to its DOI / PubMed record.