Synergistic Effect of Q203 Combined with PBTZ169 against Mycobacterium tuberculosis

ABSTRACT

Q203 is a first-in-class drug candidate against Mycobacterium tuberculosis. In its recently completed phase 2 clinical trial, Q203 reduced the number of live M. tuberculosis cells in a dose-dependent manner. This orally active small molecule blocks M. tuberculosis growth by inhibiting the cytochrome bc₁ complex, which consequently inhibits the synthesis of ATP. Here, we studied the interaction profiles of Q203 with several antituberculosis drugs or drug candidates (specifically, bedaquiline, PBTZ169, PA-824, OPC-67683, SQ109, isoniazid, rifampin, streptomycin, and linezolid) using the checkerboard method, based on resazurin microtiter assays (REMAs). In the assay, none of the interactions between Q203 and the tested drugs were antagonistic, and most of the interactions were additive. However, the interaction between Q203 and PBTZ169 was synergistic, with a fractional inhibitory concentration index of 0.5. Furthermore, Q203 (one-half the MIC₅₀) and PBTZ169 (one-half the MIC₅₀) inhibited more bacterial growth on an agar plate compared to the dimethyl sulfoxide (DMSO) control. This synergistic effect was no longer effective when the Q203-PBTZ169 combination was tested against an M. tuberculosis mutant containing a T313A mutation causing resistance to Q203, suggesting that QcrB inhibition is integral to the Q203-PBTZ169 interaction. Thus, this synergy is not an off-target mechanism. Zebrafish (Danio rerio)-Mycobacterium marinum infection and a curing model further validated the synergistic effect of Q203 and PBTZ169 in vivo. In this study, the synergy between these two new antituberculosis drugs, Q203 and PBTZ169, is an important finding that could lead to the development of a new TB regimen.

INTRODUCTION

Tuberculosis (TB) is a deadly infectious disease caused by the Mycobacterium tuberculosis complex. M. tuberculosis is carried in airborne particles, called droplet nuclei, and is spread from person to person by coughing and sneezing (1). In 2020, approximately 10 million people fell ill with TB worldwide, and 1.5 million people died from TB, including 214,000 people with human immunodeficiency virus (HIV) (2). Globally, TB is the 13th leading cause of death and the second leading infectious killer after the coronavirus disease 2019 (COVID-19) pandemic (2). For these reasons, the WHO adopted the End TB Strategy, which aims to end the global TB epidemic by 2035. This strategy strives to reduce TB mortality by 95% and TB incidence by 90% between 2015 and 2035 and to have no TB-affected families face catastrophic treatment costs by 2020. However, despite these efforts, the cumulative reduction of TB incidence was only 11% between 2015 and 2020. This is slightly more than half of the End TB Strategy milestone of 20% reduction (3).

TB can be treated by taking several drugs for 6 to 9 months. Of the approved drugs, the first line of anti-TB agents includes a four-drug combination of isoniazid (INH), rifampin (RIF), pyrazinamide (PZA), and ethambutol. However, drug resistance arises when patients receive inadequate or interrupted administration of first-line treatment (4). Multidrug-resistant TB (MDR-TB) is caused by M. tuberculosis that is resistant to at least isoniazid and rifampin, the two most potent and important TB drugs. Approximately 5% of TB cases globally are estimated to be resistant to both isoniazid and rifampin. Second-line drugs such as levofloxacin, moxifloxacin, bedaquiline, delamanid, and linezolid are used to treat MDR-TB. Furthermore, of the estimated cases of MDR-TB, approximately 10% are extensively drug-resistant (XDR) TB caused by bacteria that is resistant to isoniazid and rifampin, plus any fluoroquinolone and at least one of three injectable second-line drugs (i.e., amikacin, kanamycin, or capreomycin) (5). The treatment of MDR-TB requires prolonged administration of toxic second-line antituberculosis drugs, and it generally has a poor outcome and a high cost. XDR-TB requires treatment that is even more complex and has a higher mortality.

Q203 (Telacebec) is a novel first-in-class anti-TB drug candidate that targets M. tuberculosis cellular energy production by inhibiting the mycobacterial cytochrome bc₁ complex with high specificity. This complex is a critical component of the electron transport chain, and its inhibition disrupts a bacterium’s ability to generate energy. In an in vitro study, depletion of ATP synthesis by Q203 resulted in the inhibition of bacterial growth in active growing and hypoxic conditions (6). Furthermore, Q203 showed excellent activity against MDR and XDR-TB isolates (6). A recent phase 2 clinical trial demonstrated that increasing doses (up to 300 mg) of Q203 could reduce the viable mycobacterial sputum load. Q203 was associated with acceptable adverse-event rates, and these adverse events were distributed equally among all groups (7). Furthermore, Q203 also showed an extremely short treatment duration for Buruli ulcer in a mouse footpad infection model, consequently warranting further evaluation in clinical trials (8, 9).

However, Q203 appears to be bacteriostatic against M. tuberculosis H37Rv, even when applied at extremely high concentrations (10, 11). Thus, it requires association with another drug to produce a synergistic effect to kill M. tuberculosis. In this study, we employed Q203 as an anchor compound in combination with current anti-TB drugs. This work revealed that the addition of PBTZ169 to Q203 significantly synergized the activity in vitro against actively growing mycobacteria.

RESULTS

MIC evaluation and checkerboard assay for compound interactions.

In order to find the drug combination with the highest synergistic effect with Q203, various first-line TB drugs, second-line TB drugs, and drug candidates in the anti-TB drug discovery pipeline were tested. The MIC value of each drug was determined by resazurin microtiter assay (REMA). MIC₅₀ was defined as the MIC required to inhibit M. tuberculosis growth by 50%. The MIC₅₀ values of all tested drugs are presented in Table 1. As shown in a previous report, Q203 showed favorable activity against M. tuberculosis (MIC₅₀, 2.5 nM) (6).

TABLE 1.

MICs of selected anti-M. tuberculosis drugs against M. tuberculosis mc² 6230 and the corresponding interaction profiles with Q203

Compound	Assigned name	MIC50 (μM)a	Interaction profile with Q203
			ΣFIC	Outcome
Q203	Telacebec	0.0025
Isoniazid		0.40	1.4	Additive
Rifampin		0.0034	1.2	Additive
Streptomycin		0.27	1.4	Additive
Moxifloxacin		0.20	3.2	Antagonism
Linezolid		1.30	1.3	Additive
OPC-67683	Delamanid	0.30	1.3	Additive
Bedaquiline		0.5	3.1	Antagonism
PA-824	Pretomanid	0.30	3.1	Antagonism
PBTZ169	Macozinone	0.31	0.5	Synergistic
SQ109		0.70	3.1	Antagonism

^aEvaluated using the REMA checkerboard.

The effect of the drugs in combination with Q203 were evaluated to find the best drug pairing. As illustrated by Fig. 1, the combination of Q203 and PBTZ169 showed synergistic effects at a much lower MIC than each drug’s independent MIC. For example, one-half the MIC₅₀ of Q203 (1.25 nM) added to one-half the MIC₅₀ of PBTZ169 (0.15 μM) did not result in a resazurin color change from blue to pink, indicating that M. tuberculosis growth was inhibited (Fig. 1A). Furthermore, one-quarter the MIC₅₀ of Q203 added to one-quarter the MIC₅₀ of PBTZ169 also inhibited M. tuberculosis growth. A fractional inhibitory concentration (FIC) index was used to evaluate the synergistic effects of the Q203-PBTZ169 combination ratios against M. tuberculosis (Table 1). Synergy was defined as a fractional inhibitory concentration (FIC) index of 0.5 or less (12). Thus, the Q203-PBTZ169 combination is synergistic against M. tuberculosis. However, Q203 combined with INH, RIF, streptomycin (STR), moxifloxacin (MXF), linezolid (LZD), OPC-67683, bedaquiline (BDQ), PA-824, or SQ109 showed no synergistic anti-M. tuberculosis effect, with FIC index (FICI) values of 1.2 to 3.2 (Table 1). Conversely, FICI values of >2 were observed, indicating potential antagonism between Q203 and the tested MXF, BDQ, PA-824, or SQ109. Moreover, an additive effect was observed in combinations of Q203 with INH, RIF, STR, LZD, and OPC-67683 (Table 1; see Fig. S1 in the supplemental material).

FIG 1.

Drug-drug interactions using a checkerboard assay. (A, C, E) The drug interactions were evaluated using the MIC₅₀, one-half the MIC₅₀, one-quarter the MIC₅₀, and one-eighth the MIC₅₀ of Q203 in combination with the MIC₅₀, one-half the MIC, one-quarter the MIC₅₀, and one-eighth the MIC₅₀ of PBTZ169 (A), INH (C), and MXF (E). The dotted white lines and red text indicate the MIC₅₀ value of each compound. (B, D, F) Isobolograms of the resazurin checkerboard synergy testing method showing the synergy of Q203 with PBTZ169 (B), INH (D), and MXF (F). An additive effect was observed when Q203 interacted with INH (D) and MXF (F).

In vitro live/dead evaluation using CFU determination for drug-drug interaction.

To confirm whether the Q203-PBTZ169 synergistic effect kills M. tuberculosis, a traditional CFU determination test was carried out. As shown in Fig. 2A, the CFU evaluation of Q203-PBTZ169 was consistent with the results of the checkerboard assay. Q203-PBTZ169 showed a clear synergistic effect resulting in a significant reduction of live M. tuberculosis numbers on the 7H10^G/C/P/OADC (7H9 supplemented with 0.2% glycerol [G], 0.2% Casamino Acids [C], 24 μg/mL pantothenate [P], and Middlebrook oleic acid-albumin-dextrose-catalase [OADC]) agar plates. The results show that a combination of 1.25 nM Q203 (one-half the MIC₅₀) and 0.15 μM PBTZ169 (one-half the MIC₅₀) had clear growth inhibitory activity (4.7 log₁₀ CFU/mL reduction) compared to the activity of the DMSO control on day 7 (Fig. 2A, black bar). Conversely, Q203 combined with INH acted additively, with each combination giving similar bacterial viability inhibitions as their single agents (Fig. 2B), and Q203 combined with MXF showed an antagonistic effect that may have blocked or reduced the effectiveness of one of the drugs (Fig. 2C).

FIG 2.

Evaluation of bacterial growth inhibition by CFU counts. M. tuberculosis mc² 6230 was grown in the presence of different concentrations of Q203 or Q203 in combination with decreasing concentrations of PBTZ169 (A), INH (B), and MXF (C). The compound activity was determined by counting the CFU. A one-way analysis of variance (ANOVA) with Tukey’s multiple-comparison test was used to compare the means across multiple groups. ***, P < 0.001; ns, not significant. The gray and black bars show results for the untreated controls.

To understand whether the synergistic effect of Q203-PBTZ169 was caused by the activity of Q203 on its target, we generated a Q203-resistant mutant. The Q203-resistant mutant harbors a T313A mutation on qcrB and showed strong resistance against Q203 (Fig. 3A). However, the Q203-resistant mutant and its mother strain showed the same inhibitory curves against BDQ in a dose-dependent manner (Fig. 3B). Thus, the mutant strain showed specific resistance to Q203. To know whether the mechanism of synergy observed between Q203 and PBTZ169 could be due to on- or off-target effects, the synergistic Q203-PBTZ169 interaction was reanalyzed using a CFU determination assay with the M. tuberculosis mutant (qcrB_T313A). As shown in Fig. 3C, the synergistic effect was no longer observed when Q203 combined with PBTZ169 was used against the M. tuberculosis mutant (qcrB_T313A), unlike for its mother strain. Thus, we suggest that synergy was caused by the activity of Q203 on its target, QcrB, and inhibition of QcrB is essential for the synergy.

FIG 3.

Impact of QcrB inhibition on the synergy between Q203 and PBTZ169. (A, B) To determine whether the synergy between Q203 and PBTZ169 was due to the inhibition of QcrB by Q203 or to an off-target effect, a Q203-resistant strain (T313A) was generated under a high concentration of Q203. The Q203-resistant strain showed significant drug resistance against Q203 (A) but not BDQ (B). (C) The activities of combinations of Q203 and PBTZ169 were tested on M. tuberculosis mc² 6230 and on Q203-resistant M. tuberculosis mc² 6230, which carried a point mutation in QcrB (T313A). The compound activity was determined by counting the CFU. A one-way ANOVA with Tukey’s multiple-comparison test was used to compare the means across multiple groups. UNT, untreated.

Synergistic effect of Q203-PBTZ169 in Mycobacterium marinum-infected zebrafish.

We further verified the synergistic effect of Q203 with other anti-TB compounds in zebrafish (ZF) infected with Mycobacterium marinum carrying pTEC27 (plasmid 30182; Addgene) and expressing the red fluorescent protein, tdTomato. For this, the concentrations necessary for each drug to reduce the CFU by 1 log₁₀ in infected ZF compared to the untreated control were determined (Fig. 4A). The tested concentrations were defined as follows: Q203 (12.5 μM), INH (12.5 μM), RIF (12.5 μM), BDQ (1.5 μM), and PBTZ169 (5 nM). For the next step, an in vivo combination drug efficacy test was carried out with the determined concentrations of each drug. Q203 was used as the main drug and was paired separately with INH, RIF, BDQ, and PBTZ169. After infecting ZF larvae with M. marinum-tdTomato, treatments with each drug combination were carried out for 5 days. The progression of the tdTomato-labeled bacteria in the ZF after drug combination treatment was observed using an ImageXpress Pico automated cell imaging system. ZF treated with Q203 (12.5 μM) plus PBTZ169 (5 nM) showed a significant synergistic effect in the checkerboard assay compared to ZF treated with the other drug combinations and the control. The tdTomato-labeled M. marinum dissemination was observed throughout the bodies of the ZF in the nontreated DMSO control group (Fig. 4B). As shown in Fig. 4B, ZF treated with Q203 (12.5 μM) plus PBTZ169 (5 nM) showed almost no tdTomato fluorescence. However, the tdTomato signals were still observed in the other Q203 combinations. Therefore, these results indicate that the combination of Q203 and PBTZ169 significantly inhibited M. marinum growth in the ZF.

FIG 4.

Evaluation of the Q203-PBTZ169 synergistic effect in a ZF infection and treatment model. (A) The drug concentrations that reduced the CFU by 1 log₁₀ compared to DMSO were determined using different concentrations of Q203, INH, RIF, BDQ, and PBTZ169 in the M. marinum-infected ZF model. (B) Q203 (12.5 μM) was combined with PBTZ 169 (5 nM), INH (12.5 μM), RFB (12.5 μM), and BDQ (1.5 μM). Each drug combination was used to treat ZF infected with M. marinum-tdTomato, and the reduction of the tdTomato signal in ZF was monitored under a fluorescence microscope. (C) Therapeutic outcomes using the different drug combinations were validated using the traditional agar plate quantification method. Data are expressed as the mean log₁₀ CFU per embryo (n = 10 for each condition) from three independent experiments. (D) To determine the in vivo efficacy, a survival curve was plotted from M. marinum-tdTomato-infected ZF for 13 days (n = 20, representative of three independent experiments). Survival curves were constructed using the log-rank (Mantel-Cox) test. ***, P < 0.001. Infect/nontreated, infected but not treated control.

Next, bacterial survival in the ZF was enumerated by CFU counts after the different compound treatments. To determine whether each Q203 combination effectively reduced bacterial survival in ZF, the bacterial burden was compared with that for the Q203 (12.5 μM) plus PBTZ169 (5 nM) treatment and the other Q203 combinations (INH, RIF, and BDQ), including the nontreated DMSO control. For this, each infected and treated ZF was homogenized, and the number of bacteria was counted on a 7H10^G/OADC agar plate containing BBL MGIT PANTA antibiotic mixture. As shown in Fig. 4C, the Q203 at 12.5 μM along with PBTZ169 (5 nM) was sufficient to prevent the growth of viable M. marinum cells within the ZF. The Q203-PBTZ169 treatment showed an approximately 3.3 log₁₀ CFU reduction compared with the DMSO control group. Similarly, Q203 in combination with RIF caused the arrest of growth and reduced the mycobacterial burden (2.6 log₁₀ CFU reduction). In contrast to the combination of Q203-PBTZ169 or that of Q203-RIF, the combination of Q203 along with INH and Q203 plus BDQ only slightly reduced the CFU of M. marinum.

Next, we monitored the M. marinum-infected ZF survival after treatment with Q203 drug combinations. As shown in the Kaplan-Meier survival curve (Fig. 4D), treatment with Q203 (12.5 μM) plus PBTZ169 (5 nM) yielded a significantly higher survival rate than that for the other combinations. The Q203-PBTZ169 combination led to a 68.5% survival rate 13 days postinfection. Similarly to the CFU counts, double therapy with Q203 in combination with RIF yielded a lower mortality rate than the other combinations. The Q203 plus RIF combination led to a 45% mortality rate at 13 days after treatment. In contrast, Q203 plus INH or BDQ yielded lower survival rates (19.5% and 28.6% at 13 days postinfection [dpi], respectively). The M. marinum-infected and DMSO-treated ZF groups showed a 100% mortality rate, and the noninfected group had a 100% survival rate after 13 days. These results were consistent with those from the CFU determination (Fig. 3). Together, these data show that the inhibitory effects of Q203 are enhanced in combination with PBTZ169 or RIF in an in vivo ZF model.

DISCUSSION

TB requires at least 6 months of treatment, and MDR-TB and XDR-TB require longer treatment durations, ranging from 18 to 24 months after sputum smear and culture conversion (13, 14). Therefore, it is necessary to find a new regimen that shortens the duration of TB treatment. A possible first step for finding this new regimen is in vitro studies that combine current anti-TB drugs with drug candidates that are in clinical trials. In this context, Lechartier et al. reported that BTZ043 (the former lead compound of PBTZ169) acts synergistically with TMC207 (the former name of BDQ) via a checkerboard assay (13). They showed that TMC207 used in combination with BTZ043 at one-quarter the MIC of each compound demonstrated a stronger bactericidal effect on M. tuberculosis than TMC207 at the MIC. Makarov et al. expanded on this study by using a long-term M. tuberculosis-infected mouse model (15). An excellent synergistic effect was shown when PBTZ169 interacted with BDQ in the mouse model of chronic TB. In the study, the combination of PBTZ169 and BDQ showed a very good synergistic effect compared to the standard regimen of RIF, INH, and PZA currently used for TB treatment by reducing the number of CFU in the lungs and spleen. Furthermore, PBTZ169 anchored with BDQ or PZA, or both drugs (triple therapy), also showed superior efficacy, as it reduced the bacterial load more rapidly than the standard regimen in long-term treatment. The authors hypothesize that PBTZ169 (or BTZ043) weakens the bacterial cell wall by targeting decaprenylphosphoryl-beta-d-ribose oxidase (DprE1) and allowing for improved penetration of BDQ (TMC207) to its target, ATP synthase (AtpE).

Q203 inhibits the cytochrome bc₁ complex, leading to the depletion of ATP synthesis in M. tuberculosis, and it has successively completed phase 2 of clinical trials (6, 16). However, Q203 is a bacteriostatic compound due to the presence of an alternate terminal oxidase, the cytochrome bd oxidase. Furthermore, a previous report demonstrated that Q203 showed an antagonistic effect with isoniazid and moxifloxacin, respectively (17). In this context, we further expanded the number of test drugs to find antituberculosis drugs that can synergize with Q203 (17, 18). Here, we performed a checkerboard assay to find the best drug paired with Q203 in vitro and in a ZF model. Q203 was paired with anti-TB drugs and TB drug candidates that are in clinical trials. As presented in Table 1, Q203 showed a unique synergistic effect with PBTZ169. One-half the MIC₅₀ (1.25 nM) of Q203 in combination with one-half the MIC₅₀ (0.15 nM) of PBTZ169 had a stronger bacterial growth inhibitory effect on M. tuberculosis than Q203 alone at a concentration of 1.25 nM. This synergistic Q203-PBTZ169 combination showed a 2.7 log₁₀ CFU/mL reduction compared to PBTZ169 alone (Fig. 2A). Based on the definition, bactericidal activity was defined as a reduction of ≥3 log₁₀ of the total count of CFU/mL in the original inoculum. Therefore, the Q203 plus PBTZ169 combination was shown to be bacteriostatic against M. tuberculosis. Interestingly, this synergistic effect was not present when Q203-PBTZ169 was tested against a Q203-resistant M. tuberculosis mutant with a T313A mutation in qcrB. This result eliminated the possibility of the synergistic effect occurring due to an off-target mechanism (Fig. 3C). We also hypothesize that inhibition of DprE1, which is an important enzyme for penetration of the arabinogalactan biosynthetic pathway of the M. tuberculosis cell wall by PBTZ169, increases the bacterial cell wall permeability, allowing Q203 to penetrate and reach its target, thereby allowing the synergistic effect to occur, as Lechartier et al. reported previously (13). Furthermore, PBTZ169 has shown synergy with drugs that act on components in the electron transport chain, such as BDQ and clofazimine (CLO) (19). CLO is a prodrug, which is reduced by NADH dehydrogenase (NDH-2) to release reactive oxygen species (ROS) (20). Conversely, when lansoprazole (LPZ), which like Q203 targets the QcrB component of cytochrome bc₁ oxidase, was combined with PBTZ169 in vitro, the two were not synergistic (19). Further studies will be needed to fully understand the mechanisms of LPZ and Q203 underlying their different effects in combination with PBTZ169.

This new finding was further validated using a ZF larvae M. marinum infection model at subtherapeutic concentrations of the individual agents. For this, we determined all the compound concentrations that reduced the CFU by 1 log₁₀ in the infected ZF (Fig. 4A). Using an in vivo early embryo infection model, we injected M. marinum into the caudal veins of ZF and initiated Q203 treatment in combination with various anti-M. tuberculosis agents (INH, RIF, BDQ, and PBTZ169). In this study, the efficacy of Q203 was significantly improved when combined with PBTZ169 (Fig. 4). The Kaplan-Meier survival curves showed that Q203-PBTZ169 was the most effective combination at not only reducing bacterial proliferation but also expanding the life span of M. marinum-infected ZF (70% survival after 13 dpi; Fig. 4D). Therefore, we speculate that the drug synergism between Q203 and PBTZ169 observed in this study should be evaluated in higher organisms, such as rodents. The discovery of this Q203-PBTZ169 synergy is encouraging for the development of a new TB regimen for humans.

MATERIALS AND METHODS

Bacterial strains and culture conditions.

The strain used in this study (M. tuberculosis mc² 6230) was kindly provided by William R. Jacobs, Jr. (Department of Microbiology and Immunology, Albert Einstein College of Medicine, New York, NY). M. tuberculosis mc² 6230 and the laboratory-induced Q203-resistant strain (Q203-R) with a point mutation in qcrB (T313A) were grown at 37°C in 7H9^G/T/C/P/ADC (7H9 supplemented with 0.2% glycerol [G], 0.05% Tween 80 [T], 0.2% Casamino Acids [C], 24 μg/mL pantothenate [P], and Middlebrook albumin-dextrose-catalase [ADC] enrichment) agar or on solid 7H10^G/C/P/OADC (7H10 supplemented with 0.2% G, 0.2% C, 24 μg/mL P, and Middlebrook oleic acid-ADC [OADC]).

Drugs and chemical compounds.

Q203 was synthesized by Molport, SIA (Riga, Latvia). Isoniazid (INH), moxifloxacin (MXF), rifampin (RIF), linezolid (LZD), and streptomycin (STR) were purchased from Merck (Darmstadt, Germany). PBTZ169, bedaquiline (BDQ), PA-824, OPC-67683, and SQ109 were obtained from Adooq Bioscience (Irvine, CA, USA). Resazurin was purchased from Merck.

Determination of MICs and compound interactions using a REMA checkerboard assay.

The MIC for each compound was determined using a resazurin reduction microplate assay (REMA) as previously described (21). Briefly, 100 μL of 7H9^G/T/C/P/ADC medium was added to each well of a 96-well microtiter plate, except for the peripheral wells, which were filled with 200 μL of sterilized water to prevent evaporation during incubation. Antibiotics were added to the wells in 2-fold serial dilutions. After 7 days of incubation, to each well was added 40 μL of 0.025% resazurin solution (Sigma-Aldrich, USA), and the plates were reincubated overnight. Fluorescence was measured by excitation at 530 nm and emission at 590 nm using a SpectraMax M3 multimode microplate reader (Molecular Devices, Sunnyvale, CA, USA). The MIC was calculated using Prism v6 software for Windows (GraphPad Software, Inc., La Jolla, CA). To confirm whether the Q203-anchored drug combinations showed synergistic, antagonistic, or additive effects against M. tuberculosis, a checkerboard assay was performed using REMA as previously described (22). Q203 concentrations ranging from 0 to 20 nM (8 points) were prepared in 96-well plates through 2-fold serial dilution, and the 2.5-nM MIC₅₀ values of Q203 were placed in the middle of the concentration range. This serially diluted Q203 interacted with 10 different anti-M. tuberculosis drugs (INH, RIF, STR, MXF, LZD, OPC-67683, BDQ, PA-824, SQ109, and PBTZ169) at various concentrations based on 2-fold serial dilutions. The experiment was conducted in triplicate, and the results were consistent across all of the independent experiments. Fractional inhibitory concentrations (FICs) were calculated as follows: FICI = (MIC drug A in combination/MIC drug A alone) + (MIC drug B in combination/MIC drug B alone), where drug A was Q203 and drug B was INH, RIF, STR, MXF, LZD, OPC-67683, BDQ, PA-824, SQ109, or PBTZ169. An FICI of ≤0.5 was interpreted as synergy, >0.5 to 2 as indifference, and >2 as antagonism (23).

The checkerboard method was used to evaluate the antibacterial abilities of the combined antibacterial drugs. First, 1 μL of the 2-fold serial dilutions of each test compound (starting from 8× MIC₅₀) was prepared in a 96-well microplate (98 μL per well) (Corning, Baltimore, MD, USA). Bacterial stocks of M. tuberculosis mc² 6230 from the exponential-phase cultures were eluted to an optical density at 600 nm (OD₆₀₀) of 0.0025 and added to the plates to obtain a total volume of 100 μL. Each plate was then incubated for 7 days at 37°C before adding resazurin (0.025% [wt/vol] to 1/10 of the well volume). After overnight incubation, the fluorescence was measured using a SpectraMax M3 multimode microplate reader (Molecular Devices), with excitation at 560 nm and emission at 590 nm.

Evaluation of compound interactions using CFU determination.

M. tuberculosis was incubated in the presence of the drug combinations at their respective MICs. To enumerate the live bacteria in the various test conditions, bacterial broth (100 μL) was plated onto 7H10^G/C/P/OADC solid medium, and the CFU counts were determined after 2 weeks of incubation at 37°C (13).

Ethics.

All ZF experiments were approved by the Animal Research Ethics Committee of Gyeongsang National University (project identification code GNU-190325-E0014; approval date, 25 Mar 2019).

Zebrafish infection and drug combination efficacy assessment.

Mycobacterium marinum carrying pTEC27 (plasmid 30182; Addgene) and expressing the red fluorescent protein tdTomato were used for the zebrafish microinjection experiment. The infection procedure was conducted as previously described with slight modifications (22). Zebrafish (ZF) larvae 30 to 48 h postfertilization were dechorionated and anesthetized with 270 mg/L tricaine at room temperature. Around 3 nL of M. marinum-tdTomato (400 CFU) in 0.085% phenol red was injected into the caudal veins of the ZF using a digital microinjector (Tritech Research; MINJ-D). The infected ZF were transferred into 96-well plates (2 fish per well containing 200 μL fish water) and exposed to each drug. Q203 (12.5 μM) was combined with other anti-M. tuberculosis drugs (12.5 μM INH, 12.5 μM RIF, 1.5 μM BDQ, and 5 nM PBTZ169). The fish water and compounds were renewed daily. ZF larvae treated with DMSO were used as a negative control. The ZF in vivo combination efficacy was assessed as described previously (22). The in vivo anti-M. marinum effect of each Q203 drug combination was determined by the tdTomato dissemination inside each ZF body, the bacterial burden counts, and the Kaplan-Meier survival curve. The tdTomato concentration was assessed by capturing the M. marinum-tdTomato dissemination inside the infected ZF 5 days postinfection (dpi) using an ImageXpress Pico automated cell imaging system (Molecular Devices). For quantification of the bacterial burden, 20 infected ZF (5 dpi) per drug combination group were collected and homogenized in 2% Triton X-100-phosphate-buffered saline with Tween 20 (PBST) using a handheld homogenizer (D1000; Benchmark Scientific, Sayreville, NJ, USA). After a 10-fold serial dilution, the suspensions were plated onto 7H10^G/OADC agar plates containing BBL MGIT PANTA (polymyxin B, amphotericin B, nalidixic acid, trimethoprim, and azlocillin; Becton, Dickinson, Franklin Lakes, NJ, USA) and then incubated for 5 to 7 days at 30°C before counting the number of CFU. For Kaplan-Meier survival curves, the number of dead ZF was recorded daily until day 13. The bacterial burden inside the infected ZF and the survival curves were plotted using GraphPad Prism v5 using the Kaplan-Meier curve and the log-rank (Mantel-Cox) test, respectively, to compare the difference between the DMSO control and the drug combination-treated ZF.