ABSTRACT
The efficacy of the standardized four-drug regimen (comprising isoniazid, rifampin, pyrazinamide, and ethambutol) for the treatment of tuberculosis (TB) is menaced by the emergence of multidrug-resistant (MDR) and extensively drug-resistant (XDR) strains of Mycobacterium tuberculosis. Intensive efforts have been made to develop new antibiotics or to repurpose old drugs, and several of these are currently being evaluated in clinical trials for their antitubercular activity. Among the new candidate drugs is macozinone (MCZ), the piperazine-containing benzothiazinone PBTZ169, which is currently being evaluated in phase I/II clinical trials. Here, we determined the in vitro and in vivo activity of MCZ in combination with a range of anti-TB drugs in order to design a new regimen against active TB. Two-drug combinations with MCZ were tested against M. tuberculosis using checkerboard and CFU enumeration after drug exposure assays. MCZ was observed to have no interactions with all first- and second-line anti-TB drugs. At the MIC of each drug, MCZ with either bedaquiline (BDQ), clofazimine (CLO), delamanid (DMD), or sutezolid (STZ) reduced the bacterial burden by 2 logs compared to that achieved with the drugs alone, indicating synergism. MCZ also displayed synergism with clomiphene (CLM), a potential inhibitor of the undecaprenyl pyrophosphate synthase (UppS) in mycobacteria. For all the other drugs tested in combination with MCZ, no synergistic activity was observed. Neither antagonism nor increased cytotoxicity was found for most combinations, suggesting that MCZ could be added to different TB treatment regimens without any significant adverse effects.
INTRODUCTION
With more than 9 million people infected worldwide, tuberculosis (TB) is the major cause of death due to a bacterial pathogen (1). As recommended by the WHO, the mainstay regimen for drug-susceptible TB relies on a multidrug therapy which consists of isoniazid (INH), rifampin (RIF), pyrazinamide (PZA), and ethambutol (EMB) (1). The increasing numbers of multidrug-resistant TB (MDR-TB) and extensively resistant TB (XDR-TB) cases challenge the reliability of this regimen to treat active TB. Strains causing MDR-TB are resistant to both INH and RIF, the two drugs most active against strains causing drug-susceptible TB (DS-TB). MDR-TB treatment involves second-line injectable aminoglycosides (e.g., amikacin and capreomycin) and fluoroquinolones. In the case of XDR-TB, resistance to RIF, INH, aminoglycosides, and fluoroquinolones is observed. With more than 480,000 reported cases of MDR-TB and 9.5% of reported cases of MDR-TB worldwide classified as XDR-TB, drugs active against new Mycobacterium tuberculosis pharmacological targets are indeed needed (1).
An effective regimen for the treatment of TB consists of multiple drugs that inhibit several distinct essential cellular activities in the bacterium. The effectiveness of a regimen is further enhanced if synergism is present between the constituents. Recently, studying the combinatory effect of anti-TB agents has become of interest to the TB community in order to find more efficacious treatments. Several combinations of anti-TB drugs have demonstrated potentiation of the individual drug activity in vitro when combined. Interestingly, drugs such as cephalosporins, which have limited activity against M. tuberculosis due to the presence of the BlaC β-lactamase, were shown to be synergistic with drugs of the rifamycin family, bedaquiline (BDQ) and delamanid (DMD) (2, 3). Furthermore, the β-lactamase inhibitor clavulanate as a component of triple combinations including cephalosporins (and other beta-lactams) proved to be a key synergistic partner due to its ability to rescue the activity of RIF against a RIF-resistant strain (2). A combination of amoxicillin-clavulanate with first-line anti-TB drugs has also been explored and showed potent synergism against MDR isolates (4). Repurposing of drugs such as spectinomycin was also studied to test the concept that clinically available antibiotics with limited efficacy against M. tuberculosis might be used for TB treatment when coadministered with macrolides and azoles (5). The combinatory effect of new drugs has also been assessed, whereby MmpL3 inhibitors act synergistically with RIF, BDQ, clofazimine (CLO), and β-lactams (6). Altogether, these studies demonstrate the potential of anti-TB drugs to act synergistically and the possibility to optimize treatment regimens against DS-TB, MDR-TB, and XDR-TB or even to develop a pan-TB regimen (7).
A potential component of such regimens is the clinical candidate macozinone (MCZ), previously known as PBTZ169 (https://www.newtbdrugs.org/). MCZ is a bactericidal benzothiazinone that inhibits the essential flavoprotein DprE1 by forming a covalent bond with the active-site Cys387 residue (8), thus preventing the synthesis of decaprenyl phosphoryl ribose. Despite its lack of activity against nonreplicating bacteria, MCZ is highly potent against M. tuberculosis in vitro (9), ex vivo, and in vivo, where, alone, it significantly reduces the bacterial burden in the lungs and spleens of chronically infected mice (10). A new regimen including MCZ, BDQ, and PZA was found to be more efficacious than the standard three-drug treatment in a murine model of chronic disease (10). MCZ was also shown to act synergistically with BDQ and CLO (10, 11). BDQ is an inhibitor of the F1/Fo ATP synthase, and CLO is thought to be reduced by the mycobacterial type 2 NADH:quinone oxidoreductase (NDH-2), leading to a CLO-mediated increase in NADH oxidation and the production of reactive oxygen species (ROS) (12, 13). Other studies have also reported that CLO may act on the lipids of the bacterial cell wall, as observed in macrophages infected with M. leprae (14). However, the mechanism of action of this drug is far from understood.
As part of a new regimen, further studies are needed to evaluate the potential activity of MCZ when combined with other anti-TB agents. Although several new or repurposed drugs have been evaluated for their potential to be used with MCZ in combination against active TB, there has been no comprehensive work performed to optimize MCZ-containing regimens. In this study, we embarked on a thorough characterization and evaluation of the potential activity of MCZ in combination with other anti-TB drugs using a combination of checkerboard and CFU enumeration after drug exposure assays.
RESULTS
Activity of MCZ when combined with first- and second-line anti-TB drugs against strain H37Rv.The MICs of all compounds used in this study against the H37Rv strain of M. tuberculosis are shown in Table 1 and were determined by a resazurin-based microtiter assay (REMA) (15). All drugs tested were active against the bacterium under these conditions, except for PZA. The activity of MCZ in combination with the first-line drugs RIF, INH, and EMB was then determined using a checkerboard assay. At concentrations ranging from 0.5× MIC to 0.0625× MIC, the combination of MCZ with first-line drugs resulted in minimal fractional inhibitory concentration index (FICIm) and maximal fractional inhibitory concentration index (FICIM) values ranging from 0.75 to 2.0 (Table 2). MCZ was also tested in combination with the second-line injectable agent amikacin (AMK), the fluoroquinolones levofloxacin (LVX) and moxifloxacin (MFX), and the bacteriostatic agents d-cycloserine (DCS), ethionamide (ETO), and para-aminosalicylic acid (PAS). For all combinations, FICIm and FICIM ranged from 1.125 to 3 (Table 2). Taken together, these results show that MCZ has neither synergistic nor antagonistic interactions with the first- and second-line anti-TB drugs.
MICs of the drugs against M. tuberculosis H37Rva
Combinatory effect of MCZ with first- and second-line drugs against M. tuberculosis H37Rv in vitro (checkerboard assay)
Activity of MCZ with repurposed or new anti-TB drugs.In cases of MDR- and XDR-TB disease, the use of new anti-TB drugs or repurposed drugs is required (16). Several repurposed drugs, including the antileprosy agent CLO, the macrolide clarithromycin (CLR), the carbapenem meropenem (MEM), and the oxazolidinone linezolid (LZD), are now in trials for the treatment of MDR-TB (17–20). In addition, new anti-TB compounds are available. These drugs include the diarylquinoline BDQ, the dihydronitroimidazole DMD, the oxazolidinone sutezolid (STZ), and the QcrB inhibitor lansoprazole sulfide (LPZs). CLR, DMD, LPZs, LZD, MEM, and STZ were not synergistic when individually combined with MCZ, with FICIm and FICIM ranging from 1.033 to 2.5 (Table 3). As previously reported, benzothiazinones have a synergistic effect when combined with CLO and BDQ (10, 11). In the checkerboard assay, synergism was observed at concentrations close to the MIC in vitro (0.25× MIC and above), suggesting that MCZ synergism with CLO and BDQ is dependent on the concentration of MCZ. Interestingly, both BDQ and CLO were reported to act on M. tuberculosis by disrupting the proton motive force of the mycobacterial membrane (21). To assess the activity of MCZ with other uncouplers, we tested pairwise combinations of clomiphene (CLM) and tamoxifen (TMX), using the H37Rv-lux strain for TMX, due to a high fluorescence background at concentrations higher than the MIC when TMX susceptibility was determined by REMA (data not shown) (21). Although SQ109 has been reported to act as an uncoupler, it was not included in the study as its activity was previously shown to be indifferent in the presence of benzothiazinones (10, 21, 22). CLM and TMX alone are both active against H37Rv with MICs of 24.8 μg/ml and 14.22 μg/ml, respectively (Table 1). Synergism was observed when MCZ was combined in vitro with CLM (Table 3), whereas no combinatory effect was found with TMX (Table 3).
Combinatory effect of MCZ with repurposed or new anti-TB drugs against M. tuberculosis H37Rv in vitro (checkerboard assay)
CFU enumeration after drug exposure.MCZ is a bactericidal drug that inhibits the DprE1 enzyme (10, 23). To assess the bactericidality of MCZ in two-drug combinations in vitro, CFU enumeration was performed on H37Rv in the presence of BDQ, CLO, MFX, DMD, and STZ (Fig. 1B; see also Table S1 in the supplemental material). First, the activity of the drugs was assessed at 2× MIC, 1× MIC, 0.5× MIC, and 0.25× MIC of each compound. The activity of INH and RIF was also determined at 1× MIC and 0.25× MIC as controls (Fig. 1A). RIF and INH decreased significantly the bacterial burden by 2.2 log10 CFU/ml and to below the detection limit, respectively (Fig. 1A), compared to that for the nontreated control at day 7. MCZ, BDQ, CLO, DMD, and MFX showed dose-dependent bactericidal activity (Fig. 1B). No dose dependency was observed when H37Rv was exposed to STZ (Fig. 1B). At the MIC of each drug, the combination of MCZ with BDQ, CLO, MOX, DMD, and STZ significantly decreased the number of CFU compared to that achieved with each drug alone (Fig. 1C). In combination, MCZ significantly increased the bactericidal activity, reducing the number of CFU below that seen before treatment (day 0; Fig. 1C). In the absence of MCZ, only the combination of CLO and BDQ had this effect (data not shown). At 0.25× MIC of each drug, a significant decrease in bacterial viability was observed only when MCZ was combined with CLO (Fig. 1C). Also, a decrease of 1.08 log10 CFU/ml was observed when MCZ was combined with BDQ at 0.25× MIC of each drug (Fig. 1C). This result suggests that despite the lack of synergism at 0.25× MIC, MCZ and BDQ can potentiate their bactericidal action in vitro. Interestingly, these results imply that at optimal concentrations of each drug, the combination of MCZ either with BDQ, CLO, DEL, or MFX or with STZ can increase the bactericidal activity in vitro compared to that of the drugs used alone.
Counting of the number of CFU after exposure to MCZ in two-drug combinations. (A) The viability of nontreated (NT) M. tuberculosis H37Rv and M. tuberculosis H37Rv in the presence of INH and RIF at 1× MIC (gray bars) and 0.25× MIC (white bars) as controls was assessed on day 0 (D0) and on day 7 (D7). (B) Activities of the drugs tested individually at 2× MIC (black bars), 1× MIC (gray bars), 0.5× MIC (tiled bars), and 0.25× MIC (white bars). (C) Activities of MCZ two-drug combinations at 1× MIC (gray bars) and 0.25× MIC (white bars) of each drug. The dotted line represents the limit of detection in the assay. The dashed lines represent the number of CFU of the nontreated bacteria at day 0 and after 7 days of incubation. A decrease of at least 2 log10 in the mean number of CFU/ml was considered significant (*) compared to the results for the nontreated (NT) control after 7 days of incubation (A and B). In combination, a decrease of 2 log10 CFU/ml was considered significant (*) compared to the activities of the drugs alone (C). <l.o.d., below the limit of detection.
Cytotoxicity of potent synergistic combinations containing MCZ.The cytotoxicity of the synergistic combinations against HepG2 cells was determined. All drugs alone, except CLO, had a toxic dose that inhibits 50% of the cell viability (TD50) and a toxic dose that inhibits 99% of the cell viability (TD99) higher than 20 μg/ml (Table 4). The toxicity of two-drug combinations was similar to that of the more toxic drug in the combination for all combinations except MCZ-DMD and DMD-STZ (Table 4). The TD50 and TD99 of the MCZ-DMD combination were 7.7-fold and 7.8-fold higher than those of MCZ and DMD alone, respectively. In the case of DMD-STZ, the combination was 2.1-fold more toxic than STZ alone. Interestingly, the three-drug combination MCZ-BDQ-DMD had toxicity similar to that of the MCZ-DMD combination.
Cytotoxicity of drug combinations for human HepG2 cells
In vivo activity of MCZ in combination with DMD and STZ.Despite the lack of synergism in the checkerboard assay (Table 3), both the MCZ-DMD and MCZ-STZ combinations were able to decrease bacterial viability in vitro by more than 2 log10 CFU when added at a concentration equal to the MIC of each drug (Fig. 1C). Thus, the in vivo activity of these two drugs was evaluated in a mouse TB model (Fig. 2). At the concentrations tested, all drugs alone and in combinations significantly decreased the bacterial burden by more than 2.18 log10 in the lungs and 1.71 log10 in the spleens compared to the level on day 0, thus indicating a bactericidal effect. Although no combinatory effect was observed in these organs when MCZ, DMD, and STZ were combined in two-drug combinations, the triple combination MCZ-DMD-STZ was more active in the lungs of M. tuberculosis-infected mice than DMD-STZ (the most active of the two-drug combinations) with bactericidal activity of more than 0.69 log10 CFU (P = 0.014). Moreover, MCZ-DMD-STZ was also more active than the mainstay regimen (RIF-INH-PZA [RHZ]) in this model, reducing the burden by 1.17 log10 CFU compared to that achieved with RHZ (P = 0.004). In the spleen, the most active two-drug combination was MCZ-DMD, although no significant decrease (≤0.5 log10 CFU) in the bacterial burden could be observed in the case of the triple combination MCZ-DMD-STZ. These results suggest that in spite of the lack of synergism observed between MCZ, DMD, and STZ in this model, under the conditions tested, the MCZ-DMD-STZ combination is efficacious in treating TB infection in the mouse and is comparable to the mainstay regimen of RHZ.
Efficacy of drugs alone or in combination in a mouse model of TB. The following drugs were administered by gavage at the indicated doses: rifampin (R) at 10 mg/kg, isoniazid (H) at 25 mg/kg, pyrazinamide (Z) at 150 mg/kg, MCZ at 25 mg/kg, delamanid (DMD) at 5 mg/kg, and sutezolid (STZ) at 100 mg/kg. White and black columns correspond to the bacterial burden in the lungs and spleens, respectively, at day 0, when treatment was initiated, or day 28 (with no treatment [NT] or with treatment with the indicated regimens), when treatment ended. Bars represent the mean ± SD of the number of CFU for five BALB/c mice per group. *, P ≤ 0.03; ***, P ≤ 0.0002.
DISCUSSION
In this study, we determined the combinatory effects of MCZ not only with first- and second-line anti-TB drugs but also with new and repurposed drugs with antitubercular activity. Synergism is indeed of interest for the optimization of a multidrug regimen, as the activity of the drugs would be potentiated by the presence of another compound. More importantly, antagonistic interactions between drugs should be avoided, so as not to impair the activity of the regimen. We demonstrated that MCZ has no antagonistic effects with the first-line anti-TB drugs RIF, INH, and EMB (Table 2) or the second-line drugs (Table 2). This is consistent with previously published results obtained in vitro with combinations containing BTZ043, the lead compound from the benzothiazinone chemical class (23). MCZ was previously shown to be synergistic with BDQ and CLO in vitro and in vivo (10, 11). In this study, we observed similar synergism between MCZ, BDQ, and CLO in the checkerboard assay and an enhanced bactericidal activity of MCZ in the CFU enumeration after drug exposure assay in vitro (Table 3 and Fig. 1).
Interestingly, MCZ was also synergistic with CLM in the in vitro checkerboard assay. CLM is a nonsteroidal, ovulatory stimulant that acts as a selective estrogen receptor modulator (24). Little is known about the mechanism of action of CLM in M. tuberculosis. However, a previous study in Staphylococcus aureus showed that CLM targets the undecaprenyl diphosphate synthase (UppS), an enzyme which catalyzes the condensation of isopentenyl diphosphate with allylic pyrophosphates (25) and acts in the same pathway as DprE1 in actinobacteria. In Corynebacterium glutamicum, overexpression of UppS was able to rescue the wild-type strain and induce BTZ043 resistance by providing sufficient decaprenyl phosphate for cell wall biosynthesis. The 2.15-Å crystal structure (PDB accession number 5CQJ) of Escherichia coli UppS in the presence of CLM has already been determined (25). Despite the low identity between the M. tuberculosis and E. coli UppS sequences (39.02%), 10 of the 16 amino acids reported to interact with CLM are conserved (M25, N28, H43, G46, V50, A69, L93, L100, L107, and L139, according to the E. coli numbering), suggesting that in M. tuberculosis CLM is also an inhibitor of UppS (see Fig. S2 in the supplemental material). CLM may thus be used as a booster of MCZ activity by further inhibiting the decaprenyl phosphate pathway of actinobacteria.
CLM, BDQ, and CLO have all been reported to act as protonophore uncouplers which disrupt the mycobacterial membrane potential (21). BDQ and CLO similarly demonstrated synergism with MCZ, as with CLM, in our study, although no synergism was observed with other uncouplers, such as SQ109 and TMX (Table 3) (10). SQ109 was reported to dissipate the transmembrane proton concentration gradient (ΔpH), the membrane potential (Δψ), or both components of the proton motive force simultaneously, mainly due to its inhibitory effect on MmpL3 and other MmpL-driven processes (22). Little is known about the mechanism of action of TMX against M. tuberculosis. It was previously hypothesized that TMX may cause a loss of transmembrane potential and ultimately cell death by a significant efflux of K+ and Na+ ions from bacterial cells (26), and this may also occur in M. tuberculosis. Interestingly, BDQ and CLO, which synergize with MCZ, act on components of the electron transport chain (ETC) in the bacterial membrane. BDQ is an inhibitor of the F1/Fo ATP synthase, and CLO is thought to be reduced by the mycobacterial type 2 NADH:quinone oxidoreductase (NDH-2) enzyme, leading to a CLO-mediated increase in NADH oxidation and reactive oxygen species (ROS) production (12, 13). Other studies have also reported that CLO may act on the lipids of the bacterial cell wall, as observed in macrophages infected with M. leprae (14). On the other hand, there were compounds targeting the ETC, such as LPZs, which were not synergistic when combined with MCZ in vitro (Table 3). LPZs targets the QcrB component of the cytochrome bc1 oxidase, the loss of which is thought to be compensated for by the other terminal oxidase in M. tuberculosis, cytochrome bd (27). Further studies will be needed to fully understand the mechanism underlying the two-drug synergism of MCZ with BDQ, CLO, and CLM.
As observed from the CFU enumeration after drug exposure assay at certain optimal concentrations, the bactericidal activity of MCZ was enhanced by BDQ, CLO, DMD, MFX, and STZ (Fig. 1). In addition, BDQ and CLO potentiated their activity. Interestingly, all drugs tested were reported to have bactericidal activity against M. tuberculosis in vitro or in vivo (28–32). Background regimens with MCZ and BDQ, CLO, DMD, MFX, or STZ would thus be promising alternatives to decrease the bacterial burden in vivo. In this study, MCZ in combination with DMD and STZ was assessed in a murine infection model. Despite the lack of synergism between the drugs in vivo, this regimen was able to decrease the bacterial burden in the lungs and spleens of infected mice over 1 month of treatment to a level lower than that achieved with RHZ (Fig. 2). DMD and STZ are suitable candidates for the treatment of TB, as they have low toxicity in preclinical species and are well tolerated in humans (28, 33). However, further studies on their potential cytotoxicity when combined with MCZ should be performed before use in humans, as increased cytotoxicity for HepG2 cells was observed with this combination (Table 4).
In conclusion, MCZ enhances the activity of several potential partner drugs in vitro and in vivo. According to the results obtained in this study, BDQ, CLO, DMD, and STZ are indeed good candidates to elaborate a new MCZ-containing regimen active against all forms of TB. Moreover, boosting MCZ activity by the addition of CLM is also of interest, but further studies will be needed to determine the efficacy of this specific combination in vivo.
MATERIALS AND METHODS
Strain and culture methods.M. tuberculosis H37Rv (Institut Pasteur) was used for the in vitro assessment of drug activity against replicating H37Rv bacteria (9). Bacteria were cultured in 7H9 complete medium (Middlebrook 7H9 [Difco] broth supplemented with 10% albumin-dextrose-catalase [ADC] enrichment, 0.2% glycerol, and 0.05% Tween 80) at 37°C when needed.
Drug preparation.Amikacin (AMK; Sigma), bedaquiline (BDQ; Janssen Pharmaceutica NV), clofazimine (CLO; Sigma), clomiphene (CLM; Sigma), clarithromycin (CLR; Sigma), d-cycloserine (DCS; Sigma), delamanid (DMD; Otsuka Co.), ethambutol (EMB; Sigma), ethionamide (ETO; Sigma), isoniazid (INH; Sigma), levofloxacin (LVX; Sigma), linezolid (LZD; AstraZeneca), lansoprazole sulfide (LPZs; Toronto Research Chemicals), meropenem (MEM; Sigma), moxifloxacin (MFX; Sigma), para-aminosalicylic acid (PAS; Sigma), MCZ HCl salt (called MCZ throughout; Innovative Medicines for Tuberculosis), pyrazinamide (PZA; Fluka), rifampin (RIF; Sigma), sutezolid (STZ; Sequella), and tamoxifen (TMX; Sigma) were dissolved in dimethyl sulfoxide (DMSO) to obtain stock concentrations of 10 mg/ml. Aliquots were stored at −20°C. When needed, the drugs were diluted in DMSO to the desired concentration. Stocks were thawed a maximum of 5 times to avoid a potential loss of drug potency.
Antimycobacterial assays.The activities of drugs against the M. tuberculosis H37Rv strain were tested using the resazurin-based microtiter assay (REMA) in 96-well plates as previously described (15). Briefly, a mid-logarithmic-phase culture of H37Rv (optical density at 600 nm [OD600], approximately 0.5) was diluted in 7H9 complete medium to an OD600 of 0.0001 (approximately 3 × 104 CFU/ml). Bacteria (100 μl) were then dispensed in transparent flat-bottom 96-well plates. Twofold serial dilutions of each drug (resuspended in DMSO) were then prepared. On each plate, controls without drug and medium alone were included. The plates were incubated for 6 days at 37°C before addition of resazurin (0.025% [wt/vol] to 1/10 of the well volume). After overnight incubation, the fluorescence of the resazurin metabolite, resorufin, was determined with excitation at 560 nm and emission at 590 nm, measured using a Tecan Infinite M200 microplate reader. The MIC was calculated using GraphPad Prism software (version 7) and the Gompertz equation for MIC determination. All drugs were tested at least in duplicate.
Checkerboard assay.To assess if the compounds in two-drug combinations acted synergistically, antagonistically, or indifferently, a checkerboard assay was employed using REMA. The assay was performed in 96-well plates, as sketched in Fig. S1 in the supplemental material. Fractional inhibitory concentrations (FICs) were calculated using the following formula: FIC (compound X + compound Y) = (MIC of compound X in combination with compound Y)/(MIC of compound X).
The fractional inhibitory concentration index (FICI) was calculated as the FICI of compound X plus the FICI of compound Y to evaluate interaction profiles. FICI values of ≤0.5 are indicative of synergistic activity; FICI values of ≥4.0 are indicative of antagonism. The MIC in this assay is defined by the lowest concentration that inhibits at least 90% of the bacterial growth (MIC90). Values in between correspond to additivity (indifferent). FICIm and FICIM were also determined and reported as the lowest FICI (FICIm) and the highest FICI (FICIM) observed out of all assays performed. Also, the mean FICI was calculated for all drug combinations used at concentrations between 0.5× MIC and 0.0625× MIC. In the table, the mean FICI represents the mean of all FICI obtained between 0.5× MIC and 0.0625× MIC of MCZ HCl. The experiments were performed in biological duplicate.
When mentioned, the interaction between MCZ and the drugs was also assessed using the luminescent H37Rv(pEG200) strain, as previously described (34). A mid-logarithmic-phase culture of H37Rv(pEG200) (OD600, approximately 0.5) was diluted in 7H9 complete medium to an OD600 of 0.001 (approximately 3 × 105 CFU/ml). Luminescence was measured by using a Tecan Infinite M200 microplate reader after 6 days of incubation at 37°C. The MIC was calculated using GraphPad Prism software (version 7) and the Gompertz equation for MIC determination. All combinations were tested at least in duplicate.
CFU enumeration after drug exposure in vitro.To assess the activity of anti-TB drugs on bacterial viability, H37Rv was diluted to an OD600 of 0.0001 and exposed to the drugs using the same protocol used for REMA. After 7 days of incubation, 50 μl of bacteria was 10-fold serially diluted in phosphate-buffered saline containing 0.05% Tween 80 and plated on 7H10 agar plates containing 10% oleic acid-albumin-dextrose-catalase (OADC; Gibco). The numbers of CFU were counted after 4 weeks of incubation at 37°C. On day 0 (the day that the bacteria were exposed to the drugs) and day 7 (after 7 days of incubation with the drugs), the bacteria were also plated as a control to determine bacterial growth. For each experiment, a REMA was performed in addition to determine the MICs of the drugs tested. Significant results for the drugs tested alone were reported as a decrease of at least 2 log10 CFU/ml compared to the mean number of CFU per milliliter for the nontreated (NT) control at day 7. For combinations, significant effects were reported as a decrease of at least 2 log10 CFU/ml compared to the mean number of CFU per milliliter obtained with the drugs tested alone.
Cytotoxicity assay.The drugs were individually tested in duplicate by REMA for their cytotoxicity against human hepatic HepG2 cells (ATCC HB-8065) (10). Drugs were prepared in DMSO to a final concentration of 100 μg/ml and were then added to 96-well plates containing 4.0 × 104 cells/well in a final volume of 100 μl. The drugs were then serially diluted 2-fold. The plates were incubated for 3 days at 37°C, and after addition of resazurin, fluorescence was determined using a Tecan M200 microplate reader. The cytotoxic effect was expressed as the TD50 and TD99, which correspond to the toxic doses inhibiting the growth of 50% and 99% of the HepG2 cells, respectively.
To assess the potential cytotoxic effect of two-drug regimens combining MCZ with another drug in HepG2 cells, a mixture of MCZ (final concentration in the plate, 50 μg/ml) and each of the other drugs (final concentration, 50 μg/ml) was added to the plates containing HepG2 cells. The mixture was then 2-fold serially diluted and incubated for 3 days at 37°C before resazurin was added to the plates.
In vivo assessment of MCZ in combination with DMD and STZ.Female BALB/c mice (5 to 6 weeks old) were obtained from Charles River Laboratories. The mice were subjected to a low-dose aerosol infection of M. tuberculosis H37Rv. Treatment began 4 weeks after infection, and compounds were administered to infected mice by gavage 5 days a week for 4 weeks so as to determine the in vivo efficacy of single drugs and combinations.
The activities of DMD and STZ in combination with MCZ were first assessed at the following doses: STZ, 100 mg/kg of body weight; MCZ, 25 mg/kg; DMD, 5 mg/kg; PZA, 150 mg/kg; RIF, 10 mg/kg; and INH, 25 mg/kg. In this experiment, 5% Arabic gum in distilled water was used as the vehicle. At the end of the experiment, all mice were sacrificed and the bacterial load in the lungs and spleen was determined by plating dilutions of organ homogenates on 7H10 agar plates containing 10% OADC, cycloheximide (10 μg/ml), and ampicillin (50 μg/ml). The plates were incubated for 4 weeks at 37°C before the CFU were enumerated. CFU counts were log10 transformed before analysis as the mean number of log10 CFU ± standard deviation (SD) and compared using Student's t tests in Prism software (version 7; GraphPad). Animal experiments were approved by the Swiss Cantonal Veterinary Authority (authorization no. 3082).
ACKNOWLEDGMENTS
We thank Janssen Pharmaceutical, Otsuka Novel Products GmbH, and Sequella, Inc., for kindly providing the BDQ, DMD, and STZ used in this study, respectively.
This work was funded by the European Community's Seventh Framework Programme under grant agreement 260872.
FOOTNOTES
- Received 25 April 2018.
- Returned for modification 14 May 2018.
- Accepted 10 August 2018.
- Accepted manuscript posted online 20 August 2018.
Supplemental material for this article may be found at https://doi.org/10.1128/AAC.00840-18.
- Copyright © 2018 American Society for Microbiology.
