Novel Derivatives of Quinoxaline-2-carboxylic Acid 1,4-Dioxides as Antimycobacterial Agents: Mechanistic Studies and Therapeutic Potential

The World Health Organization (WHO) reports that tuberculosis (TB) is one of the top 10 leading causes of global mortality. The increasing incidence of multidrug-resistant TB highlights the urgent need for an intensified quest to discover innovative anti-TB medications In this study, we investigated four new derivatives from the quinoxaline-2-carboxylic acid 1,4-dioxide class. New 3-methylquinoxaline 1,4-dioxides with a variation in substituents at positions 2 and 6(7) were synthesized via nucleophilic aromatic substitution with amines and assessed against a Mycobacteria spp. Compound 4 showed high antimycobacterial activity (1.25 μg/mL against M. tuberculosis) and low toxicity in vivo in mice. Selection and whole-genomic sequencing of spontaneous drug-resistant M. smegmatis mutants revealed a high number of single-nucleotide polymorphisms, confirming the predicted mode of action of the quinoxaline-2-carboxylic acid 1,4-dioxide 4 as a DNA-damaging agent. Subsequent reverse genetics methods confirmed that mutations in the genes MSMEG_4646, MSMEG_5122, and MSMEG_1380 mediate resistance to these compounds. Overall, the derivatives of quinoxaline-2-carboxylic acid 1,4-dioxide present a promising scaffold for the development of innovative antimycobacterial drugs.


Introduction
Tuberculosis (TB), an infectious disease caused by the pathogen Mycobacterium tuberculosis, remains a significant global health challenge.When patients adhere to a correctly selected treatment regimen, the cure rates hover around 85% [1].However, factors such as inappropriate chemotherapy, patient non-compliance, HIV infection, and diabetes contribute to the development of TB drug resistance [2].Clinical cases of M. tuberculosis resistant to the latest approved antituberculosis drugs, bedaquiline and pretomanid, have already been reported [3].These drug-resistant TB strains are challenging to treat, necessitating lengthier treatment regimens that often come with more severe side effects.This underscores the urgent need to search for and develop new antituberculosis drugs that employ novel mechanisms of action [4].
It is well known that the presence of two N-oxide fragments in the structure of quinoxaline 1,4-dioxides (QdNOs) endows them a wide spectrum of biological properties, including antitumor, antibacterial, antiparasitic, anti-inflammatory, antioxidant, and herbicidal activities [5].Moreover, QdNOs exhibit high inhibitory activity against various pathogenic microorganisms such as Escherichia coli, Pasteurella multocida, dysentery spirochete [6], and Brachyspira hyodysenteriae [7].Several studies have confirmed the mechanism of antibacterial action of QdNOs associated with the production of free radicals during bio-reduction that lead to DNA damage [8].Furthermore, some derivatives of QdNOs have also exhibited excellent inhibitory activity against M. tuberculosis, indicating the great interest in this scaffold for the development of new anti-TB drugs [9].
Recent in silico studies of quinoxaline 1,4-dioxides binding to mycobacterial DNA gyrase have indicated that these compounds can bind to the same site as the well-established DNA-gyrase ligand, novobiocin [10].Some derivatives of this class are currently undergoing preclinical trials, including studies on MDR M. tuberculosis strains [11].
In prior research, we described a series of novel 2-acyl-3-trifluoromethylquinoxaline 1,4-dioxides with significant inhibitory properties against M. smegmatis.However, their activity against M. tuberculosis was limited [12].In this study, we introduce a novel series of quinoxaline-2-carboxylic acid 1,4-dioxide derivatives with substituent variations at positions 3, 6, and 7.These compounds underwent testing against M. smegmatis and the autoluminescent M. tuberculosis H37Ra (AlRa) strains.Compound 4, or 7-chloro-2-(ethoxycarbonyl)-3-methyl-6-(piperazin-1-yl)quinoxaline 1,4-dioxide, exhibited the most potent activity against both M. tuberculosis and M. smegmatis strains, leading to its selection as the lead compound for further studies on its mechanism of action and resistance.A comparative genomic analysis of spontaneous M. smegmatis mutants resistant to compound 4 pinpointed mutations in genes that encode the redox enzymes of pyruvate metabolism, which might function as pro-drug activators.Using reverse genetics, we verified that the genes MSMEG_5122 and MSMEG_4646 confer resistance to quinoxaline in M. smegmatis.Additionally, resistance to compound 4 can also be attributed to the MmpS5-MmpL5 efflux.In vivo tests of the compounds at specified doses in a mouse infection model yielded results comparable to the negative control.However, no toxic effects were detected.Consequently, quinoxaline 1,4-dioxide 4 has emerged as a promising foundation for refining anti-TB drug candidates, emphasizing the enhancement of its in vivo activity.

Chemistry
The synthesis of the designed QdNOs 4-7, 9 are presented in Schemes 1 and 2. The synthetic route included the Beirut reaction between 5,6-dichlorobenzofuroxan (1) with 1,3-dicarbonyl compounds in the presence of K 2 CO 3 as the catalyst and further treatment of the formed 6,7-dichloro-3-methylquinoxaline 1,4-dioxides 2, 3 with cyclic amines in DMF at 50 • C that resulted in target 6-aminosubstituted derivatives 4-6 (Scheme 1).The revealed regioselectivity of the nucleophilic substitution of the chlorine atom at position 6 of quinoxalines 2 and 3 (Scheme 1) can be explained by the conjugation of the C6 carbon atom with the electron-withdrawing group at position 2 of the heterocycle of these derivatives.This conjugation leads to an increase in the positive charge at C6 carbon atom as well as stabilization of the intermediate Meisenheimer complex, thereby enhancing its reactivity in nucleophilic substitution reactions [13].The structure of all synthesized compounds 4-7 (Scheme 1) was characterized by using HRMS ESI, HPLC, and NMR spectroscopies.
Pharmaceuticals 2023, 16, x FOR PEER REVIEW 3 of 20 1,3-dicarbonyl compounds in the presence of K2CO3 as the catalyst and further treatment of the formed 6,7-dichloro-3-methylquinoxaline 1,4-dioxides 2, 3 with cyclic amines in DMF at 50 °C that resulted in target 6-aminosubstituted derivatives 4-6 (Scheme 1).The revealed regioselectivity of the nucleophilic substitution of the chlorine atom at position 6 of quinoxalines 2 and 3 (Scheme 1) can be explained by the conjugation of the C6 carbon atom with the electron-withdrawing group at position 2 of the heterocycle of these derivatives.This conjugation leads to an increase in the positive charge at C6 carbon atom as well as stabilization of the intermediate Meisenheimer complex, thereby enhancing its reactivity in nucleophilic substitution reactions [13].The structure of all synthesized compounds 4-7 (Scheme 1) was characterized by using HRMS ESI, HPLC, and NMR spectroscopies.It was found that the chlorine atom at the 7-position of quinoxaline 2 also undergoes substitution with an amine under these conditions.So, the heating of dichloroderivative 2 with an excess of piperazine in DMF leads to the formation of the byproduct-7-aminosubstituted 2-carboethoxy-3-methylquinoxaline 1,4-dioxide 7 at a yield of 21% (Figure 1).At the final stage, free amines 4, 5, and 7 were transformed into target hydrochlorides by treatment of a solution of hydrochloric acid in methanol.The purity (>95%) of the target compounds was determined by HPLC.The structures of isomeric quinoxalines 4 and 7 were unambiguously confirmed by an analysis of the increment values of the signals in the 13 C NMR spectra, as described earlier for quinoxaline-2-carbonitriles [13,14].Using two-dimensional NMR spectroscopy (HSQC and HMBC), the 1 H and 13 C NMR signals of the regioisomeric derivatives 4 and 7 were assigned (Supplementary Materials, Figures S29-S38).Comparison of the 13 C NMR spectra of isomers 4 and 7 (Supplementary Materials, Figures S6 and S12, respectively) with that of the unsubstituted 2-ethoxycarbonyl-3methylquinoxaline 1,4-dioxide (Supplementary Material, Figure S16) revealed negative shifts in the signals of the carbon atoms at positions 9 and 10 for the 6-aminoderivative 4 and 7-aminoderivative 7, respectively.These shifts are consistent with previously obtained results.These shifts also correlated with the increments of substituents in these positions, characteristic of benzene derivatives and polyaromatic compounds (Supplementary Materials, Table S1) [13].The positions of the substituents have been definitely confirmed via 2D NMR using HMBC and NOESY experiments.A key aspect of the structure elucidation was the selective NOESY experiment using the signal of the CH3 group at position 3 of both isomers for selective excitation.In the case of the 6-isomer (compound 4), an enhancement of the high-field signal (at 7.89 ppm) corresponding to the CH-group in the ortho-position to the piperazine substituent, was observed (Supplementary Materials, Figures S32 and S33).Conversely, the same selective NOESY experiment revealed a lower field signal enhancement (at 8.41 ppm) for the 7-isomer (compound 7) (Supplementary Materials, Figures S37  and S38).These observations provided evidence that the piperazine substituent is located at position 6 of the quinoxaline ring in derivative 4 and at position 7 in its isomer 7, respectively.Additionally, crucial correlations in HMBC spectra of regioisomers were observed: a four-bond coupling between signals H8 and C2 for the 7-isomer and between H5 and C3 for the 6-isomer.This unequivocally verified the proposed structures of the regioisomers (Supplementary Materials, Figures S30 and S31 and S35 and S36, respectively).
It is worth noting that compounds 4-7 exhibited remarkable stability in the solid state and were quite stable in aqueous solutions.A store of aqueous solutions of derivatives 4, 6, and 7 at 37 °C for 7 days revealed a slow degradation of these compounds that did not exceed 10% (Supplementary Materials, Table S2).Surprisingly, a significantly higher decay rate was observed in these conditions for carboxamide 5.

Biological Screening of Quinoxaline 1,4-Dioxides Derivates on M. smegmatis and M. tuberculosis
Antibacterial activity of the novel compounds was tested against Gram-positive bacteria strains, specifically M. smegmatis mc 2 155 and autoluminescent M. tuberculosis H37Ra.The minimum inhibitory concentration (MIC) values of compounds 4-9 and two reference drugs are summarized in Table 1.It was found that the chlorine atom at the 7-position of quinoxaline 2 also undergoes substitution with an amine under these conditions.So, the heating of dichloroderivative 2 with an excess of piperazine in DMF leads to the formation of the byproduct-7-aminosubstituted 2-carboethoxy-3-methylquinoxaline 1,4-dioxide 7 at a yield of 21% (Figure 1).harmaceuticals 2023, 16, x FOR PEER REVIEW 1,3-dicarbonyl compounds in the presence of K2CO3 as the catalyst and fur of the formed 6,7-dichloro-3-methylquinoxaline 1,4-dioxides 2, 3 with cy DMF at 50 °C that resulted in target 6-aminosubstituted derivatives 4-6 (S revealed regioselectivity of the nucleophilic substitution of the chlorine ato 6 of quinoxalines 2 and 3 (Scheme 1) can be explained by the conjugation of atom with the electron-withdrawing group at position 2 of the heterocycle atives.This conjugation leads to an increase in the positive charge at C6 ca well as stabilization of the intermediate Meisenheimer complex, thereby en activity in nucleophilic substitution reactions [13].The structure of all syn pounds 4-7 (Scheme 1) was characterized by using HRMS ESI, HPLC, and N copies.It was found that the chlorine atom at the 7-position of quinoxaline 2 a substitution with an amine under these conditions.So, the heating of dich 2 with an excess of piperazine in DMF leads to the formation of the byprod substituted 2-carboethoxy-3-methylquinoxaline 1,4-dioxide 7 at a yield of 2 At the final stage, free amines 4, 5, and 7 were transformed into target h by treatment of a solution of hydrochloric acid in methanol.The purity (>95% compounds was determined by HPLC.The structures of isomeric quinox were unambiguously confirmed by an analysis of the increment values of the 13 C NMR spectra, as described earlier for quinoxaline-2-carbonitriles two-dimensional NMR spectroscopy (HSQC and HMBC), the 1 H and 13 C N At the final stage, free amines 4, 5, and 7 were transformed into target hydrochlorides by treatment of a solution of hydrochloric acid in methanol.The purity (>95%) of the target compounds was determined by HPLC.The structures of isomeric quinoxalines 4 and 7 were unambiguously confirmed by an analysis of the increment values of the signals in the 13 C NMR spectra, as described earlier for quinoxaline-2-carbonitriles [13,14].Using two-dimensional NMR spectroscopy (HSQC and HMBC), the 1 H and 13 C NMR signals of the regioisomeric derivatives 4 and 7 were assigned (Supplementary Materials, Figures S29-S38).Comparison of the 13 C NMR spectra of isomers 4 and 7 (Supplementary Materials, Figures S6 and S12, respectively) with that of the unsubstituted 2-ethoxycarbonyl-3-methylquinoxaline 1,4dioxide (Supplementary Material, Figure S16) revealed negative shifts in the signals of the carbon atoms at positions 9 and 10 for the 6-aminoderivative 4 and 7-aminoderivative 7, respectively.These shifts are consistent with previously obtained results.These shifts also correlated with the increments of substituents in these positions, characteristic of benzene derivatives and polyaromatic compounds (Supplementary Materials, Table S1) [13].
The positions of the substituents have been definitely confirmed via 2D NMR using HMBC and NOESY experiments.A key aspect of the structure elucidation was the selective NOESY experiment using the signal of the CH 3 group at position 3 of both isomers for selective excitation.In the case of the 6-isomer (compound 4), an enhancement of the high-field signal (at 7.89 ppm) corresponding to the CH-group in the ortho-position to the piperazine substituent, was observed (Supplementary Materials, Figures S32 and S33).Conversely, the same selective NOESY experiment revealed a lower field signal enhancement (at 8.41 ppm) for the 7-isomer (compound 7) (Supplementary Materials, Figures S37 and S38).These observations provided evidence that the piperazine substituent is located at position 6 of the quinoxaline ring in derivative 4 and at position 7 in its isomer 7, respectively.Additionally, crucial correlations in HMBC spectra of regioisomers were observed: a four-bond coupling between signals H8 and C2 for the 7-isomer and between H5 and C3 for the 6-isomer.This unequivocally verified the proposed structures of the regioisomers (Supplementary Materials, Figures S30, S31, S35 and S36, respectively).
It is worth noting that compounds 4-7 exhibited remarkable stability in the solid state and were quite stable in aqueous solutions.A store of aqueous solutions of derivatives 4, 6, and 7 at 37 • C for 7 days revealed a slow degradation of these compounds that did not exceed 10% (Supplementary Materials, Table S2).Surprisingly, a significantly higher decay rate was observed in these conditions for carboxamide 5.

Biology 2.2.1. Biological Screening of Quinoxaline 1,4-Dioxides Derivates on M. smegmatis and M. tuberculosis
Antibacterial activity of the novel compounds was tested against Gram-positive bacteria strains, specifically M. smegmatis mc 2 155 and autoluminescent M. tuberculosis H37Ra.The minimum inhibitory concentration (MIC) values of compounds 4-9 and two reference drugs are summarized in Table 1.A comparative assessment of the screening outcomes highlighted the pivotal role of substituents in determining the antimicrobial properties of the quinoxaline 1,4-dioxides 4-7.Specifically, compound 4, which contains an ethoxycarbonyl group at position 2 of quinoxaline, displayed the most pronounced activity across all tested strains.Substituting the ethoxycarbonyl group at position 2 with a carboxamide group of analogous electronic properties negatively influenced the antitubercular effectiveness of the compounds (derivatives 5 and 6) in both test cultures.Intriguingly, the inclusion of a lipophilic piperidine segment at position 6 (in compound 6) marginally enhanced the antimycobacterial activity against the M. tuberculosis AlRa strain compared to its counterpart, compound 5.The location of the amino group within the benzene ring of the quinoxaline core significantly influenced the antimicrobial activity.For instance, moving the piperazine fragment from position 6 to 7 led to a drop in the antitubercular activity of derivative 7 when compared to its regioisomer 4. Furthermore, the lead compound 4 exhibited activity levels against both mycobacterial strains that were comparable to the reference drug rifampicin (MIC values of 0.03 and 4 µg/mL, respectively).Additionally, derivatives 4-7 showed activity levels akin to the reference drug dioxidine (MIC values ranging between 10-16 µg/mL).As a result, compound 4 was selected for the subsequent identification of spontaneous resistant mutants.

M. smegmatis Drug-Resistant Mutants and Their Genomic Analysis
The fast-growing and non-pathogenic bacterial strain M. smegmatis mc 2 155 serves as a more suitable model for obtaining spontaneous resistant mutants, studying bacterial resistance, and understanding the drug's mechanism of action.We managed to procure spontaneous mutants of M. smegmatis mc 2 155 resistant to quinoxaline 1,4-dioxide 4 at a frequency of 2.3 × 10 −8 .While we made several attempts, we were unsuccessful in obtaining genuinely resistant M. tuberculosis mutants.Although individual colonies appeared on plates with higher concentrations of compound 4, their MIC values mirrored that of the parent strain.
For in-depth analysis, we randomly selected five mutants to study their cross-resistance and to perform whole-genome sequencing.Comparative genomic analysis highlighted that the mutants qdr1, qdr2, and qdr5 had between 31 and 35 unique, non-synonymous mutations, while qdr3 and qdr4 had 27 and 18, respectively (see Appendix A, Supplementary Materials).

Investigating the Effect of Individual Gene Mutations on Sensitivity to Quinoxalin-1,4-dioxide
To understand the phenotype of the spontaneous resistant mutants, we assessed the MICs of synthesized compounds on the resistant mutants M. smegmatis qdr1-qdr5.All displayed an increased MIC for quinoxaline 1,4-dioxide 4, ranging from 2-4 times.The qdr1, qdr2, and qdr3 strains exhibited resistance to its analogues 5-7 and DIOX.Rifampicin was used as a control.
Utilizing the method of homologous recombination via the suicide vector system p2Nil/pGOAL19 [20], we constructed M. smegmatis strains with the identified mutations.This led to three recombinant strains: 1.
The recombinant strains M. smegmatis 4646c and M. smegmatis 5122c showed more than double the MIC values of compound 4 and DIOX.Both also displayed resistance to compound 7, while M. smegmatis 5122c had heightened MICs for compounds 5 and 6.In contrast, the MSMEG_4648 mutation did not affect sensitivity to compound 4 but did enhance resistance to DIOX.The MIC values for these compounds and references are detailed in Table 3.
Given the identified mutation in MSMEG_1380-which encodes the transcriptional regulator of the mmpS5-mmpL5 operon-we assessed the susceptibility of our synthesized compounds to MmpS5-MmpL5 efflux.Utilizing our previously developed test system based on three M. smegmatis strains with varying MmpS5-MmpL5 expression [21], we observed M. smegmatis mc 2 155 and M. smegmatis atr9c showing identical growth inhibition halos.However, M. smegmatis ∆mmp5 was more susceptible to compounds 4-6.It appears that the MmpS5-MmpL5 efflux offers basal resistance to compound 4 and some derivatives, but its overexpression (as seen in M. smegmatis atr9c) does not heighten this resistance.Compound 7 seems unaffected by this efflux, likely due to substituent positions influencing efflux affinity.The growth inhibition halo diameters are depicted in Figure 2.

Mutations in Genes MSMEG_5122 and MSMEG_4648 Change Their Expression
The mutation in MSMEG_4648 led to a noticeable increase in the expression of g MSMEG_4648 and MSMEG_4645 (by 38% and 52%, respectively) (p < 0.01, Figure 3).H ever, the expression of MSMEG_4646 remained unchanged.This change in expres might not be sufficient to confer full resistance, as seen in the drug sensitivity profi the M. smegmatis 4648c strain.Conversely, in the M. smegmatis 5122c strain, a prom The capability of our synthesized amino derivatives of quinoxaline-2-carboxylic acid 1,4-dioxide to bypass the transmembrane transporters was studied using M. smegmatis ∆mmp5 and M. smegmatis atr9c.These compounds exhibited promising activity against the multidrug-resistant M. smegmatis atr9c strain, known for overexpressing mmpS5-mmpL5 genes.Furthermore, compounds 4-7 also displayed comparable activity against both the wild-type M. smegmatis mc 2 155 and its resistant counterpart M. smegmatis ∆mmp5 (see Figure 2).Intriguingly, the amino group's position in the quinoxaline core plays a crucial role in drug penetration, with regioisomer 7 of compound 4 showcasing similar activity levels in both M. smegmatis ∆mmp5 and the original M. smegmatis mc 2 155 test cultures.

Mutations in Genes MSMEG_5122 and MSMEG_4648 Change Their Expression
The mutation in MSMEG_4648 led to a noticeable increase in the expression of genes MSMEG_4648 and MSMEG_4645 (by 38% and 52%, respectively) (p < 0.01, Figure 3).However, the expression of MSMEG_4646 remained unchanged.This change in expression might not be sufficient to confer full resistance, as seen in the drug sensitivity profile of the M. smegmatis 4648c strain.Conversely, in the M. smegmatis 5122c strain, a promoter mutation caused the expression of gene MSMEG_5122 to drop significantly (by 2.76 times) (p < 0.01, Figure 2), resulting in resistance to quinoxaline 1,4-dioxides.

Mutations in Genes MSMEG_5122 and MSMEG_4648 Change Their Expressio
The mutation in MSMEG_4648 led to a noticeable increase in the expression of g MSMEG_4648 and MSMEG_4645 (by 38% and 52%, respectively) (p < 0.01, Figure 3).H ever, the expression of MSMEG_4646 remained unchanged.This change in expre might not be sufficient to confer full resistance, as seen in the drug sensitivity profi the M. smegmatis 4648c strain.Conversely, in the M. smegmatis 5122c strain, a prom mutation caused the expression of gene MSMEG_5122 to drop significantly (by 2.76 ti (p < 0.01, Figure 2), resulting in resistance to quinoxaline 1,4-dioxides.

Derivative 4 Displays Minimal Cytotoxicity on Human Fibroblasts
In primary cultures of human fibroblasts, the cytotoxicity of derivative 4 was assessed using MTT analysis and IC 50 determinations.The IC 50 value for 4 was 67 µM.For comparison, DIOX and mitoxantrone were used, yielding IC 50 values of 3000 µM and 0.5 µM, respectively.
An essential metric is the selectivity index (SI = IC 50 /MIC).A compound is deemed promising if its SI is above 10.Derivative 4 displayed an SI of 53.6, whereas DIOX's SI stood at 300.While derivative 4 had a lower SI than DIOX, its value was still above 10, underscoring the potential of further investigations into its antituberculosis properties [22].

In Vivo Study Activity of Derivative 4 in Murine Tuberculosis Model
Mice infected with the autoluminescent reporter strain of M. tuberculosis AlRa were treated with derivative 4 both orally and intraperitoneally (I.P.).Rifampicin (10 mg/kg) served as a positive control.The positive control exhibited slower activity but remained significantly different from the test drug across all tested doses (Figure 4).

In Vivo Study Activity of Derivative 4 in Murine Tuberculosis Model
Mice infected with the autoluminescent reporter strain of M. tuberculosis AlRa were treated with derivative 4 both orally and intraperitoneally (I.P.).Rifampicin (10 mg/kg) served as a positive control.The positive control exhibited slower activity but remained significantly different from the test drug across all tested doses (Figure 4).While derivative 4 did not exhibit immediate efficacy at any dosage, no toxicity was observed, as all treated mice survived.A statistical analysis using Dunnett's multiple test indicated that the drug's activity could be influenced by the administration route (Supplementary Materials, Table S2).

Discussion
Quinoxaline is known for its biological and pharmacological activity [5].In this study, we detailed the synthesis and antimicrobial potency of new quinoxaline 1,4-dioxide derivatives.
We evaluated the antimicrobial activity of these compounds in vitro using M. smegmatis and M. tuberculosis AlRa cultures.Derivative 4 stood out, demonstrating the lowest While derivative 4 did not exhibit immediate efficacy at any dosage, no toxicity was observed, as all treated mice survived.A statistical analysis using Dunnett's multiple test indicated that the drug's activity could be influenced by the administration route (Supplementary Materials, Table S2).

Discussion
Quinoxaline is known for its biological and pharmacological activity [5].In this study, we detailed the synthesis and antimicrobial potency of new quinoxaline 1,4-dioxide derivatives.
We evaluated the antimicrobial activity of these compounds in vitro using M. smegmatis and M. tuberculosis AlRa cultures.Derivative 4 stood out, demonstrating the lowest MICs for both cultures: 1.25 and 4 µg/mL, respectively.Consequently, we chose derivative 4 for further cytotoxicity and in vivo analyses.The MTT assay on human fibroblasts showed derivative 4's SI to be 53.6.Notably, any compound with an SI ≥ 10 is regarded as a promising candidate for further exploration.
Our in vivo study on mice infected with the autoluminescent M. tuberculosis AIRa revealed no early activity for derivative 4 at any dosage.Still, there were no recorded mortalities.Interestingly, intraperitoneal administration appeared slightly more effective than oral administration, suggesting potential challenges with oral assimilation for derivative 4.
Detailed study of the mode of actions of QdNOs will provide information about their biological targets and the mechanism of action and will be valuable to construct models to screen new drugs.Whole-genome analysis revealed numerous mutations in all obtained drug-resistant strains, which indicated the high mutagenic potential of derivative 4. One of the possible mechanisms of action of quinoxaline 1,4-dioxide derivatives is the direct introduction of single-and double-stranded breaks in the DNA of bacteria, which could explain the numerous mutations in M. smegmatis qdr strains [23].Despite this, the exact mechanism of action of QdNOs has not yet been established.We identified four mutations in genes MSMEG_1380, MSMEG_4646, MSMEG_4648, and MSMEG_5122 that were presented in several qdr-mutants simultaneously.We have previously shown that mutations in MSMEG_1380, which encodes a transcriptional regulator, lead to the overexpression of MmpS5-MmpL5.This overexpression can result in mycobacterial resistance to a wide range of antibiotics, including thiacetazone, bedaquiline and clofazimine [16], imidazo [1,2-b][1,2,4,5]tetrazines [15], and tryptanthrin [19].To confirm the role of MmpS5-MmpL5 in drug resistance, all synthesized compounds were screened in our previously described test system [21].M. smegmatis ∆mmp5 was more sensitive to compounds 4-6, while M. smegmatis mc 2 155 and M. smegmatis atr9c had the same level of resistance to all compounds.The fact that overexpression of MmpS5-MmpL5 in the atr9c strain did not result in increased resistance but absence of efflux resulted in decreased resistance might indicate that MmpS5-MmpL5 can provide a basal level of resistance.
Through a reverse genetics approach, we confirmed that mutations in MSMEG_4646 and in the promoter region of MSMEG_5122 mediated resistance of M. smegamtis to the described QdNOs.MSMEG_4646 encodes the alpha subunit of ferredoxin oxidoreductase (pyruvate synthase) involved in pyruvate metabolism.MSMEG_5122 encodes ferredoxin, which acts as an electron acceptor for pyruvate synthase during the process of oxidation of pyruvate to acetyl-CoA.MSMEG_4648 is annotated as a DNA-binding protein and can be a transcriptional regulator of the nearby operon, which encodes the alpha and beta subunits of the mentioned pyruvate synthase.The role of MSMEG_4648 was not obvious because the mutation in this gene was observed with a mutation in the promoter region of MSMEG_5122 simultaneously.The measured MIC values of the constructed strain M. smegmatis 4646c were the same as that of the w.t.strain to all characterized compounds.
We established that mutations in the promoter of MSMEG_5122 and in MSMEG_4648 changed the expression of MSMEG_5122 and MSMEG_4645 with MSMEG_4648, respectively.However, the expressions of MSMEG_4645 and MSMEG_4648 were altered insignificantly, while MSMEG_5122 was decreased by almost 3 times.Thus, the reduced expression of ferredoxin encoded by MSMEG_5122 resulted in increased resistance, which might indicate that ferredoxin takes part in the activation of 4 or in the metabolic pathways of its mechanism of action.Meanwhile, pyruvate synthase encoded by MSMEG_4646 can participate in the inactivation of derivative 4.
In conclusion, QdNOs demonstrate potential as valuable scaffolds for new drug development.The action mechanism and biological targets of QdNOs remain subjects of debate.However, our findings suggest that these compounds have significant bacterial DNA-damage capabilities and impact pyruvate oxidation metabolic pathways.Further research involving the transcriptomic analysis of M. smegmatis mc 2 155 treated with derivative 4 at various concentrations and time spans may provide additional insights on QdNOs' mechanism of action.

Materials and General Methods
NMR spectra were recorded on a Varian Mercury 400 Plus instrument operated at 400 MHz ( 1 H NMR) and 100 MHz ( 13 C NMR) or a Bruker AVANCE III 500 (Bruker Biospin, Rheinstetten, Germany) NMR spectrometer equipped with a broadband Z-gradient probehead with a direct observe BB coil (PABBO) at 500.18MHz for 1 H and 125.77MHz for 13 C, respectively.Chemical shifts were measured in DMSO-d 6 using TMS as an internal standard.Spectra for all obtained compound were recorded in DMSO-d 6 solutions at 303 K and were referenced against residual solvent signals: 2.50 ppm for DMSO-d5 for 1H and 39.50 ppm for DMSO-d 6 for 13 C, respectively.The 1D and 2D NMR spectra were processed using TopSpin 3.2 Bruker (Bruker Biospin, Rheinstetten, Germany) or ACD Laboratories Spectral Processor Academic Edition (Advanced Chemistry Development.ACD, Inc., Toronto, ON, Canada, 2012.Software program).The 1 H and 13 C signal assignment was conducted by using 1H{ 13 C} HSQC, 1H{ 13 C} HMBC, and 2D NOESY NMR experimental data.Standard pulse sequences were used.For 2D NOESY and selective NOESY experiments, mixing times of 400 and 600 ms were correspondingly used.For selective excitation, an 80 ms Gaussian-shaped pulse was used.Analytical TLC was performed on silica gel F254 plates (Merck, Rahway, NJ, USA) and column chromatography on Silica Gel Merck 60.Melting points were determined on a Buchi SMP-20 apparatus and are uncorrected.High-resolution mass spectra were recorded via electron spray ionization on a Bruker Daltonics micro OTOF-QII instrument.UV spectra were recorded on a Hitachi-U2000 spectrophotometer (Hitachi High-Technologies Corporation (HHC), Hitachinaka, Japan).IR spectra were recorded on Nicolet iS10 Fourier transform IR spectrometer (Nicolet iS10 FT-IR, Thermo Scientific, Madison, WI, USA).The purity of compounds 4-7, 9 was >95% as determined by HPLC analysis.All products were dried under vacuum at room temperature.HPLC was performed using a Shimadzu Class-VP V6.12SP1 system (GraseSmart RP-18, 6 × 250 mm) with eluents as follows: A, H 3 PO 4 (0.01 M); B, MeCN.All solutions were evaporated at reduced pressure on a Buchi-R200 rotary evaporator (Büchi Labortechnik AG, Postfach, Switzerland) at a temperature below 50 • C. All products were dried under vacuum at room temperature.All solvents, chemicals, and reagents were obtained from Sigma-Aldrich (St. Louis, MO, USA) (unless specified otherwise) and used without purification.

Synthesis 4.2.1. General Procedure for Synthesis of Compounds 2-9
The quinoxalines 2, 3 and 8, 9 used in this study were previously described and synthesized in [24].The preparation of the compounds 2, 3 and 8 was carried out by the Beirut reaction as presented in Scheme 1.The starting 5,6-dichlorobenzofuroxan (1) was obtained by using a previously described method [25].

Biology 4.4.1. Microbial Cultures and Growth Conditions
All the bacterial cultures used in this study are summarized in Table 4. M. smegmatis strains were grown in liquid Middlebrook 7H9 medium (Himedia, Mumbai, India) supplemented with oleic albumin dextrose catalase (OADC, Himedia), 0.1% Tween-80 (v/v), and 0.4% glycerol (v/v), while M290 Soyabean Casein Digest Agar (Himedia) was used as the solid media.Escherichia coli DH5α was used for plasmid propagation and was grown in liquid or on agarized LB medium with the addition of the corresponding selection antibiotics when required.Cultures in liquid medium were incubated in a Multitron incubator shaker (Infors HT, Bottmingen-Basel, Switzerland) at 37 • C and 250 rpm.Minimal inhibitory concentrations (MICs) of the studied compounds on M. smegmatis were determined in a liquid medium.M. smegmatis strains were cultured overnight in 7H9 medium and then diluted at a proportion of 1:200 in fresh medium (to approximately OD 600 = 0.05).Then, 196 µL of the diluted culture was poured in sterile non-treated 96-well flat-bottom culture plates (Eppendorf, Hamburg, Germany); 4 µL of serial twofold dilutions of the tested compounds in DMSO were added to the wells to reach final concentrations ranging from 0.5 to 32 µg/mL, while 4 µL of DMSO was used as an untreated control.The plates were incubated at 37 • C and 250 rpm for 48 h.The experiments were carried out in triplicates, and the MIC was determined as the lowest concentration of the compound with no visible bacterial growth.

MIC Determination on M. tuberculosis Strains
Serial dilutions of drug-containing solutions and autoluminescent M. tuberculosis H37Ra (AlRa) strain broth culture (OD 600 of 0.3 to 1.0) were prepared.Relative light unit (RLU) counts from the same batch of triplicate samples were measured according to the designed time points.Ninety-six-well plates were measured as rendering the amount of light by using a VeritasTM Microplate Luminometer Operating Manual (Promega, Madison, WI, USA).The MIC lux was determined as the lowest concentration that could inhibit > 90% of the RLUs compared with that of the untreated controls.The MIC lux of the autoluminescent strains was determined by detecting RLUs from these samples as described previously [26,27].The concentrations used were a two-fold series of dilutions for this purpose that ranged from 0.078 to 20 µg/mL (a total of nine concentrations).The light from each 1.5 mL tube containing a total of 500 µL of culture was detected using GloMax 20/20n (Promega, Madison, WI, USA).Each experiment above was performed in triplicate (three independent experiments) or more by independent persons and from three or more independent cultures.

Paper Disk Assay
M. smegmatis cultures were incubated at 250 rpm and 37 • C overnight until OD 600 = 1.2.Afterwards, the drug susceptibility was assessed by using a paper-disc assay: M. smegmatis culture was diluted 1:9:10 (culture:water:М290 medium) and seeded over the base agar layer on Petri dishes.The compounds were manually impregnated on sterile paper discs.The plates were incubated for 2 days at 37 • C until the bacterial lawn was fully grown.Growth inhibition halos were measured to the nearest 0.1 mm (the halo area around the disk was measured with a digital caliper and analyzed in a Mega Bio-Print 3020-WL/LC/20M X-Press, Vilber Lourmat Gel Documentation System).The experiments were carried out as three repeats, and the average diameter and standard deviation (SD) were calculated.We used CLSI standards containing information about disk diffusion (M02) and dilution (M07, M11) test procedures for bacteria and Laboratory Practices in Microbiology [28].

Generation of Resistant Mutants and Their Phenotype Characterizations
The M. smegmatis strain was grown in a liquid medium to reach OD 600 = 2.8 (~4 × 10 8 CFU/mL).Then, 200 µL of bacterial culture was plated on agar plates containing 4 at a final concentration of 32 µg/mL.The plates were incubated in a thermostat at 30 • C for three days until the emergence of single colonies.The colonies from two plates for each strain were counted to determine the frequency of drug-resistant mutants' emergence.Serial 10-fold dilutions of each bacterial culture were plated on compound-free plates to determine the exact titer.The mutants' resistance phenotypes were confirmed by streaking several colonies on M290 plates containing 32 µg/mL of 4. The w.t.strain was used as a control.The MICs on resistant mutants were determined in a liquid medium as described above.4.4.5.M. smegmatis Whole-Genomic Sequencing and Analysis Mycobacterial genomic DNA was isolated from 10 mL via enzymatic lysis as described by Belisle et al. [29].After preliminary isolation, the DNA was treated with RNase A (Thermo Fischer Scientific, Waltham, MA, USA) and extracted in the phenol-chloroformisoamyl alcohol solution (25:24:1).
A total of 250 ng of genomic DNA was taken for shotgun sequencing library preparation.After DNA sonication on a Covaris S220 System (Covaris, Woburn, MA, USA), the size (400-500 b.p.) and quality of the fragmented samples were assessed on an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA) according to the manufacturer's manual.A NEBNext Ultra II DNA Library Prep Kit (New England Biolabs, Ipswich, MA, USA) was used for pair-ended library preparation, and a NEBNext Multiplex Oligos kit for Illumina (96 index primers, New England Biolabs) was used for library indexing.The libraries were quantified with a Quant-iT DNA Assay Kit, High Sensitivity (Thermo Fischer Scientific).DNA sequencing (2 × 125 b.p.) was performed on a HiSeq 2500 platform (Illumina, San Diego, CA, USA) according to the manufacturer's recommendations.
The quality of the obtained reads was assessed with FastQC (v.0.11.8)(FastQC: A Quality Control Tool for High Throughput Sequence Data); as the quality was good, we proceeded to the assembly without trimming the reads.The reads were aligned to the reference genome (NC_008596.1,PRJNA57701) using the BWA-MEM algorithm [30].The pileup was generated with mpileup (-B-f) in SAMtools [31].Single-nucleotide variants were called by running mpileup2snp (-min-avg-qual 30-min-var-freq 0.80-p-value 0.01-outputvcf 1-variants 1) in VarScan (v.2.3.9)[32].The annotation was created using vcf_annotate.pl(developed by Natalya Mikheecheva of the Laboratory of Bacterial Genetics, Vavilov Institute of General Genetics, Moscow, Russia).The non-synonymous single-nucleotide variants found within open reading frames and absent in the wild-type strain were selected for further analysis.The raw sequencing data (SRA) are publicly available in NCBI GenBank (BioProject ID: PRJNA778794).

Construction of the M. smegmatis Recombinant Strains
To generate M. smegmatis strains carrying single mutations in the genes MSMEG_4646, MSMEG_4648, and MSMEG_5122, based on the M. smegmatis mc 2 155 strain, homologous recombination using the suicide system p2NIL/pGOAL19 [20] was applied.Briefly, using the Q5 high-fidelity PCR kit (NEB, Ipswich, MA, USA), primers containing specific restriction sites, genomic DNA from spontaneous M. smegmatis mutants with target mutations, corresponding fragments containing mutations in the target genes, and ~1500 bp homologous arms for homologous recombination were amplified.After digestion with appropriate restriction endonucleases (Thermo Fisher Scientific, Waltham, MA, USA) of the amplicons and the p2NIL plasmid, ligation and selection of the target clones were performed using the method described above.Next, the target plasmids were cloned with a cassette containing marker genes from the pGOAL19 plasmid at the PacI restriction site.The final plasmids were electroporated into M. smegmatis mc 2 155 cells.Selection of single crossovers was performed on tryptone-soy agar containing kanamycin (50 µg/mL), hygromycin (50 µg/mL), and X-Gal (50 µg/mL).Since the p2NIL plasmid lacks mycobacterial replication origin, the function of marker genes was only possible upon plasmid integration into the chromosome via homologous recombination.Blue single-crossover colonies were grown overnight in liquid Middlebrook 7H9 medium and then plated onto tryptone-soy agar containing X-Gal (50 µg/mL) and 10% sucrose for selection of double crossovers, and their sensitivity to kanamycin was tested to confirm the complete removal of the cassette containing marker genes from the chromosome.For final selection of recombinants carrying the desired single-nucleotide replacement, the target region was sequenced via Sanger sequencing.

RNA Extraction and Real-Time PCR
To isolate the total RNA, M. smegmatis cultures were grown in 10 mL Middlebrook 7H9 medium until reaching an optical density (OD 600 ) of approximately 1.0, after which they were washed twice with 3 mL of RNAprotect Bacteria Reagent (QIAGEN, Germantown, MD, USA).The total bacterial RNA was extracted using the RNeasy Plus Mini Kit (QIA-GEN, USA) following the manufacturer's protocol with the use of Lysing Matrix B silicon beads (MP Biomedicals, Santa Ana, CA, USA) for homogenization.After extraction and purification, the RNA was treated with TURBO DNAse (Thermo Fisher Scientific, USA), and the concentration was measured using the Qubit instrument (Invitrogen, Waltham, MA, USA) with an RNA BR Assay Kit (Thermo Fisher Scientific, USA).The integrity of the RNA was evaluated via agarose gel electrophoresis.The first-strand cDNA synthesis was performed using the iScript Select cDNA Synthesis Kit (Bio-Rad, USA) with 300 ng of total RNA.
For quantitative real-time PCR, the qPCR-HS SYBR Kit (Eurogen, Moscow, Russia) and 5 ng of cDNA were used on the CFX96 Touch instrument (Bio-Rad, Hercules, CA, USA).The results were analyzed using CFX Manager V 3.1 software (Bio-Rad, USA), and the relative normalized expression was calculated as ∆∆Cq using the sigA and ftsZ genes as references.

Cytotoxic Measurement Assay
The cytotoxic effects of various concentrations of derivative 4 on human dermal fibroblasts (HDF) were measured using an MTT assay.This assay is based on the measurement of succinate dehydrogenase activity, a mitochondrial enzyme, and is widely used for in vitro toxicity testing of potential drugs.Using the MTT assay data, the half-maximum inhibition concentration (IC 50 ) was calculated.The primary HDF cells were obtained from a healthy donor.The HDF cells were cultured in DMEM medium (10% FBS, 2 mM glutamine, and 1% gentamicin) in plastic flasks under sterile conditions and incubated at 37 • C in the presence of 5% CO 2 .A stock solution of 4 (50 mM) was prepared in DMSO; before adding it to the cells, it was diluted to the desired concentrations in the culture medium, and HDF cells were seeded into the wells of 96-well plates at a density of 3.5 × 10 3 cells per well.Cells were allowed to attach for 14 h, after which different concentrations of the test compound or DMSO (as a control sample) were injected in triplicate using the titration method.The final volume in the wells was 100 µL.After 48 h of sample addition, cell viability was measured using the MTT reagent (Sigma, Livonia, MI, USA) at a volume of 10 µL per well.In each well, the working solution (7 mg/mL) was added (100 µL per well), and the wells were incubated for 3 h, after which the medium was replaced with DMSO solution.The optical density of each well was determined at 570 nm using a plate spectrophotometer (TECAN Infinite M Plex) with subtraction of the background absorbance.The concentration value and the inhibitory dose (IC 50 ) were determined from the dose-response curves.4.4.9.In Vivo Activity against M. tuberculosis A rapid drug susceptibility screening of the compound was carried out in vivo using AlRa.Briefly, at Day -1 (D -1), each of the BALB/c mice was intravenously infected with 250 µL of a mid-log phase culture of AlRa at ~2,000,000 relative light units per milliliter (RLU/mL).In addition, each mouse was identified with a color tag for individual followup of the drug effect on bacterial growth.Compound 4 was dissolved in 10% Tween 80, and solvent and rifampicin were used as negative and positive controls, respectively.Daily treatment was initiated a day after infection (Day 0 (D 0)), and bacterial growth was monitored at two-day intervals and on the last day of treatment (Day 5 (D 5)).Drugs were administered through the following routes at the following doses (mg/kg): rifampicin, 10 (oral); 4, 30 (oral); 4, 100 (oral); 4, 100 (intraperitoneally); and 4, 200 (oral).

Figure 2 .
Figure 2. Growth inhibition halos (mm) for derivatives 4-7 in the M. smegmatis test system columns represent the mean value, while the error bars represent the standard deviations ( from three independent experiments.

Figure 2 .
Figure 2. Growth inhibition halos (mm) for derivatives 4-7 in the M. smegmatis test system.The columns represent the mean value, while the error bars represent the standard deviations (SDs) from three independent experiments.

Figure 2 .
Figure 2. Growth inhibition halos (mm) for derivatives 4-7 in the M. smegmatis test system columns represent the mean value, while the error bars represent the standard deviations from three independent experiments.

Figure 3 .
Figure 3. Relative alterations in the expressions of genes MSMEG_4645, MSMEG_4646, MSMEG_4648, and MSMEG_5122 in recombinant strains M. smegmatis 4648c and M. smegmatis 5122c.The error bars represent the standard deviation calculated from three independent replicates.

Figure 4 .
Figure 4.In vivo activity of derivative 4 against AlRa.All drugs were administered orally except where shown in parentheses.Administered dosages (mg/kg) are shown after drug names.RLU, relative light units; D 0, day 0 and so on; IP, intraperitoneal route.

Figure 4 .
Figure 4.In vivo activity of derivative 4 against AlRa.All drugs were administered orally except where shown in parentheses.Administered dosages (mg/kg) are shown after drug names.RLU, relative light units; D 0, day 0 and so on; IP, intraperitoneal route.

Table 1 .
MICs of studied compounds on mycobacterial strains.DIOX) served as the primary reference derivative of quinoxaline 1,4-dioxide used in clinical settings, while rifampicin (RIF) was used as the antimycobacterial reference drug.Compound 4 exhibited the highest activity against M. tuberculosis AlRa, with MIC values 8 times lower than the reference DIOX.Moreover, it displayed equivalent activity against M. smegmatis mc 2 155.Generally, the remaining derivatives showed reduced activities against M. smegmatis, with MIC values of at least 16 µg/mL and diminished effectiveness against M. tuberculosis AlRa.Except for compound 4, all other synthesized compounds had a limited antimycobacterial effect.Notably, compound 9 did not exhibit any MICs against either mycobacterial species and was consequently excluded from further MIC measurements.

Table 4 .
Microbial cultures used in the study.