Both Nitro Groups Are Essential for High Antitubercular Activity of 3,5-Dinitrobenzylsulfanyl Tetrazoles and 1,3,4-Oxadiazoles through the Deazaflavin-Dependent Nitroreductase Activation Pathway

3,5-Dinitrobenzylsulfanyl tetrazoles and 1,3,4-oxadiazoles, previously identified as having high in vitro activities against both replicating and nonreplicating mycobacteria and favorable cytotoxicity and genotoxicity profiles were investigated. First we demonstrated that these compounds act in a deazaflavin-dependent nitroreduction pathway and thus require a nitro group for their activity. Second, we confirmed the necessity of both nitro groups for antimycobacterial activity through extensive structure–activity relationship studies using 32 structural types of analogues, each in a five-membered series. Only the analogues with shifted nitro groups, namely, 2,5-dinitrobenzylsulfanyl oxadiazoles and tetrazoles, maintained high antimycobacterial activity but in this case mainly as a result of DprE1 inhibition. However, these analogues also showed increased toxicity to the mammalian cell line. Thus, both nitro groups in 3,5-dinitrobenzylsulfanyl-containing antimycobacterial agents remain essential for their high efficacy, and further efforts should be directed at finding ways to address the possible toxicity and solubility issues, for example, by targeted delivery.


■ INTRODUCTION
Although tuberculosis (TB) is a curable and preventable disease, it remains among the top causes of death worldwide and indeed recently became the second leading infectious killer after COVID-19.Moreover, the COVID-19 pandemic has reduced the access to TB diagnosis and treatment, which resulted in an increase in TB deaths.Only 6.4 million people newly diagnosed with TB were reported in 2021 from an estimated 10.6 million people who developed the disease, and the number of TBrelated deaths increased from 1.4 million in 2019 to 1.6 million in 2021. 1 The COVID-19 pandemic also reduced the number of people provided with treatment for drug-resistant TB by approximately 15%, and only one in three people with drugresistant TB received treatment in 2020, with a slight recovery in 2021 (7.5% increase).Globally, approximately 3−4% of newly diagnosed TB cases are classified as multidrug-resistant strains (MDR-TB), and in the case of patients previously treated for TB, the proportion of MDR-TB is higher than 18%.Current therapy for drug-resistant TB has a low success rate (about 60% in 2019) and consists of prolonged multidrug regimens, which can last up to 24 months of taking five or more different anti-TB drugs. 1 Such treatment regimens have many unpleasant side effects and drug−drug interactions (especially with antiretro-viral drugs in the case of HIV co-infection) which cause poor compliance and hamper the coadministration of antiretroviral and anti-TB drugs.This, together with the fact that the most affected regions are those with relatively poor medical care, increases the risk of the formation and spread of MDR and extensively drug-resistant (XDR) strains.Therefore, addressing the availability and effectiveness of treatment for drug-resistant TB remains a major concern, and new, highly efficient, and better-tolerated drugs are needed.Recently, two nitro groupcontaining agents, delamanid 2 and pretomanid, 3 have been approved for the treatment of MDR/XDR-TB.Both these agents have a nitro group-dependent mechanism of action as they are bioreductively activated by deazaflavin-dependent nitroreductase (Ddn) in mycobacteria. 4Other compounds with nitro group-dependent antimycobacterial activity are the benzothiazinones, 5 which are inhibitors of mycobacterial decaprenylphosphoryl-β-D-ribofuranose 2′-oxidase (DprE1). 6,7wo benzothiazinone derivatives, BTZ-043 and PBTZ-169 (macozinone), are currently undergoing evaluation in the clinic. 8ur research group has developed several structural types of new antitubercular agents with high and selective antimycobacterial activity.These compounds typically contain a fivemembered heterocycle and a 3,5-dinitrophenyl 9−11 or 3,5dinitrobenzylsulfanyl moiety. 12,13The latter group, 3,5dinitrobenzylsulfanyl tetrazoles (1) and oxadiazoles (2) (Figure 1), showed excellent activity against both drug-susceptible and drug-resistant strains.The best oxadiazole derivatives of structure 2 had minimum inhibitory concentrations (MIC) of 0.03 μM against replicating Mycobacterium tuberculosis (M.tb.) strains and were also highly effective against the nonreplicating M.tb.SS18b-Lux strain. 13Interestingly, despite the structural similarity to known DprE1 inhibitors, 6,14 3,5-dinitrobenzylsulfanyl oxadiazoles 2 did not affect the function of this enzyme, and the actual mechanism of action remained elusive. 13The in vitro antimycobacterial efficiency of tetrazole derivatives 1 was lower compared to their oxadiazole counterparts; their MIC values reached 1 μM concentration . 12Despite the presence of two nitro groups in the molecules, these lead compounds did not suffer from cytotoxicity to various cell lines, including isolated human hepatocytes, and did not exhibit genotoxicity in several assays.
The results of the above-mentioned studies indicated that the 3,5-dinitrobenzyl moiety is the fragment responsible for high in vitro antimycobacterial activity.It was found that 2,4dinitrobenzyl isomers had substantially lower antimycobacterial activity compared to 3,5-dinitro compounds, and 3-nitro-5-(trifluoromethyl)benzyl or 3-amino-5-nitrobenzyl analogues lost antimycobacterial activity altogether. 12,13,15However, the presence of two nitro groups could be the main obstacle to the further development of these potent antimycobacterial agents.Despite the long history of nitro-containing drugs and recent findings of bioreductive activation, 16 medicinal chemists typically try to avoid nitro groups in drug design due to concerns about toxicity and solubility.
Therefore, the first aim of this work was to elucidate the mode of action of oxadiazoles 2 to (a) determine whether the presence of a nitro group is essential for antimycobacterial activity and (b) rationalize the design of new analogues.Second, the structure− activity relationships were explored.In Part A (Figure 2A), one nitro group in the two lead compounds 3,5-dinitrobenzylsul-fanyl tetrazole (1) and oxadiazole (2) was replaced by other electron withdrawing groups.Thus, chloro-, fluoro-, bromo-, cyano-, methoxycarbonyl-, carbamoyl-, and pyrrol-1-yl-analogues of tetrazoles 1a−e and/or oxadiazoles 2a−e were prepared, and their in vitro antimycobacterial activities were evaluated.Trifluoromethyl analogues were also prepared to complete the series. 13The 3,5-dinitrobenzyl moiety was also replaced by a heterocyclic (5-nitropyridin-3-yl)methyl or (5nitrofuran-2-yl)methyl group.
Parts B and C of the structure−activity relationship study focused on the position of the nitro groups on the benzyl moiety, which appeared to be crucial for the antimycobacterial activity of compounds 1 and 2. In addition to the previously investigated 2,4-dinitrobenzyl analogues, in this work we shifted just one nitro group of the parent compounds 1a−e and 2a−e and prepared their 3,4-and 2,5-dinitrobenzyl analogues (Figure 2B).As the preliminary experiments showed high in vitro antimycobacterial activity of the compounds with the 2,5dinitrobenzyl moiety, their mononitro analogues with 2-nitro-5-(trifluoromethyl)benzyl and 5-nitro-2-(trifluoromethyl)benzyl groups were also synthesized (Figure 2C).
In part D, a methyl or methoxy group was introduced to the 3,5-dinitrobenzyl moiety to explore the effect of additional substitution and steric hindrance of one or both neighboring nitro groups on the antimycobacterial activity of lead compounds (Figure 2D).Structure−activity relationships with respect to the substituent R on the tetrazole or oxadiazole core have been fully elucidated in our previous studies; 11−13 thus in this work we selected five lipophilic substituents R (a−e, Figure 2) and used them in all series prepared and studied in this work to obtain easily comparable results.

■ RESULTS AND DISCUSSION
Mode of Action of 3,5-Dinitrobenzylsulfanyl Oxadiazoles 2. Previously we proved that 3,5-dinitrobenzylsulfanyl oxadiazoles and thiadiazoles do not affect the mycobacterial DprE1 and may target the synthesis of mycobacterial nucleic acids. 13To elucidate the mechanism of action of these compounds, mutants of M.tb.Erdman resistant to 3,5dinitrobenzylsulfanyl 1,3,4-oxadiazole T6030 (11i in ref 13) and 1,3,4-thiadiazole T6053 (14g in ref 13) were generated using concentrations 10 times and 20 times higher than their MIC values.Whole genome sequencing followed by bioinformatics analysis showed that all mutant colonies carried a different nonsynonymous single nucleotide polymorphism in the fgd1 gene (rv0407) encoding F 420 -dependent glucose-6phosphate dehydrogenase (FGD1) (Table 1), similarly as in M.tb.mutants resistant to nitroimidazoles pretomanid and delamanid, 17−19 FDA-approved anti-TB drugs.Mutations in FGD1 disrupt the reduction of cofactor F 420 to F 420 -H 2 , which inhibits the function of Ddn and blocks the reductive activation of nitroimidazoles. 17o further confirm that the antimycobacterial activity of compounds T6030 and T6053 rely on the Ddn-activation, we determined the MIC values in Ddn-and FbiC-deficient M.tb.mutants.We found that both mutant strains showed resistance to both T6030 and T6053 (>3-and 10-fold increase in MIC values, respectively, compared to wild-type M.tb.H37Rv), as well as to pretomanid.
These results indicated that 3,5-dinitrobenzylsulfanyl oxadiazoles 2 are activated in a similar way as nitroimidazoles pretomanid and delamanid and proved that their antimycobacterial activity is nitro group-dependent.This conclusion is in

Journal of Medicinal Chemistry
agreement with a recent study of van Calenbergh et al., who experimentally proved that the antimycobacterial activity of closely related quinazolinones bearing the key 3,5-dinitrobenzylsulfanyl group depends on the reductive activation of the 3,5-dinitrobenzyl moiety by Ddn as in the case of the nitroimidazoles (Figure 3). 20hese findings proved that at least one nitro group must be maintained in the structure of 3,5-dinitrobenzylsulfanyl heterocycles such as tetrazoles 1 and oxadiazoles 2 and drove the design of their mononitro analogues prepared in this work.the preparation of the corresponding 3-nitro-5-substituted benzoic acids 3−8, followed by the reduction of the carboxylic acid group using borane in THF (Scheme 1 and 2). 21-Nitro-5-trifluoromethylbenzoic acid (3) was obtained by nitration of 3-trifluoromethylbenzoic acid in excellent yield (Scheme 1).13 Synthesis of 3-chloro-(5) or 3-fluoro-5nitrobenzoic acid (6) started from 3,5-dinitrobenzoic acid.Its reduction by sodium sulfide hydrate in the presence of ammonium chloride provided 3-amino-5-nitrobenzoic acid 4, which, after diazotization and substitution with chlorine or
In Vitro Antimycobacterial Activity.In vitro antimycobacterial activity of all final compounds of series 52−83 were evaluated against M.tb.CNCTC My 331/88 (H37Rv) and against nontuberculous mycobacterial strains of M. avium CNCTC My 330/88 and M. kansasii CNCTC My 235/80 and compared with in vitro antimycobacterial activity of lead compounds of series 1 and 2. The antimycobacterial activities of all compounds were evaluated after 7, 14, or 21 days of incubation and are expressed as minimum inhibitory concentration (MIC) in micromolar.
The first aim of this work was to explore the possibility of the replacement of one nitro group for another electron-withdrawing and/or (bio)isosteric group in 3,5-dinitrobenzylsulfanyl tetrazole 1 and/or oxadiazole 2 antitubercular agents (Part A).Therefore, derivatives with trifluoromethyl-(52a−e, 57a−e), chloro-(53a−e, 58a−e), fluoro-(54a−e, 59a−e), bromo-(55a−e, 60a−e), and cyano-(56a−e, 61a−e) groups instead of one nitro group in the 3,5-dinitrobenzyl part were prepared.In the case of oxadiazole lead compounds 2a−e, which displayed outstanding activities, additional analogues with methoxycarbonyl-(62a−e), carbamoyl-(63a−e, 64a−e), and pyrrole (65a−e) groups were prepared.However, a strong decrease of the antimycobacterial activity was observed in all cases, regardless of the introduced functional group or heterocycle involved; indeed the majority of compounds completely lost their antitubercular activity (Tables 2 and 3).Among the prepared analogues, 3-cyano-5-nitrobenzyl derivatives of series Another possibility to reduce the number of nitro groups in the lead compounds was the replacement of the 3,5dinitrobenzyl fragment with heterocyclic (5-nitropyridin-3yl)methyl and (5-nitrofuran-2-yl)methyl moieties, especially the latter, since the 5-nitrofuran-2-yl group has previously been identified as a key moiety responsible for high antimycobacterial effect of several series of potent anti-TB agents. 28,29Thus, oxadiazole-type series 82a−e and 83a−e were prepared.Despite good antimycobacterial activity found with some of the prepared analogues, especially in the case of 5-nitrofuran-2-yl analogue 83e, lead compounds of series 2 were always in excess of 10 times more active (Table 4).
Because all the efforts to remove or replace one nitro group in the lead compounds 1 and 2 resulted in substantial decrease of antimycobacterial activity, we decided to explore more deeply the role of the position of both nitro groups in antimycobacterial activity.In our previous work, we proved that 2,4-dinitrobenzyl analogues showed lower antimycobacterial activity compared to their 3,5-dinitro counterparts. 12,13,15Therefore, 3,5-dinitrosubstituted compounds served as the lead compounds in following studies. 11,30Thus, in Part B we focused on the remaining variants with a nitro group in position 3 (or 5), i.e. 2,5-dinitro and 3,4-dinitro analogues.Positive hits could open a new path to further structural modifications and the possibility of nitro group replacement, which was not the case with 3,5dinitrobenzyl lead compounds.In the case of 3,4-dinitrobenzyl analogues of series 66 and 68, we found a decrease in antimycobacterial activity when compared to those of the lead compounds of series 1 and 2. Nonetheless, 2,5-dinitro analogues of series 67 and 69 showed very good activities comparable to that of INH, i.e., comparable to those of lead compounds 1a−e but lower than oxadiazole-based lead compounds 2a−e.Interestingly, activities of 3,4-dinitro and especially 2,5-dinitro analogues were not influenced by the type of the heterocycle.Tetrazole-based and oxadiazole-based compounds 67a−e and 69a−e, respectively, showed very similar activities.As 2,5dinitrobenzylsulfanyl maintained high antimycobacterial activities, we preliminarily checked the possibility of replacing one nitro group for another electron-withdrawing group: trifluoromethyl.Thus, in Part C, 2-nitro-5-(trifluoromethyl) derivatives 70a−e and 72a−e and 5-nitro-2-(trifluoromethyl) derivatives 71a−e and 73a−e were prepared and evaluated for their antimycobacterial efficacy.Unfortunately, significant decrease of activity or its complete loss was observed for both tetrazole and oxadiazole series, similarly to the case of trifluoromethyl analogues of lead compounds 1 and 2 (Tables 5 and 6).
found in the tetrazole series, where 4-substituted derivatives 76a−e showed the highest antimycobacterial activities within tetrazole series 74−77.Antimycobacterial activities of oxadiazoles 79a−e and 81a−e were comparable to those of tetrazoles 1a−e and INH but still significantly lower compared to the most efficient 3,5-dinitrobenzylsulfanyl oxadiazoles 2a−e (Table 8).
To further inspect the antimycobacterial activities of the most active derivatives prepared in this study, 14 compounds, tetrazoles 56c, 67a, 67b, 67c, and 67e and oxadiazoles 61b, 69a, 69b, 69c, 69e, 79a, 79e, 81a, and 81e, were selected, and their activity against seven clinically isolated MDR/XDR M.tb.strains was evaluated (Table 9).The activities of studied compounds against these resistant strains were comparable with those against the standard M.tb.strain indicating that these derivatives acted through a Ddn-activation pathway similar to the parent oxadiazoles 2. Consistently, the highest activities were found in the series of 2,5-dinitrobenzylsulfanyl derivatives 67 and 69, regardless of the substituent R on the tetrazole or oxadiazole, respectively.
Mode of Action of 2,5-Dinitrobenzylsulfanyl Tetrazoles 67a−e and Oxadiazoles 69a−e.Due to the very small difference in the structure of 2,5-dinitro-and 3,5-dinitrobenzylsulfanyl derivatives, we first checked whether their mechanism of action is consistent.However, in contrast to the parent 3,5dinitrobenzylsulfanyl derivatives T6030 and T6053, selected 2,5-dinitrobenzylsulfanyl tetrazoles 67b and 67c and oxadiazoles 69c and 69e showed the same inhibitory activity against wild-type M.tb.H37Rv as against Ddn-and FbiC-deficient mutants indicating that 2,5-dinitro compounds of series 67 and 69 acted via a Ddn-independent pathway.Thus, we turned our attention to DprE1, another important target of nitro-groupcontaining anti-TB agents including 3,5-dinitrophenyl-containing entities. 11,14First, we inspected the effects of 2,5-dinitrobenzylsulfanyl tetrazoles 67b and 67c and oxadiazoles 69c and 69e on the biosynthesis of lipids of M.tb.H37Rv via the [ 14 C]acetate radiolabeling experiments in the presence of 10 times or 100 times the MIC of selected compound.The effects of parent T6030 and T6053 were also reassessed (11i and 14g in ref 13, respectively) as the reference.As shown in Figure 4, tetrazole 67b and oxadiazole 69e caused accumulation of trehalose monomycolates (TMMs) and trehalose dimycolates (TDMs) in mycobacteria, which is a typical phenomenon for DprE1 inhibitors including BTZ-043. 11Treatment of mycobacteria with derivatives 67c and 69c led to the accumulation of TMM only.As expected, treatment with 3,5-dinitrobenzylsulfanyl derivatives T6030 and T6053 did not affect the [ 14 C]labeled lipid profiles in mycobacteria (Figure 4).To confirm that the antimycobacterial activity of 2,5-dinitrobenzylsulfanyl heterocycles of series 67 and 69 is related to DprE1 inhibition, we determined their MIC values in M.tb.H37Ra overproducing DprE1/2, with BTZ-043, one of the most efficient DprE1 inhibitors, used as a control.As shown in Table 10, the activity of 2,5-dinitrobenzylsulfanyl tetrazole 67b and oxadiazole 69e against mycobacteria overproducing DprE1/2 dropped more than 10 times, while the activity of tetrazole 67c and oxadiazole 69c was not significantly affected.As expected, the activity of BTZ-043 dropped significantly, while the original 3,5-dinitro compounds T6030 and T6053 showed similar activity regardless of the level of DprE1/2 production.
In Vitro Effects of Studied Compounds on Mammalian Cell Viability.The effects of selected final compounds on mammalian cell viability were tested using HepG2 (human hepatocellular carcinoma) cells.In the cases when the IC 50 exceeded 30 μM, the data are presented as the relative viability at a concentration of 30 μM compared to control vehicle-treated samples (100% viability).All 2,5-dinitrobenzylsulfanyl tetra-

Journal of Medicinal Chemistry
zoles (67b, 67c, 67e) and oxadiazoles (69a−c, 69e) that showed the highest antimycobacterial activities within compounds in this SAR study showed the highest toxicity/ antiproliferative activity to HepG2 cells (Table 11), which was not the case for parent 3,5-dinitrobenzylsulfanyl tetrazoles 1 12 and mainly oxadiazoles 2, which did not affect HepG2 cell viability at 50 μM concentrations after 48 h of incubation. 13CONCLUSIONS The presence of nitro groups has often discouraged further development of hit compounds as drugs, because nitro groups can increase the risk of toxicity (mainly genotoxicity/ mutagenicity), decrease the solubility of these compounds, and lead to their rapid metabolization. 16However, 3,5dinitrobenzylsulfanyl-substituted heterocycles have been identified by us and others as readily accessible compounds with excellent antimycobacterial activities and acceptable toxicity profiles. 12,13,15,20Here, we first examined the role of the nitro groups in the mode of antimycobacterial action of these compounds.Whole genome sequencing of spontaneously resistant colonies showed that they harbored mutations in the fgd1 (Rv0407) gene encoding FGD1.Mutations in FGD1 disrupt the reduction of cofactor F 420 to F 420 -H 2 , which inhibits the function of Ddn and blocks the reductive activation of nitrogroup-containing drugs like pretomanid or delamanid. 17ecreased activity of 3,5-dinitrobenzylsulfanyl derivatives T6030 and T6053 toward Ddn-and FbiC-deficient M.tb.mutants proved that 3,5-dinitrobenzylsulfanyl heterocycles have a nitro-group-dependent mode of action that relies on Ddnreductive activation.In the second part of this work, we have thoroughly investigated the structure−activity relationships of 3,5-dinitrobenzylsulfanyl tetrazoles and 1,3,4-oxadiazoles to see if we can replace/relocate one of the two nitro groups.Thus, various electron-withdrawing groups were attached instead of one nitro group.Moreover, the isosteric pyrrol-1-yl group, which has been successfully used to replace nitro group in various types of anti-TB agents, 31 was utilized.Finally, the entire 3,5dinitrophenyl group was replaced by nitro-substituted heterocyclic groups.However, the majority of the prepared compounds had significantly decreased activity as compared to their parent tetrazole and especially oxadiazole compounds.Thus, in the next step, we investigated the role of the relative position of the two nitro groups to possibly open the way for further structural optimization.We found that 2,5-dinitrobenzylsulfanyl tetrazoles 67a−e and oxadiazoles 69a−e showed consistently high antimycobacterial activity with MIC values around 1 μM against drug-susceptible and also MDR/XDR clinically isolated strains, i.e., activities comparable to those of parent tetrazoles 1a−e but lower compared to oxadiazoles 2a−e.Interestingly, shifting the nitro group from position 3 to position 2 led to a change in the dominant mechanism of antimycobacterial action.2,5-Dinitrobenzylsulfanyl tetrazoles of series 67 and oxadiazoles of series 69 acted as DprE1 inhibitors as demonstrated by the accumulation of TMMs and TDMs in treated mycobacteria and by decreased activity of these compounds in mycobacteria overproducing DprE1/2.However, all 2,5-dinitro analogues showed significant toxicity to HepG2 cells, which was not the case for the parent 3,5-dinitro compounds.The replacement of one nitro group for a trifluoromethyl group in 2,5-dinitrobenzyl derivatives also led to a significant decrease or complete loss of antimycobacterial activity.The last attempt to modify the structure of compounds 1 and 2 was the introduction of an additional methyl or methoxy substituent adjacent to the 3,5-dinitrophenyl group, which can sterically hinder one or both nitro groups.However, these modifications also led to a significant decrease in antimycobacterial activity.
In conclusion, both nitro groups in 3,5-dinitrobenzylsulfanylcontaining antimycobacterial agents remain essential for their high efficacy.Further efforts should therefore be directed at fine-

Journal of Medicinal Chemistry
tuning the activity/toxicity ratios and finding ways to address the solubility issues, for example, by targeted delivery, rather than avoiding nitro groups.
■ EXPERIMENTAL SECTION General.The prepared compounds were characterized using 1 H NMR and 13 C NMR spectroscopy.The purity of all prepared compounds was >95% as determined using elemental analysis (fluorine-free compounds) or HPLC−HRMS experiments (fluorinecontaining compounds and oily compounds).All chemicals used in the syntheses were obtained from Sigma-Aldrich (Schnelldorf, Germany) and PENTA s.r.o.(Prague, Czech Republic) and were used as received.TLC separations were performed on Merck aluminum plates with silica gel 60 F 254 .Merck Kieselgel 60 (0.040−0.063 mm) was used for column chromatography.Melting points were recorded with a Buchi B-545 apparatus (BUCHI Labortechnik AG, Flawil, Switzerland) and are uncorrected. 1H and 13 C NMR spectra were recorded using Varian Mercury Vx BB 300, VNMR S500 NMR (Varian, Palo Alto, CA, USA) or Jeol JNM-ECZ600R (JEOL Ltd., Akishima, Tokyo, Japan) spectrometers.Chemical shifts are reported as δ values in parts per million (ppm) and were indirectly referenced to tetramethylsilane (TMS) via the solvent signal.Elemental analyses were performed on an Automatic Microanalyzer EA1110CE (Fisons Instruments S.p.A., Milano, Italy).HPLC−HRMS (ESI) experiments were performed using an HRMS system Acquity UPLC I-class and a Synapt G2Si Q-TOF mass spectrometer (Waters, Milford, MA, USA).
Cell Proliferation/Viability Assay.HepG2 cells were cultivated in DMEM supplemented with 10% fetal bovine serum and sodium pyruvate (1 mM).The viability assay was carried out using the CellTiter 96 Aqueous One Solution Cell Proliferation Assay (Promega) according to the manufacturer's protocol.Briefly, the cells were seeded onto the 96-well plates at the density of 30 000 cells/well and allowed to attach for 24 h.After that, the cells were treated with the tested compounds that were predissolved in DMSO to a 1000× concentration and then dissolved in cultivation medium to 1× concentration and vehicle control (0.1% DMSO).The cells were treated for 48 h.After that, the reagent was added to the wells, and the plates were incubated in 37 °C, 5% CO 2 for 1 h.After incubation, the absorbance was measured at 490 nm using the Synergy 2 Biotek plate reader (Biotek, Winooski, VT).
Isolation and Characterization of M. tuberculosis Erdman Mutants Resistant to T6030 or T6053.3,5-Dinitrobenzylsulfanyl oxadiazole mutants of M. tuberculosis H37Rv were isolated from 7H9 cultures over 5 passages with increasing concentrations of T6030 or T6053 starting from 2×, 5×, and 10× MIC to final concentrations of 50× and 100× MIC.Single colonies were obtained from three independent cultures by streaking on 7H10 agar plates, and resistance to T6030 and T6053 was measured by REMA.Genomic DNA extraction was performed using the QiaAMP UCP pathogen minikit (Qiagen) as per the manufacturer's instructions.Whole-genome sequencing was performed using Illumina technology with sequencing libraries prepared using the KAPA HyperPrep kit (Roche) and sequenced on an Illumina HiSeq 2500 instrument.All raw reads were adapter and quality trimmed with Trimmomatic v0.33 32 and mapped onto the M. tuberculosis H37Rv reference genome (RefSeq no.NC_000962.3)using Bowtie2 v2.2.5. 33The bamleftalign program from the FreeBayes package v0.9.20−18 34 was used to left-align indels.Reads with a mapping quality below 8 and duplicate reads were omitted.
Variant Analysis.Variant calling was done using VarScan v2.3.9 35 using the following cutoffs: minimum overall coverage of 10 nonduplicated reads, minimum of 5 nonduplicated reads supporting the SNP, base quality score of >15, and an SNP frequency above 30%.The rather low thresholds, especially the SNP frequency, were deliberately chosen to avoid missing potential variants in regions where alignment was difficult or in the case of a mixed population.All putative variants unique to the mutant strains were manually checked by inspecting the alignments.
Evaluation of the Effects of the Selected Target Compounds on M.tb.H37Rv Lipids by [ 14 C]Acetate Metabolic Labeling.M.tb.H37Rv was grown shaking (120 rpm) at 37 °C in 7H9 medium supplemented with 10% ADC and 0.05% Tween 80 until OD 600 = 0.18.The culture aliquots (100 μL) were transferred to Eppendorf tubes containing the tested compounds (2 μL of the stock solutions in DMSO) to achieve the final concentrations corresponding to 10× and
susceptible to the given antibiotic drug.R, Strain resistant to the given antibiotic drug.n.d., not determined.

Table 2 .
In Vitro Antimycobacterial Activities of the Final Tetrazole-Based Compounds of Series 52−56 Expressed as MIC (μM) and Their Comparison with Those of Parent Tetrazoles 1a−e 12 a 14/21 days.b 7/14/21 days.