MmpS5-MmpL5 Transporters Provide Mycobacterium smegmatis Resistance to imidazo[1,2-b][1,2,4,5]tetrazines

The emergence and spread of drug-resistant Mycobacterium tuberculosis strains (including MDR, XDR, and TDR) force scientists worldwide to search for new anti-tuberculosis drugs. We have previously reported a number of imidazo[1,2-b][1,2,4,5]tetrazines–putative inhibitors of mycobacterial eukaryotic-type serine-threonine protein-kinases, active against M. tuberculosis. Whole genomic sequences of spontaneous drug-resistant M. smegmatis mutants revealed four genes possibly involved in imidazo[1,2-b][1,2,4,5]tetrazines resistance; however, the exact mechanism of resistance remain unknown. We used different approaches (construction of targeted mutants, overexpression of the wild-type (w.t.) and mutant genes, and gene-expression studies) to assess the role of the previously identified mutations. We show that mutations in MSMEG_1380 gene lead to overexpression of the mmpS5-mmpL5 operon in M. smegmatis, thus providing resistance to imidazo[1,2-b][1,2,4,5]tetrazines by increased efflux through the MmpS5-MmpL5 system, similarly to the mechanisms of resistance described for M. tuberculosis and M. abscessus. Mycobacterial MmpS5-MmpL5 transporters should be considered as an MDR-efflux system and they should be taken into account at early stages of anti-tuberculosis drug development.


Introduction
Tuberculosis (TB), caused by Mycobacterium tuberculosis, is currently the leading killer among the infectious diseases caused by one infectious agent, responsible for an estimate of 1.2 million deaths in 2018 [1]. According to WHO, 1.7 billion people globally are infected with M. tuberculosis, and thus are at risk of developing the disease [1]. The emergence and spread of multidrug resistant TB (MDR-TB, defined as TB resistant to rifampicin and isoniazid), extensively drug-resistant TB (defined as MDR-TB with resistance to the fluoroquinolones and second-line injectables) and totally drug-resistant TB (TDR-TB) is a global threat to world-wide TB control [2][3][4][5]. Long treatment times (six months for drug-susceptible TB and up to two years for DR-TB) often lead to bad patient compliance, which is one of the causes of drug resistance development and results in worse treatment outcomes. Thus, researchers are forced to search for novel anti-TB drugs and shorter regimens [6].
Eukaryotic-type serine-threonine protein-kinases (ESTPKs) play a key role in M. tuberculosis life cycle regulation, controlling some of its vital aspects such as cell division and survival within host macrophages, and, therefore, they represent attractive targets for drug development [7,8]. We have previously described a number of imidazo [1,2-b] [1,2,4,5]tetrazines with a promising antibacterial Pathogens 2020, 9,166 2 of 8 activity on M. tuberculosis and M. smegmatis [9]. Most of these compounds showed activity as potential ESTPK inhibitors in the original M. smegmatis aphVIII+ test-system [10,11], and two of them were able to bind to the M. tuberculosis PknB adenine-binding pocket according to docking studies [10]. Despite the predicted activity as ESTPK-inhibitors, both the exact mechanism of action and the mechanism of resistance to these compounds are still unknown.
Pathogens 2020, 9,166 2 of 9 activity on M. tuberculosis and M. smegmatis [9]. Most of these compounds showed activity as potential ESTPK inhibitors in the original M. smegmatis aphVIII+ test-system [10,11], and two of them were able to bind to the M. tuberculosis PknB adenine-binding pocket according to docking studies [10]. Despite the predicted activity as ESTPK-inhibitors, both the exact mechanism of action and the mechanism of resistance to these compounds are still unknown. We were able to obtain spontaneous M. smegmatis mutants resistant to four imidazo [1,2b] [1,2,4,5]tetrazines (3a, 3c, 3h and 3n, Figure 1), which had cross-resistance among them, suggesting a common mechanism of drug-resistance [9]. Whole-genomic sequencing and comparative genomic analysis revealed mutations in MSMEG_0641 (binding-protein-dependent transporters inner membrane component) in 1 mutant, in MSMEG_1601 (hypothetical protein) in seven mutants, in MSMEG_2087 (transporter small conductance mechanosensitive ion channel (MscS) family protein) in one mutant [12], while all the mutants carried different mutations in MSMEG_1380 (AcrR/TetR_N transcriptional regulator) -1 nonsynonymous SNP, 2 insertions leading to a frameshift, 2 duplications (6 and 501 base pairs-long) and one deletion [13].

Mutations in MSMEG_1380 Gene Lead to imidazo
The list of nonsynonymous mutations found in spontaneous drug-resistant M. smegmatis mutants used in this study is presented in Table 1. We were able to construct targeted M. smegmatis mutants harboring each mutation in genes MSMEG_0641, MSMEG_1601, and MSMEG_2087, as well as five mutations found in MSMEG_1380 gene using the p2NIL/pGOAL19 suicide system [14] for homologous recombination (Table 1).  [9] 3a 3c 3h 3n In this article we describe the investigation of these mutations' role in mycobacterial drug resistance to imidazo [1,2-b] [1,2,4,5]tetrazines by different approaches: construction of targeted mutants, overexpression of the wild-type (w.t.) and mutant genes, and gene-expression studies. [1,2,4,5]tetrazines Resistance in M. smegmatis

Mutations in MSMEG_1380 Gene
The list of nonsynonymous mutations found in spontaneous drug-resistant M. smegmatis mutants used in this study is presented in Table 1. We were able to construct targeted M. smegmatis mutants harboring each mutation in genes MSMEG_0641, MSMEG_1601, and MSMEG_2087, as well as five mutations found in MSMEG_1380 gene using the p2NIL/pGOAL19 suicide system [14] for homologous recombination (Table 1).
We examined the minimal inhibitory concentrations (MICs) of the four compounds on the recombinant M. smegmatis strains and found that mutations in MSMEG_1380 gene lead to elevated MICs as compared to the w.t. strain (4×MIC for the compound 3a, at least 2×MIC for the compound 3c, at least 2×MIC for the compound 3h, and 4×MIC for the compound 3n), while recombinants harboring mutations in genes MSMEG_0641, MSMEG_1601, and MSMEG_2087 had the same MICs as the w.t. strain (Table 2).  [15].
We used the paper-disc assay to assess the drug susceptibility of M. smegmatis strains to imidazo [1,2-b] [1,2,4,5]tetrazines and found that the overexpression of the w.t. MSMEG_1380 gene increases M. smegmatis susceptibility to the tested compounds, while the overexpression of its mutant variants had no effect on the phenotype (Table 3), thus suggesting that the disruption of MSMEG_1380 protein's function leads to the drug-resistant phenotype.  Figure 2). The structure of this operon is conserved in different mycobacterial species: the mmpS5-mmpL5 genes are controlled by a TetR-family transcriptional repressor, encoded by a gene located upstream the operon. Mutations in genes encoding the TetR-repressor, which lead to the upregulation of the mmpS5-mmpL5 genes, are involved in M. abscessus resistance to the derivatives of thiacetazone [16], as well as in cross-resistance of M. tuberculosis to bedaquiline and clofazimine [17]. We have also identified a possible operator sequence in the 5 -untranslated region of MSMEG_1381, similar to the one described in [16]: 5 -AAGCGGATTGACCTTATCCACTT-3 .
We used the paper-disc assay to assess the drug susceptibility of M. smegmatis strains to imidazo [1,2-b] [1,2,4,5]tetrazines and found that the overexpression of the w.t. MSMEG_1380 gene increases M. smegmatis susceptibility to the tested compounds, while the overexpression of its mutant variants had no effect on the phenotype (Table 3), thus suggesting that the disruption of MSMEG_1380 protein's function leads to the drug-resistant phenotype.  Figure 2). The structure of this operon is conserved in different mycobacterial species: the mmpS5-mmpL5 genes are controlled by a TetR-family transcriptional repressor, encoded by a gene located upstream the operon. Mutations in genes encoding the TetR-repressor, which lead to the upregulation of the mmpS5-mmpL5 genes, are involved in M. abscessus resistance to the derivatives of thiacetazone [16], as well as in cross-resistance of M. tuberculosis to bedaquiline and clofazimine [17]. We have also identified a possible operator sequence in the 5′-untranslated region of MSMEG_1381, similar to the one described in [16]: 5′-AAGCGGATTGACCTTATCCACTT-3′.  To test the hypothesis that resistance to imidazo[1,2-b] [1,2,4,5]tetrazines in M. smegmatis has a similar origin to the ones described for M. tuberculosis and M. abscessus [16,17], we analyzed the expression of MSMEG_1380 gene and mmpL5 (MSMEG_1382) genes in different conditions.
All the spontaneous M. smegmatis mutants had increased mmpL5 expression (54.16-80.45 times) as compared to the w.t. M. smegmatis mc2 155 strain ( Figure 3A). The overexpression of the w.t. MSMEG_1380 gene, cloned into the pMINDKmplasmid led to a 7.90-fold repression of the mmpL5 gene expression (p < 0.001, Figure 3B), confirming that MSMEG_1380 encodes the repressor of the mmpS5-mmpL5 operon, and explaining the drug-susceptible phenotype, observed in the MSMEG_1380 overexpressing strain. On the contrary, the expression of MSMEG_1380 was upregulated in the mutant strains ( Figure 3A), indicating that this transcriptional repressor is self-regulatory and that mutations lead to the loss of its function.
We also observed that the addition of subinhibitory concentrations of the compound 3a upregulated the expression of mmpL5 in a dose-dependent manner ( Figure 3C). mmpS5-mmpL5 operon, and explaining the drug-susceptible phenotype, observed in the MSMEG_1380 overexpressing strain. On the contrary, the expression of MSMEG_1380 was upregulated in the mutant strains ( Figure 3A), indicating that this transcriptional repressor is selfregulatory and that mutations lead to the loss of its function.
We also observed that the addition of subinhibitory concentrations of the compound 3a upregulated the expression of mmpL5 in a dose-dependent manner ( Figure 3C).

Discussion
Deorphaning phenotypic screening hits, that is determining their mechanism of action and/or resistance, is a key part in the early-stage anti-TB drug development [18]. Here we determine the mechanism of M. smegmatis imidazo [1,2-b] [1,2,4,5]tetrazines resistance based on the previously obtained whole-genome sequencing data for 12 spontaneous mutants with cross-resistance to four compounds [9,13].
The construction of targeted mutants showed that only mutations in MSMEG_1380 are responsible for drug resistance. In M. smegmatis, MSMEG_1380 encodes a TetR-family transcriptional repressor, which controls the mmpS5-mmpL5 operon, encoding transmembrane transporters, conserved throughout mycobacterial species. Mutations occurring in MSMEG_1380 led to the upregulation of the mmpS5-mmpL5 operon and increased efflux of the drug-candidates from the cells, similarly to the mechanisms described for M. tuberculosis and M. abscessus [16,17]. Overexpression of the w.t. MSMEG_1380 led to the repression of the mmpS5-mmpL5 operon and expectedly to an increased drug susceptibility phenotype. Interestingly, we observed a dose-dependent upregulation of the mmpS5-mmpL5 operon upon the addition of one of the compounds, which may indicate the ability of this compound to bind to the MSMEG_1380 protein, inhibiting its affinity to the operator sequence; however, this needs to be examined in vitro in future studies.
The tested compounds showed activity as ESTPK inhibitors [9,11]; however, we have not observed any mutations in ESTPK genes. One or more ESTPKs might still be the biotargets of imidazo [1,2-b] [1,2,4,5]tetrazines but determining them by spontaneous mutagenesis might be difficult: some of the ESTPKs may fulfill the functions of others in the situation when they might be inhibited [8], and there is a possibility that more than one mutation might be required.
The primary biological role of the MmpS5-MmpL5 system consists in siderophore transport, which is crucial for M. tuberculosis survival under low-iron conditions within macrophages [19]. Yet, this efflux system has also shown itself to be an important factor of drug resistance: besides the mentioned efflux-mediated resistance to thiacetazone derivatives, bedaquiline and clofazimine [16,17], it has also been reported to provide M. tuberculosis resistance to azoles [20]. We can expect that M. tuberculosis strains resistant to bedaquiline and clofazimine might also be resistant to imidazo [1,2-b] [1,2,4,5]tetrazines; however, a 36% mismatch in the amino acid sequences of the

Discussion
Deorphaning phenotypic screening hits, that is determining their mechanism of action and/or resistance, is a key part in the early-stage anti-TB drug development [18]. Here we determine the mechanism of M. smegmatis imidazo [1,2-b] [1,2,4,5]tetrazines resistance based on the previously obtained whole-genome sequencing data for 12 spontaneous mutants with cross-resistance to four compounds [9,13].
The construction of targeted mutants showed that only mutations in MSMEG_1380 are responsible for drug resistance. In M. smegmatis, MSMEG_1380 encodes a TetR-family transcriptional repressor, which controls the mmpS5-mmpL5 operon, encoding transmembrane transporters, conserved throughout mycobacterial species. Mutations occurring in MSMEG_1380 led to the upregulation of the mmpS5-mmpL5 operon and increased efflux of the drug-candidates from the cells, similarly to the mechanisms described for M. tuberculosis and M. abscessus [16,17]. Overexpression of the w.t. MSMEG_1380 led to the repression of the mmpS5-mmpL5 operon and expectedly to an increased drug susceptibility phenotype. Interestingly, we observed a dose-dependent upregulation of the mmpS5-mmpL5 operon upon the addition of one of the compounds, which may indicate the ability of this compound to bind to the MSMEG_1380 protein, inhibiting its affinity to the operator sequence; however, this needs to be examined in vitro in future studies.
The tested compounds showed activity as ESTPK inhibitors [9,11]; however, we have not observed any mutations in ESTPK genes. One or more ESTPKs might still be the biotargets of imidazo [1,2-b] [1,2,4,5]tetrazines but determining them by spontaneous mutagenesis might be difficult: some of the ESTPKs may fulfill the functions of others in the situation when they might be inhibited [8], and there is a possibility that more than one mutation might be required.
The primary biological role of the MmpS5-MmpL5 system consists in siderophore transport, which is crucial for M. tuberculosis survival under low-iron conditions within macrophages [19]. Yet, this efflux system has also shown itself to be an important factor of drug resistance: besides the mentioned efflux-mediated resistance to thiacetazone derivatives, bedaquiline and clofazimine [16,17], it has also been reported to provide M. tuberculosis resistance to azoles [20]. We can expect that M. tuberculosis strains resistant to bedaquiline and clofazimine might also be resistant to imidazo [1,2-b] [1,2,4,5]tetrazines; however, a 36% mismatch in the amino acid sequences of the MmpL5 proteins in M. smegmatis and M. tuberculosis may affect the drug-specificity of the transporter, and this should be additionally examined in future studies. Still, the MmpS5-MmpL5-mediated resistance mechanism needs to be considered during early stages of anti-TB drug development, and convenient in silico and in vitro test-systems for rapid analysis should be developed.

Bacterial Strains and Growth Conditions
M. smegmatis strains described in this study are presented in Table 1. Middlebrook 7H9 medium (Himedia, India) supplemented with OADC (Himedia, India), 0.1% Tween-80 (v/v), and 0.4% glycerol (v/v) was used as liquid medium, while the M290 Soyabean Casein Digest Agar (Himedia, India) was used as solid medium. Escherichia coli DH5α was used for plasmids propagation. Cultures in liquid medium were incubated in the Multitron incubator shaker (Infors HT, Switzerland) at 37 • C and 250 rpm.

MIC Determination
MICs of the studied compounds on M. smegmatis were determined in liquid medium. M. smegmatis strains were cultured overnight in 7H9 medium, then diluted in the proportion of 1:200 in fresh medium (to approximately OD 600 = 0.05). 196 µl of the diluted culture was poured in sterile nontreated 96-well flat-bottom culture plates (Eppendorf, Germany) and 4 µL of serial two-fold dilutions of the tested compounds in DMSO were added to the wells. The plates were incubated at 37 • C and 250 rpm for 48 h. The MIC was determined as the lowest concentration of the compound with no visible bacterial growth.