Rv3488 of Mycobacterium tuberculosis H37Rv has been assigned to the phenolic acid decarboxylase repressor (PadR) family of transcriptional regulators that play key roles in multidrug resistance and virulence of prokaryotes. The binding of cadmium, zinc, and several other metals to Rv3488 was discovered and characterized by isothermal titration calorimetery to be an exothermic process. Crystal structures of apo-Rv3488 and Rv3488 in complex with cadmium or zinc ions were determined by X-ray crystallography. The structure of Rv3488 revealed a dimeric protein with N-terminal winged-helix-turn-helix DNA-binding domains composed of helices α1, α2, α3, and strands β1 and β2, with the dimerization interface being formed of helices α4 and α1. The overall fold of Rv3488 was similar to PadR-s2 and metal sensor transcriptional regulators. In the crystal structure of Rv3488–Cd complex, two octahedrally coordinated Cd2+ ions were present, one for each subunit. The same sites were occupied by zinc ions in the structure of Rv3488–Zn, with two additional zinc ions complexed in one monomer. EMSA studies showed specific binding of Rv3488 with its own 30-bp promoter DNA. The functional role of Rv3488 was characterized by expressing the rv3488 gene under the control of hsp60 promoter in Mycobacterium smegmatis. Expression of Rv3488 increased the intracellular survival of recombinant M. smegmatis in murine macrophage cell line J774A.1 and also augmented its tolerance to Cd2+ ions. Overall, the studies show that Rv3488 may have transcription regulation and metal-detoxifying functions and its expression in M. smegmatis increases intracellular survival, perhaps by counteracting toxic metal stress.

Introduction

Tuberculosis is one of the most fatal diseases that affects millions of people all over the world. There have been 1.3 million deaths reported worldwide in 2016 [1]. In recent years, there has been a growth in multiple drug resistance (MDR) Mycobacterium tuberculosis, which is a cause of great concern. In most of the clinical isolates, antibiotic resistance is often associated with overexpression of efflux pumps [2], which is controlled at the level of transcription. The transcription regulators of MDR transporters in bacteria belong to one of the following four regulatory protein families: AraC, MarR, MerR, and TetR [3]. The multiple antibiotic resistance family (MarR) of transcriptional regulators consists of a large number of repressors like EmrR and MexR of Escherichia coli and Pseudomonas aeruginosa, respectively [4,5].

PadR-like family of transcription regulators are closely related to the MarR family [6]. Phenolic acids induce the phenolic acid stress response (PASR) in Gram-positive bacteria [711], which regulates the expression of genes associated with detoxification mechanism. Phenolic acid decarboxylase (Pad) activity was first characterized in Pediococcus pentosaceus [12]. It was shown that this mechanism involved two proteins that were designated as phenolic acid decarboxylase or PadA (PadC in Bacillus subtilis) and phenolic acid decarboxylase repressor or PadR. It was further shown that the deletion of padA led to growth inhibition in the presence of phenolic acids, while deletion of padR led to constitutive overexpression of padA and to high resistance to phenolic acids [1315]. Based on the size of the C-terminal domain, PadR family proteins have been classified into two distinct subfamilies. PadR-like proteins of subfamily 1 (PadR-s1) are ∼180 amino acids in length and include LadR, AphA, MtrY, VanR, and PadR from Listeria monocytogenes, Vibrio cholerae, Streptomyces argillaceus, Corynebacterium glutamicum, and B. subtilis, respectively [7,8,1618]. They have relatively large C-terminal domains of ∼80–90 amino acids containing multiple α-helices. While subfamily 2 PadR-like proteins (PadR-s2) possess ∼110 amino acids and include LmrR, BcPadR1, BcPadR2, Pex, CdPadR1, and BF2549 from Lactococcus lactis, Bacillus cereus, Cyanobacterium circadian, Clostridium difficile R20291, and Bacteroides fragilis, respectively [911,19,20]. They have significantly smaller C-terminal domains of ∼20–30 amino acids forming a single α-helix. A common structural feature of both the subfamilies is a conserved N-terminal (∼70 amino acids) winged-helix-turn-helix (wHTH) DNA-binding domain. The wHTH motif, which is composed of α1–α2–α3 helices and β1–β2 wing strands of PadR family, binds with the operator DNA.

Several members within the PadR-s1 subfamily have been structurally and functionally characterized. The PadR of B. subtilis (BsPadR) has a wHTH region, which represses the expression of padC by interacting with the operator DNA of padC gene [8]. Recently, the crystal structures of BsPadR protein in complex with p-coumaric and ferulic acids, and also with a 28-bp ds-DNA containing a palindromic sequence in the padC operator, have been reported [8]. The palindromic sequence of the ds-DNA was recognized by wHTH motifs of dimeric PadR (composed of α1–α2–α3–β1–β2–α4). The α2 and α3 helices of individual motifs interacted with the major groove of ds-DNA, whereas the wings (β1–β2) of the wHTH motifs interacted with the adjacent minor grooves of the operator DNA. The α5–α6–α7 helices together formed the C-terminal coiled-coil dimerization domain, mainly involving antiparallel interactions between the α7 helices of two monomers. p-Coumaric acid was bound in a narrow interdomain pocket between the N-terminal and C-terminal domains formed by the α2–α3 loop of NTD and helices α6 and α7 of the CTD. Furthermore, a dimeric PadR-type repressor, VanR, was shown to be involved in the regulation of vanABK operon that encodes the vanillate utilization components in C. glutamicum [7]. The structure of VanR in complex with vannilate was determined recently [21]. VanR has wHTH NTD and CTD similar to PadR. However, VanR contains an additional α-helix, αi, that functions as an interdomain cross-linker between the NTD and the CTD and contributes a key arginine residue that binds to the vanillate molecule. Among the other characterized members of the PadR-s1 family, AphA was shown to activate the transcription of the tcpPH promoter on the Vibrio pathogenicity island, in conjunction with AphB, and initiate a transcriptional cascade that culminated in the production of TCP and cholera toxin. While, in S. argillaceus, the PadR-like regulator MtrY was shown to play a dual role in the mithramycin gene cluster by repressing the expression of resistance genes in the absence of mithramycin and by enhancing the expression of mithramycin biosynthesis genes when mithramycin was present [18]. For L. monocytogenes, it was shown that the expression of multidrug efflux pump, MdrL, was repressed at the transcriptional level, under standard growth conditions, by LadR, a PadR-like transcriptional regulator, and the expression of MdrL was induced in the presence of rhodamine [16].

The PadR-s2 subfamily proteins contain a single α-helix in the C-terminal effector-binding/oligomerization domain. Among the PadR-s2 subfamily members, BcPadR1 and BcPadR2, from B. cereus, have been structurally and functionally characterized. BcPadR2 was suggested to regulate the expression of a putative antibiotic ABC efflux pump [10]. In L. lactis, the expression of its major MDR transporter, LmrCD, was shown to be regulated by a transcription regulator, LmrR, of the PadR-like family. The crystal structures of LmrR, with and without bound H33342 or daunomycin, were determined. In the dimeric structure, each monomer had topology α1–α2–α3–β1–β2–α4. The helix α4 formed a protruding arm, which interacted with α1′, α2′, and the α2′–α3′ loop of the other monomer. This dimeric arrangement resulted in the formation of a large hydrophobic pore running through the dimer center. In the drug-bound structures, the flat ring systems of the drugs were wedged in between the W96 and W96' side-chains forming aromatic stacking interactions with each of the two indole systems. This binding changed the distance between the two DNA-binding helices α3 and α3′ [9]. The structure of the PadR-s2 protein BF2549 from B. fragilis was determined recently [11]. The structure consists of the minimal α1–α2–α3–β1–β2–α4 domain. Homodimerization of BF2549 was mediated by the N-terminal helix α1 and the C-terminal helix α4. It was suggested that the PadR structures were characterized by an elongated tip comprising of 19 residues in the wing region.

The genome of M. tuberculosis H37Rv has three members of the PadR-like transcriptional regulators family. Based on sequence length and homology, Rv0047c and Rv1176c belong to the PadR-s1 subfamily, while Rv3488 belongs to the PadR-s2 subfamily. The rv3488 gene is located downstream of lipF [22] and encodes a 107 amino acid protein, which has been annotated as a member of the PadR family of regulators. Rv3488 has been shown to bind at a minimal acid-inducible promoter region of LipF [23].

To characterize the structure and function of Rv3488, we cloned, expressed, and crystallized Rv3488, and determined its crystal structure in the apo-form by X-ray crystallography. We were not able to establish and demonstrate the binding of Rv3488 to phenolic acids, like p-coumaric acid, ferulic acid, and caffeic acid, through isothermal titration calorimetery (ITC). However, during the course of purification, the affinity of untagged Rv3488 for the Ni2+-NTA matrix was noticed. In M. tuberculosis, the concentration of nickel is regulated by a metalloregulatory repressor, NmtR, which is a member of the ArsR/SmtB family of metal-sensing proteins. Other members of this family, like ArsR, SmtB, KmtR, MntR (SirR), and CmtR, are involved in regulation of various metals like arsenic, bismuth, lead, zinc, nickel, cobalt, manganese, and cadmium. In addition, the regulation of copper and mercury concentration in M. tuberculosis is carried by the CsoR and MerR proteins, respectively. Given the binding of Rv3488 to nickel and similarity of its structure to several of the ArsR/SmtB family proteins, we systematically established the metal-binding affinities of Rv3488 for several metals including nickel, manganese, zinc, cadmium, copper, and mercury, by using ITC. The highest binding affinities of Rv3488 were observed for cadmium, manganese, and zinc. Therefore, the cadmium- and zinc-binding sites were characterized by determining the structure of Rv3488 in complex with cadmium or zinc.

Furthermore, binding of Rv3488 specifically to its own promoter DNA was characterized by electrophoretic mobility shift assays (EMSA). We investigated the functional role of Rv3488 by generating a recombinant Mycobacterium smegmatis strain heterologously expressing M. tuberculosis rv3488 gene under the control of hsp60 promoter. Further study revealed that Rv3488 increased the intracellular survival of recombinant M. smegmatis without affecting the rate of growth in the J774A.1 macrophages and also augmented the tolerance of the bacilli against sublethal concentration of Cd2+. Overall, the studies show that Rv3488 may have transcription regulation and metal-detoxifying functions through which it may enhance intracellular survival of mycobacteria.

Experimental procedures

Materials

Reagents for cloning, pfu DNA polymerase, and protein molecular mass markers were obtained from Fermentas (U.S.A.). Restriction enzymes, DNA size markers, and T4 DNA ligase were from NEB (U.S.A.). Gel extraction kit, plasmid miniprep kit, and nickel–nitrilotriacetic acid (Ni2+-NTA) superflow metal-affinity chromatography matrix were obtained from Qiagen. Agar powder and Luria–Bertani broth were from Hi-Media (India). Isopropyl-β-d-1-thiogalactopyronoside (IPTG) was from Merck Biosciences (Germany). Ultrafiltration membranes used for protein concentration were from Pall Corporation (U.S.A.). Polyvinylidene difluoride membrane and Enhanced Chemiluminescence kit (ECL) were from GE Healthcare. Middlebrook 7H9 broth, MB7H10 agar, ADC, and OADC enrichments were from BD Biosciences. Rest of the chemicals were obtained from Sigma.

Cloning of rv3488

The rv3488 gene was amplified by PCR using the primers 5′-CCAGAATTCATGCGGGAGTTTCAGCGGGC-3′ and 5′-CCCAAGCTTTCAGGTCCCATTACCAGCGGT-3′ containing EcoRI and HindIII restriction enzyme sites. The PCR-amplified product was subjected to restriction digestion with the respective enzymes and ligated into pET-NH6 vector. The presence of the correct rv3488 sequence in the plasmid construct was confirmed by DNA sequencing.

Expression and purification of Rv3488

The recombinant plasmid (pET-NH6::rv3488) was transformed into E. coli strain BL21 (λDE3) cells. Cells containing pET-NH6::rv3488 were grown in 1.5 l of LB medium supplemented with 100 µg/ml ampicillin, at 37°C. At OD600 nm of 1, the cells were induced by adding 0.4 mM IPTG and were further allowed to grow at 22°C for ∼14 h. Cells were harvested by centrifugation and homogenized by sonication in lysis buffer (50 mM Tris–HCl, pH 8.0, containing 300 mM NaCl, 10 mM imidazole, and 100 µM phenyl methanesulfonyl fluoride). The ruptured cells were removed by centrifugation at 9000 rpm and the cleared lysate was applied to a Ni2+-NTA resin column. The 6× His-tag protein was eluted using 300 mM imidazole in Tris–HCl lysis buffer. His-tag was removed by digestion with TEV protease. The Rv3488 protein and TEV were maintained in the concentration of 10 : 1 for the digestion reaction, which was continued for 16 h at 22°C. After the digestion of His-tag, imidazole was removed by dialysis in buffer containing 50 mM Tris–HCl, pH 8.0, 300 mM NaCl, and 10 mM imidazole. The dialyzed protein was again purified with a Ni2+-NTA column. The uncut proteins as well as the TEV protease (which has a 6× His-tag) were bound to the column, while untagged-Rv3488 was recovered from the flow through.

Size-exclusion chromatography

His-tag free Rv3488 was further purified by size-exclusion chromatography (SEC) using a Superdex 75-10/30GL column (Amersham Pharmacia Biotech, Sweden; exclusion limit 75 kDa for proteins) on the Bio-Rad FPLC system. Rv3488 was injected into the column equilibrated in buffer containing 20 mM Tris–HCl, 100 mM NaCl, and 5% glycerol, pH 7.5. Rv3488 protein sample (1 ml) was loaded on the column at a flow rate of 0.4 ml/min, and the eluted protein was detected online by following the absorbance at 280 nm, at room temperature. The column was calibrated with standard molecular mass markers like Conalbumin (75.0 kDa), Ovalbumin (43.0 kDa), Chymotrypsinogen (25.0 kDa), and Ribonuclease A (13.7 kDa) (GE Healthcare Lifesciences). The apparent molecular mass of the Rv3488 protein was determined from the calibration curve. Fractions containing pure Rv3488 were pooled and concentrated to 10 mg/ml for crystallization using a 3 kDa cutoff centricon.

Crystallization, data collection, and structure determination of Rv3488

Preliminary crystallization conditions of Rv3488 protein were screened with the hanging drop vapor diffusion method in 24-well plates at 296 K. Rv3488 crystallized under the condition in which 2 µl of 10 mg/ml Rv3488 protein was mixed with 2 µl of reservoir solution of 0.1 M sodium acetate pH 4.6, and 2 M sodium formate, and equilibrated against the reservoir solution. The crystals grew in the drops within 2 weeks. For the X-ray data collection, a single crystal was mounted on nylon loop and flash-cooled in a nitrogen stream at 100 K. All diffraction data were collected with Rigaku FR-E+ SuperBright with a wavelength λ = 1.5418 Å using the R-AXIS IV++ detector. A complete data set (430°, 0.75° oscillation diffraction data) was collected and diffraction data were indexed, integrated, and scaled using the HKL-2000 package [24]. The Rv3488 protein structure was determined using the molecular replacement method with the program MOLREP [25] with structure co-ordinates (PDB ID: 4EJO) of a PadR family protein from Eggerthella lenta (ElPadR1) as a search model. The Rv3488 protein structure was iteratively built and refined using COOT [26] and REFMAC5 [27] programs, respectively, and progress was monitored by evaluating the Rfree factor.

For the Rv3488–Cd and Rv3488–Zn-bound structures, prior to the data collection, good quality crystals of Rv3488 were soaked in mother liquor (0.1 M sodium acetate, pH 4.6, and 2 M sodium formate) containing 10% glycerol and 10 mM CdCl2 or 10 mM ZnCl2 for 30 min. The soaked crystals were mounted directly from the mother liquor, flash-cooled, and X-ray diffraction data were collected. These structures were refined in a similar manner that was employed for the native Rv3488.

Differential scanning calorimetry

Differential scanning calorimetry (DSC) experiments were performed using VP-DSC calorimeter from MicroCal (GE). The volume of sample and reference cells was 0.5 ml. All experiments were carried out in Tris–HCl buffer. Before each sample run, buffer–buffer baseline run was recorded under exactly same conditions. Rv3488 protein was dialyzed in Tris–HCl buffer. Both the sample and buffer solutions were thoroughly degassed for 20 min just before the experiment. DSC scans were performed from 30 to 80°C at a heating rate of 60°C per hour. Buffer scans were subtracted from the sample scans, and the data were normalized with respect to protein concentration, scan rate, and electrical calibration of the calorimeter, to generate the excess heat capacity versus temperature thermogram of the sample. The baselines before and after transition were selected for the thermogram with the ORIGIN 7.0 program. The transition enthalpy (ΔH), T1/2, and the temperature at the midpoint of the unfolding transition (Tm) were analyzed by integration and nonlinear curve fitting to a two-state model. All experiments were repeated at least three times.

Isothermal titration calorimetry

ITC experiments were performed at 25°C on a VP-ITC calorimeter (MicroCal™ U.S.A.). The thermodynamic parameters governing complex formation between Rv3488 and metal chlorides were determined. The buffer condition for both ion solution and protein solution was 20 mM Tris–HCl, pH 7.5, 100 mM NaCl, and 5% glycerol. The protein and buffer solutions were degassed for 20 min before ITC experiments. For titrations with ZnCl2, MnCl2, and NiCl2, the sample cell was loaded with 1.45 ml of 60 µΜ Rv3488 solution, while 292 µl of a 600 µΜ solution of the respective metal salt solution, in the same buffer, was placed in the syringe. For titration with cadmium, the syringe was filled with 60 µΜ solution of the CdCl2, while 3 µΜ Rv3488 protein solution in the same buffer was placed into the sample cell. The reference cell was loaded with the degassed buffer. During titrations, the solution was stirred with a syringe at 416 rpm in sample cell without foaming of the protein solution. Aliquots (10 µl) of metal chloride were added sequentially at an interval of 180 s. Control experiments were determined by titration of metal chloride solution into plain buffer solution. The heats of titration of the control experiment were subsequently subtracted from the data of metal chloride and Rv3488 titration. The analysis of ITC data was carried out using the ORIGIN Version 7.0 software provided by MicroCal. The amount of the heat generated per titration was determined by using the integration of the area under the peak.

Electrophoresis mobility shift assays for Rv3488

To determine the binding of Rv3488 with the promoter region, EMSAs were performed. A 30 base oligonucleotide (5′-GAATCGCAAGCCGATATCGGCTTGGTCACC-3′) and its complementary oligonucleotide, encompassing the region from −41 to −12 of the rv3488 promoter, were purchased from Integrated DNA Technologies, U.S.A. The 30-bp oligonucleotide, which contained an 18-bp palindromic region, was labeled at 5′-end with Cy3. A random 30-bp fragment was used as the control DNA, and it was also labeled at 5′-end with Cy3. The ds-DNA was prepared by annealing in the annealing buffer [20 mM Tris–HCl (pH 7.5), 50 mM NaCl, and 1 mM EDTA] at 95°C for 15 min. The Cy3-labeled DNA fragments (0.1 µM) were incubated with the purified Rv3488 protein at concentrations ranging from 0.6 to 3 µM and were premixed on ice in the binding buffer containing 20 mM Tris–HCl (pH 7.5), 100 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol, 0.25 mg/ml bovine serum albumin as a non-specific protein competitor, and 5 µg/ml of Poly [d(I-C)] as a nonspecific DNA competitor. Samples were loaded on a 6% polyacrylamide gel after 15 min incubation. Native PAGE was run in 1× TBE buffer containing 0.089 M Tris–HCl, 0.089 M boric acid, and 0.022 M EDTA, at 50 V for 90 min. After the electrophoresis, the Cy3-labeled DNA was detected using ImageQuant LAS4000 (GE Healthcare Life Sciences Ltd). For the competition experiments, different amounts (50- and 80-fold molar excesses) of unlabeled 30-bp DNA were added as a competitor during the binding step. The ligand-binding experiment was performed by incubating 20 µM of Rv3488 protein with 2.5 mM cadmium chloride for 15 min prior to adding the Cy3-labeled 30-bp DNA (ds-DNA) strands for the assay.

To determine the binding of Rv3488 protein with the promoter region of rv1999c, the necessary oligonucleotides were also purchased from Integrated DNA Technologies. The unlabeled DNA fragment was prepared by annealing as mentioned above and incubated with different concentrations of purified Rv3488 protein (from 6 to 15 µM) in DNA-binding buffer and loaded on a 6% polyacrylamide gel, as described above. After the electrophoresis, polyacrylamide gels were stained with ethidium bromide and visualized under UV light.

Immunization of rabbit for polyclonal antibody production

Female adult white New Zealand rabbit was immunized with 200 µg of purified Rv3488 protein, emulsified in Freund's incomplete adjuvant, by subcutaneous injection [28]. The rabbit was given four booster doses of the antigen on days 0, 15, 30, and 45, respectively. On day 60, the rabbit was bled, and the serum was separated by incubating at 37°C for 2–3 h. The antiserum was confirmed for its reactivity to Rv3488 by Western blotting. The immunization of the rabbit was performed according to guidelines of Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA) and CSIR-Central Drug Research Institute, Lucknow, India (ethical approval number: IAEC/2016/6).

Subcloning of rv3488 in pMV361: an integrative mycobacterial expression vector

For subcloning of the M. tuberculosis rv3488 gene into pMV361 shuttle vector, primer pairs (FP: 5′-CCAGAATTCATGCGGGAGTTTCAGCGGGC-3′, RP: 5′-CCCAAGCTTAGGACTGGTATTCGGCGGTT-3′) with added EcoRI and HindIII sites were used to amplify the complete rv3488 operon from M. tuberculosis H37Rv genomic DNA, as described above. The amplified fragment was digested and ligated into E. coli–Mycobacterium shuttle vector pMV361. pMV361 is a shuttle plasmid commonly used for cloning and expression of genes in mycobacteria. The clone was subsequently verified by DNA sequencing using hsp60 promoter primers and electroporated in M. smegmatis mc2155 as described previously [29]. The recombinant colonies were selected on kanamycin and screened for the presence of recombinant vector by PCR.

Overexpression of Rv3488 in M. smegmatis mc2155

The expression of Rv3488 in M. smegmatis was confirmed by Western blotting [30]. Briefly, the PCR-positive colonies selected on kanamycin plate were inoculated in MB7H9 media supplemented with 10% ADC and kanamycin (25 µg/ml) and grown up to mid-log-phase at 37°C on shaking conditions. The bacterial cells were harvested by centrifugation at 6000 rpm for 10 min at room temperature. The cells were lysed by sonication and centrifuged at 8000 rpm for 45 min at 4°C. The supernatant was collected and the protein was quantified by the Bradford assay using BSA as a standard. The protein samples were resolved by 12% SDS–PAGE, transferred on to the PVDF membrane, and immunoblotted with the Anti-Rv3488 polyclonal primary antibodies and HRP-conjugated antirabbit secondary antibodies.

Growth response of MS-pMV361::rv3488 recombinant strain

The primary cultures of MS-pMV361 and MS-pMV361::rv3488 were inoculated in 10 ml of Sauton's media containing kanamycin (25 µg/ml), for 48 h at 37°C. Optical density did not exceed 1.0 at OD600nm. Cells were diluted to an OD600nm of 0.005 in Sauton's medium right before the assay was performed. The cultures were grown with shaking at 180 rpm at 37°C. Optical densities at OD600nm were measured in a spectrometer to assess growth at different time points after inoculation, i.e. 0, 4, 8, 12, 24, 36, 60, and 80 h. Statistical analysis was done by GraphPad Prism 5.0.

Expression profiling of Rv3488 protein in mycobacteria

The rv3488 gene expression was profiled in different species of mycobacteria. M. tuberculosis H37Rv, M. tuberculosis H37Ra, M. smegmatis, and Mycobacterium bovis BCG cells were grown in MB7H9 media, and the lysates of log-phase culture were prepared. Whole cell lysates (∼40.0 µg) were resolved on 15% SDS–PAGE. The rv3488 gene expression was analyzed by Western blotting.

Microplate alamar blue assay of mycobacterial viability

Microplate alamar blue assay (MABA) was performed by using a standardized procedure [31], to estimate the influence of metal ions such as cadmium chloride, zinc chloride, mercury chloride, and copper chloride on the viability of M. smegmatis cells. Serial dilutions of metal chloride were mixed in 150 µl of MB7H9 medium with the bacterial suspension (1 × 105 cells per well) of MS-pMV361::Rv3488 and empty vector transformed MS-pMV361 cells in 96-well plates. The tested divalent metal cation concentrations were Hg2+ (0–3 µm), Cu2+ (0–125 µM), Zn2+ (0–10 mM), and Cd2+ (0–3.2 µM). The cultures were incubated at 37°C for 48 h. Resazurin (10% v/v; 0.3% w/v) was added to each well and further incubated for 1–2 h at 37°C and fluorescence was measured at 535 /590 nm in a spectrophotometer. Untreated and rifampicin-treated cells served as negative and positive controls, respectively. Experiments were performed in triplicates.

Time-dependent in vitro growth and intracellular survival studies

To assess the impact of the Rv3488 protein on the in vitro culture and intracellular growth of surrogate M. smegmatis, the intracellular growth of MS:rv3488 cells, constitutively expressing Rv3488, was monitored in murine-derived J774A.1 cells, as described previously [32,33]. Briefly, 12-well plates were seeded with 1 × 105 cells per well and infected at 1 : 10 m.o.i (multiplicity of infection). After 2 h, the cells were washed thrice with incomplete RPMI-1640 medium. To kill the non-phagocytosed bacilli, the cells were treated with 100 µg/ml amikacin. For quantification of intracellular bacteria, the cells were lysed in sterile distilled H2O at 0, 12, and 24 h post-infection and lysates were plated on MB7H10 agar plates supplemented with 10% OADC and 25 µg/ml kanamycin. The phagocytosis and intracellular persistence of mycobacteria were estimated by plating and determining the CFU/ml.

Results

Multiple sequence alignment of Rv3488 with PadR-s2 family proteins

Rv3488 has similarity with PadR-s2 transcriptional regulators. It possesses 107 amino acids and shares 39, 31, 21, 31, 30, and 36% protein sequence identity with ElPadR (PDB ID: 4EJO), LmrR (PDB ID: 3F8B), CdPadR (PDB ID: 5DYM), BcPadR1 (PDB ID: 4ESB), BcPadR2 (PDB ID: 4ESF), and BF2549 (PDB ID: 5H20), respectively. The sequences of these proteins were retrieved and multiple sequence alignment was performed with the Rv3488 sequence as the query sequence using ClustalW (http://www.ebi.ac.uk/clustalomega/). The start site of Rv3488 is marked as an asterisk and conserved residues are colored in red bars, as shown in Figure 1. The five strictly conserved residues are Y44, R49, R68, Y71, and G77. Y44 and R49 are on helix α3; R68 and Y71 are on β-strand 2. These four residues are involved in binding to the promoter ds-DNA. G77 stabilizes the conformation of the wing β-strands relative to the DNA recognition helix and dimerization domain.

Multiple sequence alignment of Rv3488 from M. tuberculosis H37Rv with PadR-s2 subfamily proteins of known crystal structures, as represented by their PDB IDs as follows: 4EJ0 (E. lenta), 3F8B (L. lactis), 5DYM (C. difficile), 4ESB (B. cereus strain 14579), 4ESF (B. cereus strain 10987), and 5H20 (B. fragilis).

Figure 1.
Multiple sequence alignment of Rv3488 from M. tuberculosis H37Rv with PadR-s2 subfamily proteins of known crystal structures, as represented by their PDB IDs as follows: 4EJ0 (E. lenta), 3F8B (L. lactis), 5DYM (C. difficile), 4ESB (B. cereus strain 14579), 4ESF (B. cereus strain 10987), and 5H20 (B. fragilis).

The beginning of Rv3488 sequence is indicated with an asterisk. Conserved amino acids are colored in red bars. The gaps in the alignment are represented as dots. The sequence alignment figure was made with ESPript 3 program.

Figure 1.
Multiple sequence alignment of Rv3488 from M. tuberculosis H37Rv with PadR-s2 subfamily proteins of known crystal structures, as represented by their PDB IDs as follows: 4EJ0 (E. lenta), 3F8B (L. lactis), 5DYM (C. difficile), 4ESB (B. cereus strain 14579), 4ESF (B. cereus strain 10987), and 5H20 (B. fragilis).

The beginning of Rv3488 sequence is indicated with an asterisk. Conserved amino acids are colored in red bars. The gaps in the alignment are represented as dots. The sequence alignment figure was made with ESPript 3 program.

Crystal structure of Rv3488

The cloning, expression, and purification of Rv3488 are described in Supplementary Figure S1A–C. The dimeric nature of Rv3488 in solution was established by SEC (Supplementary Figure S1D). Thermal stability of Rv3488 was characterized by a Tm of 67.9°C, which was determined by using DSC (Supplementary Figure S2). Details of cloning, expression, purification, SEC, and DSC are given in Supplementary Material.

The structure of Rv3488 (PDB ID: 5ZHC) was solved by molecular replacement and refined to 1.97 Å resolution (Table 1). Each asymmetric unit contained two subunits making a biologically relevant dimer. The average B-factors for subunit-A and subunit-B were 34.57 and 33.40 Å2, respectively. The crystallographic B-factor distribution over the Rv3488 dimer reveals that the dimerization region is quite well ordered compared with the DNA-binding domain. Rv3488 forms a flat triangular-shaped dimer and adopts similar overall fold like all other PadR-s2 family members, as shown by the cartoon representation in Figure 2. Each Rv3488 subunit is composed of four α helices and two β strands with a topology of α1 (residues 2–19)–α2 (residues 24–33)–α3 (residues 40–52)–β1 (residues 56–63)–β2 (residues 66–73)–α4 (residues 75–95). The final model contains residues 1–96, along with six additional N-terminal amino acid residues overhang remaining after the rTEV protease cleavage, one acetate, two chloride ions, and 75 water molecules. The root-mean-square deviation (RMSD) between subunit-A and subunit-B of the apo-Rv3488 was 0.5 Å (residues 1–96). In both subunits of the dimer, the C-terminal residues 97–107 were not visible in the electron density map and were, therefore, not included in the structure.

Overall structure of apo-Rv3488 homodimer.

Figure 2.
Overall structure of apo-Rv3488 homodimer.

The protein structure is shown as a cartoon diagram with subunit-A in rainbow and subunit-B in gray. Dimeric structure of Rv3488 with acetate (ACT) molecule shown in ball-and-stick and chloride (Cl) molecules shown as green spheres. The secondary structure elements are labeled. The N- and C-terminal ends of Rv3488 subunit-A are labeled as N-terminal and C-terminal, respectively.

Figure 2.
Overall structure of apo-Rv3488 homodimer.

The protein structure is shown as a cartoon diagram with subunit-A in rainbow and subunit-B in gray. Dimeric structure of Rv3488 with acetate (ACT) molecule shown in ball-and-stick and chloride (Cl) molecules shown as green spheres. The secondary structure elements are labeled. The N- and C-terminal ends of Rv3488 subunit-A are labeled as N-terminal and C-terminal, respectively.

Table 1
Summary of diffraction data and structure refinement for the crystal structures of apo-Rv3488, Rv3488–Cd, and Rv3488–Zn
 Rv3488 Rv3488–Cd Rv3488–Zn 
Diffraction data 
 Wave length (Å) 1.5418 1.5418 1.5418 
 Space group P212121 P212121 P212121 
Cell parameters 
a (Å) 30.63 30.50 30.47 
b (Å) 49.10 49.29 49.21 
c (Å) 127.32 127.34 128.39 
α(°) = β(°) = γ(°) 90.0 90.0 90.0 
No. of protein molecule per asymmetric unit 
VM3 Da−11.89 1.86 1.89 
Solvent content (%) 35.05 35.69 35.05 
Resolution range (Å) 63.66–1.97 (2.01–1.97) 63.67–2.20 (2.28–2.20) 64.19–2.40 (2.49–2.40) 
No. of unique reflections 14 217 (685) 9947 (764) 8086 (768) 
Average redundancy 10.9 (6.8) 10.1 (6.0) 6.3 (5.6) 
Average I/σ 34.04 (2.06) 19.17 (2.32) 17.50 (2.37) 
Completeness (%) 99.80 (98.40) 95.60 (76.80) 99.90 (99.60) 
Rmerge (%) 7.5 (65) 12 (52) 11(66) 
Refinement statistics 
 Rfactor 0.185 0.191 0.210 
 Rfree 0.237 0.239 0.266 
Deviation from ideal geometry 
 Bond length (Å) 0.013 0.017 0.010 
 Bond angle (°) 1.57 1.81 1.41 
Protein model 
 No. of subunit in A.U. 
 No. of atoms 1720 1709 1664 
 Protein atoms 1639 1650 1617 
B-factors 
 Main chain 31.84 32.08 38.64 
 Side-chain 37.34 37.04 41.22 
No. of atoms 
 Zinc – – 
 Chloride – – 
 Cadmium – – 
 Acetate  – 
 Water molecules 75 57 43 
Ramachandran plot (%) 
 Mostly favored regions 100.0 100.0 100.0 
 Additionally allowed regions 0.0 0.0 0.0 
 Generously allowed regions 0.0 0.0 0.0 
 Disallowed regions 0.0 0.0 0.0 
PDB ID 5ZHC 5ZI8 5ZHV 
 Rv3488 Rv3488–Cd Rv3488–Zn 
Diffraction data 
 Wave length (Å) 1.5418 1.5418 1.5418 
 Space group P212121 P212121 P212121 
Cell parameters 
a (Å) 30.63 30.50 30.47 
b (Å) 49.10 49.29 49.21 
c (Å) 127.32 127.34 128.39 
α(°) = β(°) = γ(°) 90.0 90.0 90.0 
No. of protein molecule per asymmetric unit 
VM3 Da−11.89 1.86 1.89 
Solvent content (%) 35.05 35.69 35.05 
Resolution range (Å) 63.66–1.97 (2.01–1.97) 63.67–2.20 (2.28–2.20) 64.19–2.40 (2.49–2.40) 
No. of unique reflections 14 217 (685) 9947 (764) 8086 (768) 
Average redundancy 10.9 (6.8) 10.1 (6.0) 6.3 (5.6) 
Average I/σ 34.04 (2.06) 19.17 (2.32) 17.50 (2.37) 
Completeness (%) 99.80 (98.40) 95.60 (76.80) 99.90 (99.60) 
Rmerge (%) 7.5 (65) 12 (52) 11(66) 
Refinement statistics 
 Rfactor 0.185 0.191 0.210 
 Rfree 0.237 0.239 0.266 
Deviation from ideal geometry 
 Bond length (Å) 0.013 0.017 0.010 
 Bond angle (°) 1.57 1.81 1.41 
Protein model 
 No. of subunit in A.U. 
 No. of atoms 1720 1709 1664 
 Protein atoms 1639 1650 1617 
B-factors 
 Main chain 31.84 32.08 38.64 
 Side-chain 37.34 37.04 41.22 
No. of atoms 
 Zinc – – 
 Chloride – – 
 Cadmium – – 
 Acetate  – 
 Water molecules 75 57 43 
Ramachandran plot (%) 
 Mostly favored regions 100.0 100.0 100.0 
 Additionally allowed regions 0.0 0.0 0.0 
 Generously allowed regions 0.0 0.0 0.0 
 Disallowed regions 0.0 0.0 0.0 
PDB ID 5ZHC 5ZI8 5ZHV 

Rmerge(I) = ∑hkli|Ii(hkl) − 〈I(hkl)〉|/∑hkliIi(hkl) for n independent reflections and i observations of a given reflection. 〈I(hkl)〉 is the average intensity of the i observations.

Rwork and Rfree = ∑h||F(h)o| − |F(h)c||/∑h|F(h)o|, where F(h)o and F(h)c are the observed and calculated structure-factor amplitudes, respectively. Rfree was calculated using 5% of data.

Values in the parentheses are for the highest resolution range.

The analysis of crystal structure of Rv3488 protein reveals that each subunit is composed of DNA-binding and dimerization domains. The DNA-binding domain is composed of three helices α1, α2, α3 (DNA recognition helix), strand β1, and strand β2, which adopt the wHTH fold. The ‘wing’ consists of two antiparallel beta strands β1 and β2. The dimerization domain is formed by two long helices: N-terminal α1 and C-terminal α4 that join together to form a head-to-head dimer. The long α1 and α4 helices from each monomer come together to form a four helical bundle and the dimer is held together by a salt bridge, hydrophobic interactions, and hydrogen bonds. Upon dimerization, Rv3488 buries an accessible surface area of ∼1630 Å2 per monomer. Six and eight salt bridges are observed in subunit-A and subunit-B of the crystal structure, respectively (Table 2). Residue H34 of subunit-A is involved in two salt bridges with E94 of subunit-B of the structure, while residue H12 of both subunits are involved in intermolecular salt bridges with E94, as shown in Supplementary Figure S3 (Table 3).

Table 2
Salt bridges found in the crystal structure of the Rv3488 analyzed with ESBRI server
Rv3488 crystal structure (subunit-A) Distance (Å) Rv3488 crystal structure (subunit-B) Distance (Å) 
NH2: R10/OD2: D53 3.04 NH1: R10/OD2: D53 3.14 
NH1: R61/OE1: E59 3.59 NH2: R10/OD2: D53 3.10 
NH2: R61/ OE1: E21 3.79 NH1: R33/OE1: E30 3.01 
NH1: R72/ OE1: E59 3.34 NH1: R61/OE1: E21 2.78 
NH2: R72/ OE1: E59 2.80 NH2: E61/OE1: E21 3.20 
NH2: R61/OE1: E21 3.15 NH2: R61/OE1: E59 3.62 
  NH1: R72/OE1: E59 3.34 
  NH2: R72/OE1: E59 2.80 
Rv3488 crystal structure (subunit-A) Distance (Å) Rv3488 crystal structure (subunit-B) Distance (Å) 
NH2: R10/OD2: D53 3.04 NH1: R10/OD2: D53 3.14 
NH1: R61/OE1: E59 3.59 NH2: R10/OD2: D53 3.10 
NH2: R61/ OE1: E21 3.79 NH1: R33/OE1: E30 3.01 
NH1: R72/ OE1: E59 3.34 NH1: R61/OE1: E21 2.78 
NH2: R72/ OE1: E59 2.80 NH2: E61/OE1: E21 3.20 
NH2: R61/OE1: E21 3.15 NH2: R61/OE1: E59 3.62 
  NH1: R72/OE1: E59 3.34 
  NH2: R72/OE1: E59 2.80 
Table 3
Salt bridges found between subunit-A and subunit-B in the crystal structure of the Rv3488 analyzed with the ESBRI server
Salt bridge between subunit-A and subunit-B of Rv3488 crystal structure Distance (Å) 
NE2: H12 A/OE1: E94 B 3.22 
ND1: H34 A/ OE1: E94 B 3.51 
NE2: H34 A/ OE1: E94 B 3.54 
OE1: E94 A/NE2: H12 B 2.73 
Salt bridge between subunit-A and subunit-B of Rv3488 crystal structure Distance (Å) 
NE2: H12 A/OE1: E94 B 3.22 
ND1: H34 A/ OE1: E94 B 3.51 
NE2: H34 A/ OE1: E94 B 3.54 
OE1: E94 A/NE2: H12 B 2.73 

In the difference Fourier (Fo − Fc) and (2Fo − Fc) maps of the Rv3488 structure, extra electron density was found at three sites, in which one acetate and two chloride ions were fitted (Figure 2). The acetate and two chloride ions were refined on the basis of electron density shape, and chemical and geometric properties of the ligand-binding site. Acetate site 1 is located at dimerization domain, while chloride site 2 is located at the DNA-binding domain. Chloride ion may have been captured by the recombinant Rv3488 protein during expression and purification in E. coli, while acetate was provided as a crystallizing reagent. The co-ordination of acetate is shown in Supplementary Figure S4, and the co-ordinating distances are shown in Supplementary Table S1.

Description of the cadmium-binding sites

The crystal structure of the Rv3488–Cd (PDB ID: 5ZI8) has same space group P212121, unit cell dimensions, and overall fold as that of the apo-Rv3488 structure. For the Rv3488–Cd structure, Matthew's calculation (1.86 Å3 Da−1, 35.69% solvent content) indicates two molecules in the asymmetric unit. The RMSD between subunit-A and subunit-B of the Rv3488–Cd was 0.6 Å. The RMSD between subunit-A of Rv3488–Cd and subunit-A of apo-Rv3488 was 0.3 Å. In the crystal structure of Rv3488–Cd, two Cd2+ ions were present, one for each subunit (designated here as Cd1A and Cd1B) (Figure 3). In subunit-A, cadmium site (Cd1A) is located between the DNA-binding and dimerization domains of Rv3488, where it is co-ordinated by the side-chain atoms of NE2 of H16, OE2 of E30, and NE2 of H34. A similar co-ordination is observed for cadmium in site Cd1B of subunit-B. However, H101 (of subunit-B) is not traceable for the Cd1A site in subunit-A, while in subunit-B, cadmium ion is co-ordinated by NE2 atom of H101 of subunit-A, in addition to the three residues of subunit-B. In addition, the oxygen atoms of water molecules (26, 39, and 55 for Cd1A, and 28 and 29 for Cd1B) co-ordinate with the Cd2+ ions (Figure 4C,D). B-factors for the Cd2+ ions were 39.07 Å2 for both Cd1A and Cd1B. The co-ordination sphere of these sites can be described as octahedral geometry. All the metal-to-ligand distances are summarized in Table 4. The electron density around cadmium is shown in Figure 4A,B. Subunit-A and subunit-B have seven and nine intramolecular salt bridges, respectively. However, the salt bridge formed between H34 and E94 is lost upon cadmium binding. Additional salt bridges are formed between R33 and E30 in both the subunits.

Cartoon representation of the crystal structure of Rv3488–Cd dimer with subunit-A in pink and subunit-B in blue.

Figure 3.
Cartoon representation of the crystal structure of Rv3488–Cd dimer with subunit-A in pink and subunit-B in blue.

The Cd2+ ions are shown as red spheres. Cd2+-binding residues at subunit-A and subunit-B are shown as grey sticks. The secondary structure elements are labeled. The N- and C-terminal ends of Rv3488–Cd are labeled as N-terminal and C-terminal, respectively.

Figure 3.
Cartoon representation of the crystal structure of Rv3488–Cd dimer with subunit-A in pink and subunit-B in blue.

The Cd2+ ions are shown as red spheres. Cd2+-binding residues at subunit-A and subunit-B are shown as grey sticks. The secondary structure elements are labeled. The N- and C-terminal ends of Rv3488–Cd are labeled as N-terminal and C-terminal, respectively.

The co-ordination of Cd2+ ions in the crystal structure of Rv3488-Cd.

Figure 4.
The co-ordination of Cd2+ ions in the crystal structure of Rv3488-Cd.

(A) Cd2+-binding site at subunit-A. (B) Cd2+-binding site at subunit-B. The Cd2+ ions and Cd-coordinating residues are shown as red spheres and pink sticks, respectively. The water molecules are represented as gray spheres. The 2Fo − Fc electron density maps are countered at 1σ and are shown as light blue mesh. (C) Structural comparisons for Cd2+-binding residues. Superimposition of the Rv3488 (pink) and CdCl2 bound Rv3488 (green) structures showing metal-binding center. Cd2+-binding site at subunit-A. (D) Cd2+-binding site at subunit-B. Stick representation of Rv3488–Cd is shown in pink line, apo-Rv3488 in green line. The interactions between cadmium ion and residues of Rv3488–Cd are shown as black dashes.

Figure 4.
The co-ordination of Cd2+ ions in the crystal structure of Rv3488-Cd.

(A) Cd2+-binding site at subunit-A. (B) Cd2+-binding site at subunit-B. The Cd2+ ions and Cd-coordinating residues are shown as red spheres and pink sticks, respectively. The water molecules are represented as gray spheres. The 2Fo − Fc electron density maps are countered at 1σ and are shown as light blue mesh. (C) Structural comparisons for Cd2+-binding residues. Superimposition of the Rv3488 (pink) and CdCl2 bound Rv3488 (green) structures showing metal-binding center. Cd2+-binding site at subunit-A. (D) Cd2+-binding site at subunit-B. Stick representation of Rv3488–Cd is shown in pink line, apo-Rv3488 in green line. The interactions between cadmium ion and residues of Rv3488–Cd are shown as black dashes.

Table 4
Geometric parameters of the cadmium co-ordination at the two sites in the structure of Rv3488–Cd
Cd1A Distance (Å) Cd1B Distance (Å) 
His16 (NE2)/A 2.34 His16 (NE2)/B 2.33 
Glu30 (OE2)/A 2.16 Glu30 (OE2)/B 2.20 
His34 (NE2)/A 2.47 His34(NE2)/B 2.36 
Water 26 2.67 His101 (NE2)/A 2.33 
Water 39 2.31 Water 28 2.92 
Water 55 2.85 Water 29 2.38 
Cd1A Distance (Å) Cd1B Distance (Å) 
His16 (NE2)/A 2.34 His16 (NE2)/B 2.33 
Glu30 (OE2)/A 2.16 Glu30 (OE2)/B 2.20 
His34 (NE2)/A 2.47 His34(NE2)/B 2.36 
Water 26 2.67 His101 (NE2)/A 2.33 
Water 39 2.31 Water 28 2.92 
Water 55 2.85 Water 29 2.38 

Description of zinc-binding sites

The presence of Zn2+ ions in the crystal structure of Rv3488 (PDB: 5ZHV) was confirmed by difference electron density Fourier (Fo − Fc) and (2Fo − Fc) maps. It revealed that four Zn2+ ions (designated here as Zn1A, Zn1B, Zn2, and Zn3) were present in Rv3488–Zn crystal structure as shown in Figure 5. The B-factor for the Zn2+ ions is 70.0, 42.0, 96.4, and 99.6 Å2 for Zn1A, Zn1B, Zn2, and Zn3, respectively. For the Rv3488–Zn structure, Matthew's calculation (1.89 Å3 Da−1, 35.05% solvent content) indicates two molecules in the asymmetric unit. The RMSD between subunit-A and subunit-B of the Rv3488–Zn is 0.6 Å. The RMSD between subunit-A of Rv3488–Zn and subunit-A of apo-Rv3488 is 0.3 Å. The crystal structure of Rv3488–Zn subunit-A has three Zn2+ ion-binding sites (Zn1A, Zn2, and Zn3). The electron density of each Zn2+ ions is depicted in Figure 6A–D. Two Zn2+ ions (Zn1A and Zn3) are located in the DNA-binding domain and one (Zn2) is located at the junction of the DNA-binding and dimerization domains. Zn1A site is co-ordinated by the side-chain atoms of NE2 of H16, the OE1 of E30 and the NE2 of H34, as shown in Figure 6E,F. The Zn1B site co-ordination residues are also similar to that of Zn1A site in Rv3488–Zn structure. In addition, one oxygen atom of a water molecule co-ordinates the Zn1B site. All the metal-to-ligand distances are summarized in Table 5. According to the CheckMyMetalserver (http://csgid.org/csgid/metal_sites), the co-ordination sphere of this Zn1B site can be described as tetrahedral geometry, which is the most common geometry observed in many other Zn2+ ion-binding proteins [3439]. While the co-ordination sphere of Zn1A site can be described as octahedral.

Cartoon representation of Rv3488–Zn crystal structure showing Zn2+ ion-binding sites.

Figure 5.
Cartoon representation of Rv3488–Zn crystal structure showing Zn2+ ion-binding sites.

The structure of Rv3488–Zn dimer with subunit-A in salmon and subunit-B in blue. The Zn ions are shown as yellow spheres, and metal-binding residues at subunit-A and subunit-B are shown as grey sticks. The secondary structure elements are labeled. The N- and C-terminal ends of Rv3488–Zn are labeled as N-terminal and C-terminal, respectively.

Figure 5.
Cartoon representation of Rv3488–Zn crystal structure showing Zn2+ ion-binding sites.

The structure of Rv3488–Zn dimer with subunit-A in salmon and subunit-B in blue. The Zn ions are shown as yellow spheres, and metal-binding residues at subunit-A and subunit-B are shown as grey sticks. The secondary structure elements are labeled. The N- and C-terminal ends of Rv3488–Zn are labeled as N-terminal and C-terminal, respectively.

The co-ordination of Zn2+ ions in the crystal structure of Rv3488-Zn.

Figure 6.
The co-ordination of Zn2+ ions in the crystal structure of Rv3488-Zn.

(A) The 2Fo − Fc electron density map around Zn2+ countered at 1σ and shown as light blue mesh. Zn1A-binding site at subunit-A. (B) Zn1B-binding site at subunit-B. (C) Zn2 ion-binding site at the interface of subunits-A and B. (D) Zn3 ion-binding site at subunit-A. The zinc and water molecules are represented as yellow and gray spheres, respectively. (E) Structural comparisons at Zn2+ ion-binding residues. Superimposition of the Rv3488 (green) and ZnCl2 soaked Rv3488 (orange) structures showing metal-binding centers. Zn2+ ion-binding sites at subunit-A. (F) Zn2+ ion-binding site at subunit-B. Stick representation for Rv3488–Zn is shown in orange line and for apo-Rv3488 in green line. The interaction between zinc ions and residues of Rv3488–Zn is shown in black dashes.

Figure 6.
The co-ordination of Zn2+ ions in the crystal structure of Rv3488-Zn.

(A) The 2Fo − Fc electron density map around Zn2+ countered at 1σ and shown as light blue mesh. Zn1A-binding site at subunit-A. (B) Zn1B-binding site at subunit-B. (C) Zn2 ion-binding site at the interface of subunits-A and B. (D) Zn3 ion-binding site at subunit-A. The zinc and water molecules are represented as yellow and gray spheres, respectively. (E) Structural comparisons at Zn2+ ion-binding residues. Superimposition of the Rv3488 (green) and ZnCl2 soaked Rv3488 (orange) structures showing metal-binding centers. Zn2+ ion-binding sites at subunit-A. (F) Zn2+ ion-binding site at subunit-B. Stick representation for Rv3488–Zn is shown in orange line and for apo-Rv3488 in green line. The interaction between zinc ions and residues of Rv3488–Zn is shown in black dashes.

Table 5
Geometric parameters of the Zn2+ ions co-ordination at the four Zn2+-binding sites identified in the Rv3488–Zn crystal structure
Zn1A Distances (Å) Zn1B Distances (Å) Zn2 Distances (Å) Zn3 Distances (Å) 
His16 NE2/A 2.29 His16 NE2/B 2.11 His12 NE2/A 2.48 His15 ND1/A 2.02 
Glu30 OE1/A 2.44 Glu30 OE2/B 1.90 His34 ND1/A 2.68   
His34 NE2/A 2.38 His34 NE2/B 1.98 Glu94 OE1/B 1.96   
  W26 2.11 Glu94 OE2/B 2.47   
Zn1A Distances (Å) Zn1B Distances (Å) Zn2 Distances (Å) Zn3 Distances (Å) 
His16 NE2/A 2.29 His16 NE2/B 2.11 His12 NE2/A 2.48 His15 ND1/A 2.02 
Glu30 OE1/A 2.44 Glu30 OE2/B 1.90 His34 ND1/A 2.68   
His34 NE2/A 2.38 His34 NE2/B 1.98 Glu94 OE1/B 1.96   
  W26 2.11 Glu94 OE2/B 2.47   

Zn2 site located between the dimerization and DNA-binding domains of Rv3488–Zn is found to be co-ordinated with the side-chain residues of H12, H34 of subunit-A, and E94 of subunit-B. The co-ordination sphere of Zn2 site can be described as octahedral with the side-chain of E94, which is ligated as a symmetrical bidentate ligand. The Zn3 site is poorly co-ordinated by H15 of subunit-A, which is located in the DNA-binding domain.

Structurally, conformational changes observed in all of the key metal-binding sites carboxylic and amino residues (H16, E30, and H34) in Rv3488–Zn structure are similar to that observed for Rv3488–Cd structure. H12, H15, H16, and H34 are shifted towards the Zn2+ ions bringing the amino groups of the residues closer to Zn2+ ions, thereby allowing a direct interaction with the Zn2+ ions as shown in Figure 6E,F. More specifically, the carboxylic side-chain of E94 is tilted towards Zn2 site for better orientation for the Zn2+ ion binding. The distance between the Zn2+ ions is 6.76 Å (Zn1–Zn2), 6.35 Å (Zn2–Zn3) and 6.67 Å (Zn1–Zn3). Recognition helices α3 in the Rv3488–Zn dimer are approximately separated by 33.64 Å (Cα–Cα distance between G41/subunit-A and G41/subunit-B), while Cα–Cα distance in apo-Rv3488 and Rv3488–Cd are 33.16 and 32.88 Å, respectively.

Characterization of binding of metal ions to Rv3488 by ITC

The Cd2+, Zn2+, Ni2+, and Mn2+ binding to Rv3488 were explored at 25°C, pH 7.5, by ITC. The raw ITC data and integrated areas under each peak versus molar ratio of metal ion to Rv3488 are shown in Figure 7. The binding isotherm curves were best fitted with a two or single set of binding model for the determination of thermodynamic parameters. The parameters used in fitting were the stoichiometries of association (n), association constant (Ka), and enthalpy change (ΔH). The entropy change (ΔS) and Gibbs free energy (ΔG) are calculated using the following equation:

 
formula

where T is the thermodynamic temperature and R is gas constant as shown in Table 6. Calorimetric titrations of Rv3488 with Cd2+ were an exothermic process characterized by binding constant Ka1 = 0.36 ± 0.22 nM−1, n1 = 0.66 ± 0.028, Ka2 = 12 ± 3.5 µM−1, n2 = 0.68 ± 0.047. This is consistent with the qualitative assessment that there are two independent sites for cadmium binding. One tight (site-n1) and one weaker (site-n2) site can be distinguished by a 30-fold difference in their binding affinities. The binding of Zn2+ to Rv3488 was an exothermic process and thermodynamic parameters suggest two binding sites per monomer. The best-fit binding model for Zn2+ binding is found to have a stoichiometry n1 = 1.11 ± 0.95, n2 = 0.96 ± 0.54. The binding of Cd2+ and Zn2+ to H16A and H34A mutants, respectively, revealed that the mutations abolished the cadmium and zinc binding, as shown in Supplementary Figure S5. Similarly, Ni2+ and Mn2+ binding to Rv3488 were also an exothermic process. The ITC results reveal a binding stoichiometry of 0.92 Ni2+/Rv3488 monomer and 0.98 Mn2+/Rv3488 monomer, suggesting that Rv3488 protein binds Ni2+ and Mn2+ approximately in an 1 : 1 molar ratio. Dissociation constant Rv3488 for Cd2+, Zn2+, Ni2+, and Mn2+ are 1.47 × 10−8 M, 3.14 × 10−6 M, 5 × 10−6 M, and 5.59 × 10−8 M, respectively, indicating the formation of tight complexes, as shown in Table 6. Rv3488 binds to Cd2+ with a higher affinity than Mn2+, Zn2+, and Ni2+.

ITC characterization of binding of metal ions to Rv3488.

Figure 7.
ITC characterization of binding of metal ions to Rv3488.

Isothermal calorimetric titration experiments of (A) Rv3488–CdCl2, (B) Rv3488–ZnCl2, (C) Rv3488–MnCl2 and (D) Rv3488–NiCl2. The data points (▪) were obtained by integration of heat signals plotted against the molar ratio of Rv3488 to the respective metal salts in the reaction cell. The solid line represents a calculated curve using the best-fit parameters obtained by a nonlinear least square fit.

Figure 7.
ITC characterization of binding of metal ions to Rv3488.

Isothermal calorimetric titration experiments of (A) Rv3488–CdCl2, (B) Rv3488–ZnCl2, (C) Rv3488–MnCl2 and (D) Rv3488–NiCl2. The data points (▪) were obtained by integration of heat signals plotted against the molar ratio of Rv3488 to the respective metal salts in the reaction cell. The solid line represents a calculated curve using the best-fit parameters obtained by a nonlinear least square fit.

Table 6
Comparative analyses of thermodynamic parameters governing complex formation of Rv3488 with CdCl2, ZnCl2, NiCl2, and MnCl2
Thermodynamic parameters Rv3488–CdCl2 Rv3488–ZnCl2 Rv3488–NiCl2 Rv3488–MnCl2 
n n1: 0.66 ± 0.03
n2: 0.68 ± 0.047 
n1: 1.11 ± 0.95
n2: 0.96 ± 0.54 
0.92 ± 0.045 0.978 ± 0.00 
Ka (M−1Ka1: 3.63 × 108 ± 2.18 × 108
Ka2: 1.27 × 107 ± 3.5 × 106 
Ka1: 9.38 × 104 ± 7.09 × 104
Ka2: 1.08 × 106 ± 3.30 × 106 
2.0 × 105 ± 7.97 × 104 1.79 × 107 ± 7.61 × 106 
Kd (M) Kd1: 2.75 × 10−9 ± 1.65 × 10−9
Kd2: 7.87 × 10−8 ± 2.17 × 10−8 
Kd1: 1.07 × 10−5 ± 8.0 × 10−6
Kd2: 9.2 × 10−7 ± 2.83 × 10−6 
5.00 × 10−6 ± 1.99 × 10−6 5.59 × 10−8 ± 2.37 × 10−8 
K1/2 (M) 1.47 × 10−8 3.14 × 10−6   
ΔH (kcal/mol) ΔH1: −2.15 × 103 ± 0.45 × 103
ΔH2: −1.34 × 104 ± 0.11 × 104 
ΔH1: −0.29 × 104 ± 0.023 × 104
ΔH2: −0.12 × 104 ± 0.04 × 104 
−0.38 × 104 ± 0.025 × 104 −1.10 × 104 ± 0.017 × 104 
ΔS (kcal/mol K) ΔS1: 31.9
ΔS2: −12.4 
ΔS1: 12.9
ΔS2: 23.4 
11.5 −3.94 
ΔG (kcal/mol) ΔG1: −11.67 ± 11.3
ΔG2: −9.68 ± 8.92 
ΔG1: −6.78 ± 6.61
ΔG2: −8.22 ± 8.8 
−7.22 ± 6.68 −9.88 ± 9.38 
Calculated from  
Thermodynamic parameters Rv3488–CdCl2 Rv3488–ZnCl2 Rv3488–NiCl2 Rv3488–MnCl2 
n n1: 0.66 ± 0.03
n2: 0.68 ± 0.047 
n1: 1.11 ± 0.95
n2: 0.96 ± 0.54 
0.92 ± 0.045 0.978 ± 0.00 
Ka (M−1Ka1: 3.63 × 108 ± 2.18 × 108
Ka2: 1.27 × 107 ± 3.5 × 106 
Ka1: 9.38 × 104 ± 7.09 × 104
Ka2: 1.08 × 106 ± 3.30 × 106 
2.0 × 105 ± 7.97 × 104 1.79 × 107 ± 7.61 × 106 
Kd (M) Kd1: 2.75 × 10−9 ± 1.65 × 10−9
Kd2: 7.87 × 10−8 ± 2.17 × 10−8 
Kd1: 1.07 × 10−5 ± 8.0 × 10−6
Kd2: 9.2 × 10−7 ± 2.83 × 10−6 
5.00 × 10−6 ± 1.99 × 10−6 5.59 × 10−8 ± 2.37 × 10−8 
K1/2 (M) 1.47 × 10−8 3.14 × 10−6   
ΔH (kcal/mol) ΔH1: −2.15 × 103 ± 0.45 × 103
ΔH2: −1.34 × 104 ± 0.11 × 104 
ΔH1: −0.29 × 104 ± 0.023 × 104
ΔH2: −0.12 × 104 ± 0.04 × 104 
−0.38 × 104 ± 0.025 × 104 −1.10 × 104 ± 0.017 × 104 
ΔS (kcal/mol K) ΔS1: 31.9
ΔS2: −12.4 
ΔS1: 12.9
ΔS2: 23.4 
11.5 −3.94 
ΔG (kcal/mol) ΔG1: −11.67 ± 11.3
ΔG2: −9.68 ± 8.92 
ΔG1: −6.78 ± 6.61
ΔG2: −8.22 ± 8.8 
−7.22 ± 6.68 −9.88 ± 9.38 
Calculated from  

Electrophoresis mobility shift assay for Rv3488

To confirm the binding of Rv3488 at its own promoter region, a 30-bp sequence from −41-bp to −12-bp upstream of the rv3488 start site was synthesized with 5′-Cy3 label and subjected to EMSA with purified Rv3488 protein. As expected, a clear mobility shift of the DNA–protein complex was observed as shown in Figure 8A. Interestingly, the DNA used for EMSA has an 18-bp palindromic (5′-CAAGCCGATATCGGCTTG-3′) sequence. The presence of this palindromic sequence suggests potential binding site for the Rv3488 regulator. As a control, EMSA was done in the presence of cold DNA probe. Liberation of Cy3-labeled probe was observed, which indicates specific binding of Rv3488 to its own promoter region. A similar 10-bp palindromic sequence, 5′-CCGATATCGG-3′, was found in the promoter region of a putative cationic amino acids transporter protein Rv1999c (Figure 8B). Thus, EMSA was performed using 30-bp unlabeled DNA fragment (from −44 to −15 bp) upstream of the rv1999c start site containing the palindrome sequence, which was incubated with different concentrations of purified Rv3488 protein. The shift of the DNA band was dependent upon the protein concentration. In addition, control experiments were also carried out with random unlabeled DNA as a competitor, which did not affect the shift of Rv3488 bound to its promoter DNA (Figure 8C, lanes 3–6). No mobility shift was observed upon incubation of Rv3488 and labeled random DNA fragments (Figure 8D). The absence of Rv3488 protein-dependent mobility shift in the control with random sequence signifies the lack of nonspecific binding of the protein. Overall, the data indicate that the 30-bp DNA fragment potentially forms the specific binding site for rv3488 promoter region.

Gel images showing EMSA.

Figure 8.
Gel images showing EMSA.

(A) EMSA were performed using 0.1 μM Cy3-labeled 30-bp DNA of rv3488 promoter incubated with different concentrations (0, 0.6, 0.7, 1.2, 2, and 3 µM) of the recombinant Rv3488 protein (lanes 1–6, respectively). With Rv3488 concentration of 3 µM, the binding was competed by the unlabeled-specific 30-bp DNA (concentrations of 0.05, 0.1, 0.2, 0.5, and 1 µM, lanes 7–11, respectively). Free and bound DNA are indicated by arrows. (B) EMSA were performed using 5 µM 30-bp DNA promoter region of rv1999c incubated with increasing concentrations (0, 6, 8, 10, and 15 µM) of Rv3488. (C) Control experiment of EMSA were performed using 0.1 µM Cy3-labeled 30-bp of rv3488 promoter DNA with 3 µM Rv3488, the binding was competed by two different unlabeled random 30-bp DNA (5 and 8 µM, lanes 3–6, respectively), and lane 7 shows that CdCl2 dissociates the Rv3488–rv3488 promoter complex. 2.5 mM CdCl2 was incubated with 20 µM of purified Rv3488 for 15 min before the EMSA experiment. The final Rv3488 protein and 30-bp DNA concentrations used in the assay were 3 and 0.1 µM, respectively. (D) Control experiment of EMSA was performed by using 0.1 µM Cy3-labeled random 30-bp DNA fragment with increasing concentrations of Rv3488 protein (0, 0.6, 0.7, 1.2, 2, and 3 µM, respectively).

Figure 8.
Gel images showing EMSA.

(A) EMSA were performed using 0.1 μM Cy3-labeled 30-bp DNA of rv3488 promoter incubated with different concentrations (0, 0.6, 0.7, 1.2, 2, and 3 µM) of the recombinant Rv3488 protein (lanes 1–6, respectively). With Rv3488 concentration of 3 µM, the binding was competed by the unlabeled-specific 30-bp DNA (concentrations of 0.05, 0.1, 0.2, 0.5, and 1 µM, lanes 7–11, respectively). Free and bound DNA are indicated by arrows. (B) EMSA were performed using 5 µM 30-bp DNA promoter region of rv1999c incubated with increasing concentrations (0, 6, 8, 10, and 15 µM) of Rv3488. (C) Control experiment of EMSA were performed using 0.1 µM Cy3-labeled 30-bp of rv3488 promoter DNA with 3 µM Rv3488, the binding was competed by two different unlabeled random 30-bp DNA (5 and 8 µM, lanes 3–6, respectively), and lane 7 shows that CdCl2 dissociates the Rv3488–rv3488 promoter complex. 2.5 mM CdCl2 was incubated with 20 µM of purified Rv3488 for 15 min before the EMSA experiment. The final Rv3488 protein and 30-bp DNA concentrations used in the assay were 3 and 0.1 µM, respectively. (D) Control experiment of EMSA was performed by using 0.1 µM Cy3-labeled random 30-bp DNA fragment with increasing concentrations of Rv3488 protein (0, 0.6, 0.7, 1.2, 2, and 3 µM, respectively).

EMSA was also performed with recombinant Rv3488 protein and 30-bp Cy3-labeled DNA fragment in the presence of CdCl2. The presence of CdCl2 leads to partial release of the DNA from the protein–DNA complex. The result indicates that the binding of CdCl2 to Rv3488 causes conformation change to the structure, which reduces Rv3488 binding to the promoter probe (Figure 8C, lane 7).

Expression of Rv3488 in M. smegmatis using recombinant MS-pMV361::rv3488

The protein BLAST against MS mc2155 genome did not reveal any protein encoded within the M. smegmatis genome with enough similarity to Rv3488 to be called as its ortholog. The absence of Rv3488 homolog in M. smegmatis prompted us to use the organism as a surrogate model to express Rv3488 for characterization of its biological functions. The chromosomal alignment of regions close to rv3488 for M. tuberculosis H37Rv and M. smegmatis is shown in Figure 9A. To generate the recombinant Rv3488-expressing M. smegmatis strain, the ORF of Rv3488 was amplified from the genomic DNA of M. tuberculosis H37Rv and cloned into pMV361 vector at EcoRI and HindIII sites. The recombinant vector was introduced into electrocompetent M. smegmatis cells. The expression of gene was confirmed by Western blotting. M. smegmatis carrying pMV361::rv3488 was able to express the protein, whereas no corresponding band was detected in vector control (Figure 9B).

Expression of Rv3488 protein in various mycobacteria.

Figure 9.
Expression of Rv3488 protein in various mycobacteria.

(A) position map of Rv3488 in mycobacterial genome (Tuberculist database). Arrows indicate the orientation of open reading frames. Protein functions are colored: orange, intermediary metabolism and respiration; red, conserved hypotheticals; green, virulence, detoxification, and adaptation. (B) Expression of Rv3488 in MS-pMV361::rv3488. SDS–PAGE of the total cell lysate (∼40.0 µg each). From right side: lane 1 — recombinant purified Rv3488 protein, lane 2 — MS-pMV361::rv3488, and lane 3 — MS-pMV361. (C) Expression of Rv3488 in mycobacteria. Immunodetection of Rv3488 in mycobacterial whole cell lysates: lane 1 — MS-pMV361::rv3488; lane 2 — MS-pMV361; lane 3 — BCG; lane 4 — Mtb H37Ra; lane 5 — Mtb H37Rv, and lane 6 — MW Marker.

Figure 9.
Expression of Rv3488 protein in various mycobacteria.

(A) position map of Rv3488 in mycobacterial genome (Tuberculist database). Arrows indicate the orientation of open reading frames. Protein functions are colored: orange, intermediary metabolism and respiration; red, conserved hypotheticals; green, virulence, detoxification, and adaptation. (B) Expression of Rv3488 in MS-pMV361::rv3488. SDS–PAGE of the total cell lysate (∼40.0 µg each). From right side: lane 1 — recombinant purified Rv3488 protein, lane 2 — MS-pMV361::rv3488, and lane 3 — MS-pMV361. (C) Expression of Rv3488 in mycobacteria. Immunodetection of Rv3488 in mycobacterial whole cell lysates: lane 1 — MS-pMV361::rv3488; lane 2 — MS-pMV361; lane 3 — BCG; lane 4 — Mtb H37Ra; lane 5 — Mtb H37Rv, and lane 6 — MW Marker.

In this study, we also evaluated the rv3488 gene expression profiles in different species of mycobacteria. M. tuberculosis H37Rv, M. tuberculosis H37Ra, M. smegmatis mc2155, and M. bovis BCG were grown in MB7H9 media, and the lysates of log-phase cultures were prepared. Whole cell lysates (∼40.0 µg) were resolved on 15% SDS–PAGE. The rv3488 gene expression, as analyzed by Western blotting, is shown in Figure 9C. Immunoblotting of mycobacterial lysates with anti-Rv3488 serum revealed that Rv3488 protein was expressed in BCG and M. tuberculosis H37Rv, whereas no such expression was found in M. smegmatis mc2155. In case of M. tuberculosis H37Ra, a faint band at the corresponding position was observed, but the expression of Rv3488 could not be further ascertained.

M. smegmatis mc2155 cells expressing Rv3488 exhibit enhanced intracellular survival in J774A.1 cells

The overexpression of Rv3488 in M. smegmatis did not affect its growth in culture media under standard growth conditions (Figure 10A). To study the role of Rv3488 protein in intracellular survival of mycobacteria, we infected J774A.1 macrophage cells with MS-pMV361 and MS-pMV361::rv3488, respectively, at a 1 : 10 m.o.i. The CFUs recovered from the cells infected with MS-pMV361::rv3488 at 0 h post-infection were comparable to the vector control, suggesting that the protein does not influence phagocytosis. Whereas, when intracellular replication of bacilli was monitored by estimating the CFU counts at 12 and 24 h post-infection, a significant increase in CFU was noticed in case of MS-pMV361::rv3488 when compared with the vector control at 12 h. Furthermore, a sudden decline with no significant difference between them was noticed at 24 h (Figure 10B). However, in the light of previous studies on dynamics of M. smegmatis growth in J774A.1, the sudden fall in CFU count was not surprising. Overall, the results indicate that the M. tuberculosis Rv3488 influences the intracellular persistence of mycobacteria.

Growth responses of M. smegmatis transformed with pMV361::rv3488.

Figure 10.
Growth responses of M. smegmatis transformed with pMV361::rv3488.

(A) Growth curves of MS-pMV361::rv3488. MS-pMV361 and MS-pMV361::rv3488 recombinants were grown in Sauton's medium and growths were measured at OD600nm. (B) Infection in J774A.1 macrophage cells with M. smegmatis recombinants MS-pMV361 and MS-pMV361::rv3488. Statistical significance based on the difference between the mean intracellular survival of mycobacteria with MS-pMV361::rv3488 and control are indicated as *P < 0.05; **P < 0.01 and ***P < 0.001. (C) M. smegmatis viability in the presence of CdCl2. (D) M. smegmatis viability in the presence of HgCl2. (E) M. smegmatis viability in the presence of ZnCl2. (F) M. smegmatis viability in the presence of CuCl2. For the viability evaluation of MS-pMV361 and MS-pMV361::rv3488, cells grown without heavy metal cations were considered as 100% growth. The data are the representation of three independent experiments. The statistical analysis was done by using Graphpad prism 5.0 software.

Figure 10.
Growth responses of M. smegmatis transformed with pMV361::rv3488.

(A) Growth curves of MS-pMV361::rv3488. MS-pMV361 and MS-pMV361::rv3488 recombinants were grown in Sauton's medium and growths were measured at OD600nm. (B) Infection in J774A.1 macrophage cells with M. smegmatis recombinants MS-pMV361 and MS-pMV361::rv3488. Statistical significance based on the difference between the mean intracellular survival of mycobacteria with MS-pMV361::rv3488 and control are indicated as *P < 0.05; **P < 0.01 and ***P < 0.001. (C) M. smegmatis viability in the presence of CdCl2. (D) M. smegmatis viability in the presence of HgCl2. (E) M. smegmatis viability in the presence of ZnCl2. (F) M. smegmatis viability in the presence of CuCl2. For the viability evaluation of MS-pMV361 and MS-pMV361::rv3488, cells grown without heavy metal cations were considered as 100% growth. The data are the representation of three independent experiments. The statistical analysis was done by using Graphpad prism 5.0 software.

Growth response of MS-pMV361::rv3488 recombinant in the presence of different metal ions

To evaluate the role of M. tuberculosis Rv3488 in imparting tolerance to divalent heavy metal cations, MS-pMV31::rv3488 was grown in the presence of different divalent metal cations at the indicated concentrations. The viability of the cells was tested as described in the Experimental procedures section. The growth of both MS-pMV31::rv3488 and the empty vector control strain, MS-pMV361, was diminished by ∼80–90% in the presence of viable concentrations of Hg2+ (0–3 µm), Cu2+ (0–125 µM), Zn2+ (0–10 mM), and Cd2+ (0–3.2 µM). Noticeably, the survival of MS-pMV361::rv3488 was approximately four-folds when compared with that of the control cells at 0.4 µM cadmium concentration. However, the cell viability of both the strains was diminished by ∼95% in the presence of Cd2+ ion concentration greater than 1.6 µM, as shown in Figure 10C. On the other hand, no significant growth difference was found at viable concentration of Hg2+, or Zn2+ or Cu2+ metal ions (Figure 10D–F). Therefore, the increased tolerance of MS-pMV361::rv3488 to the cadmium ions points to an important role of Rv3488 in mycobacterial survival under conditions of cadmium toxicity.

Discussion

The functional roles of the members of the PadR-s1 and PadR-s2 subfamilies are adequately highlighted in the introduction. Members of the PadR family of transcription factors are critical for bacterial cells to respond to phenolic acids as chemical signals and to convert such signals into changes in gene activity. Rv3488 has been assigned to the PadR-s2 subfamily. While the sequence identity between the members of the PadR-s2 subfamily is modest, the secondary structural elements and particularly the residues involved in binding to the major and minor grooves of ds-DNA align very well among the members (Figure 1), including Rv3488. In DALI search, the Rv3488 structure was found to be closely related to other PadR-s2 family proteins. The binding and/or regulation of PadR-s2 subfamily members by phenolic acids have not been established and it remains a puzzle.

The structure of dimeric Rv3488 displays that each subunit of Rv3488 is composed of a wHTH DNA-binding domain and a C-terminal α4 helix, which is mainly involved in dimerization along with helix α1. Analysis of structural homology of all available crystal structures of PadR-s2 family suggests that the wHTH domain superposes well onto Rv3488, while α1 and α4 helices display significant positional deviations. An overlap of the structures of Rv3488 and six other PadR-s2 subfamily proteins is shown Figure 11. The wing of the DNA-binding domain is formed by two antiparallel β1 and β2 strands. In the crystal structure of B. subtilis PadR–DNA complex, the major groove interaction site consists of residues from the N-terminal regions of the α2 (Y20) and α3 helices (Y42), while residues on the β2 strand (K66, K67) of the wing directly interacted with the minor groove of the ds-DNA [8]. It is interesting that tyrosine Y42 of BsPadR is conserved for all of the proteins shown in Figure 1, including Rv3488 (Y44). Another residue R49 is close and also highly conserved and may be interacting with DNA. In the β2 strand of the wing of Rv3488, residues R68 and Y71 are highly conserved and may interact with the minor groove of DNA. G77, along with G54, has been suggested to have a crucial structural role in stabilizing the conformation of the wing β-strands relative to the DNA recognition helix and dimerization domain. In the BsPadR structure, the effector p-coumaric acid is located in a narrow interdomain pocket between the NTD and CTD. Structural comparison indicated that binding of the effector could disrupt the PadR–DNA interactions through an allosteric mechanism involving translational and rotational changes that dislodge the DNA-binding residues in the α2 and α3 helices and the β2 strand. For such a mechanism to operate in Rv3488 and other PadR-s2 subfamily proteins, there has to be a binding pocket between the regions formed by helices α2 and α3 and helices α1 and α4. In the structure of LmrR, the drugs daunomycin and Hoechst 33342 are bound in a central pore with their aromatic rings sandwiched between two symmetry-related tryptophan residues. This is a very specialized mode of binding and does not seem to be a general mechanism related to the regulation of phenolic acid decarboxylation. Moreover, it is important to mention that the apo- and drug-bound LmrR structures show movement of the DNA-binding helices in such a way that both are not compatible with DNA binding [9]. Unlike the structure of LmrR, Rv3488 structure does not reveal a central pore for binding of effectors.

Superposition of the single subunit structures of PadR family proteins for structural comparison.

Figure 11.
Superposition of the single subunit structures of PadR family proteins for structural comparison.

Structures were taken from the PDB using the indicated accession numbers: Rv3488 (blue, M. tuberculosis), 4EJ0 (sky blue, E. lenta), 3F8B (dark magenta, L. lactis), 5DYM (gray, C. difficile), 4ESB (deep pink, B. cereus strain 14579), 4ESF (salmon, B. cereus strain 10987), and 5H20 (orange, B. fragilis).

Figure 11.
Superposition of the single subunit structures of PadR family proteins for structural comparison.

Structures were taken from the PDB using the indicated accession numbers: Rv3488 (blue, M. tuberculosis), 4EJ0 (sky blue, E. lenta), 3F8B (dark magenta, L. lactis), 5DYM (gray, C. difficile), 4ESB (deep pink, B. cereus strain 14579), 4ESF (salmon, B. cereus strain 10987), and 5H20 (orange, B. fragilis).

The rv3488 gene has previously been suggested to bind to a 59-bp minimal acid-inducible promoter region of lipF [40], which is located 86-bp upstream of the coding region for the rv3488 gene. The rv3488 gene is transcribed in a divergent direction from the nearest upstream gene lipF [22,23]. Rv3488 could potentially be a transcriptional regulator of lipF and also serve to autoregulate itself. Analysis of the rv3488 promoter region (from −41 to −12) has revealed the existence of a perfect 18-bp palindromic sequence 5′-CAAGCCGATATCGGCTTG-3′, which could be the target site for Rv3488 binding. Fascinatingly, a shorter 10-bp palindromic sequence within the 18-bp palindromic sequence was found in the promoter region of rv1999c gene, where it is situated downstream from the −15-bp box near the transcription start site. Our EMSA studies have established that Rv3488 binds to the rv1999c promoter region. Rv1999c is annotated as a probable conserved integral membrane protein of the APC family, possibly involved in the transport of cationic amino acids across the membrane. Therefore, it would be interesting to characterize Rv1999c and its regulation by Rv3488.

Structural comparison of the apo-Rv3488 and Rv3488–Cd (PDB ID: 5ZHC and 5ZI8, respectively) reveals that there are no large conformational changes in the structure between Rv3488–Cd and apo-Rv3488. However, subtle conformational changes in all key metal-binding site residues (H16, E30, and H34) and conserved DNA-binding residues are observed. For E30 and H34, the binding stabilizes one of their alternative conformations identified in apo-Rv3488. E30 is shifted 1.14 Å at CG position towards the cadmium ions site, which brings the carboxyl groups of the residue closer to the metal ion, thus allowing a direct interaction with the cadmium ion. Furthermore, the side-chains of H16 and H34 are tilted towards the metal ion for better orientation for cadmium binding (Figure 4). In the Rv3488–Cd crystal structure, the side-chain of H34 of subunit-A moves away from E94 of subunit-B after cadmium binding, which leads to loss of the salt bridge between H34 of subunit-A and E94 of subunit-B. However, new salt bridges are formed between R33 and E30 of both subunits of Rv3488–Cd structure. Recognition helices α3 in the Rv3488–Cd dimer are separated by 32.88 Å (Cα–Cα distance between G41/subunit-A and G41/subunit-B), which is lower by 0.28 Å from the distance in the apo-Rv3488 dimer (33.16 Å). It is possible that the metal-induced changes in the positioning of DNA recognition helices could be larger than that observed between apo- and metal-bound Rv3488. This is because Rv3488 was crystallized as an apo-protein and the crystals were soaked with metal solutions. Larger changes may be realized if Rv3488–metal complex was performed and then co-crystallized. The flexibility of the two protein systems can be analyzed through the average B-factors. The average B-factors for main chain atoms for the apo-Rv3488, Rv3488–Zn and Rv3488–Cd crystals are plotted for individual residues in Figure 12. The average B-factor plot analysis shows that the DNA-binding domain has higher B-factor compared with dimerization domain for both the crystals, perhaps because the flexibility of the DNA-binding domain may be important in the binding of DNA during transcription regulation. Other significant regions of difference are the loop between α2 and α3 helices, and β1 strand that shows higher average B-factor (Figure 12). On comparing the B-factor of apo-Rv3488 and Rv3488–Cd, it is found that the residues H34 and R37 of the DNA-binding domain show lower B-factors for Rv3488–Cd in comparison with apo-Rv3488. However, such ordering is not observed in the case of Rv3488–Zn structure. Moreover, for this structure, B-factors are significantly higher for the V38–T42 residues stretch.

Comparison of the residue-wise average main chain B-factors for the apo-Rv3488, Rv3488-Cd, and Rv348-Zn crystals.

Figure 12.
Comparison of the residue-wise average main chain B-factors for the apo-Rv3488, Rv3488-Cd, and Rv348-Zn crystals.

Average B-factors for the apo-Rv3488 (blue squares), Rv3488–Cd (red squares), and Rv3488–Zn (green squares) crystal structures.

Figure 12.
Comparison of the residue-wise average main chain B-factors for the apo-Rv3488, Rv3488-Cd, and Rv348-Zn crystals.

Average B-factors for the apo-Rv3488 (blue squares), Rv3488–Cd (red squares), and Rv3488–Zn (green squares) crystal structures.

The primary Cd2+ and Zn2+ co-ordination site in Rv3488 consists of residues H16–E30–H34–H101. This stereochemistry of binding site, although rare, is observed in the structures of astrovirus serine protease (PDB ID: 2W5E), Bacillus anthracis acireductone dioxygenase (PDB ID: 4QGM), and Geobacter sulfurreducens Zn-dependent peptidase (PDB ID: 3C37). In the cadmium-binding metallorepressor, CadC, the regulatory metal-binding site is composed of four cysteine residues, while a cluster of three cysteines has been shown to form the cadmium-binding site in CmtR [41,42]. Several structural zinc-binding sites are composed of His, Asp, and Glu residues. ITC studies have shown that Rv3488 binds sequentially to two Cd2+ ions. Such sequential binding has been observed for zinc in the case of B. subtilis Zur, where it has been suggested that stepwise activation by zinc serves to broaden the physiological response to a wider range of metal concentrations [43]. Streptomyces coelicolor Zur has been shown to have three zinc-binding sites, with one having a structural role, while the other two having functional roles as an on-off switch and a fine-tuner. A combination of the latter two allows graded expression of zinc-responsive genes based on zinc availability in S. coelicolor [44]. E. coli Zur also has two zinc-binding sites and zinc occupancy of both sites is necessary for attaining a DNA-binding conformation [45].

Many metal sensor proteins bind to a fairly broad range of metal ions. It is likely that the proteins may shift their functions depending on the metal type and concentration in their environment. Cadmium is amongst the most toxic metal encountered by bacteria, plant, and the animal cells. Metals such as Cd, Mn, and Zn form a distinct group and it is for these metals that highest binding affinity has been observed for Rv3488. These metals are able to activate the Zur promoter [46] and also the promoter of MntR [47]. Cd2+ does not have a dedicated uptake system and relies on the Mn2+ uptake system for entry into the cells. For M. smegmatis, apparent MIC concentration, at which no cell growth was observed, was in the mM range for Zn2+ and Mn2+, while it was reported as 15 µM for Cd2+. The resting Cd2+ concentration in macrophages is <2 µM, although macrophages can accumulate Cd2+ up to a concentration of 20 µM [48]. Sequestration and efflux are the two prominent mechanisms for handling cadmium toxicity and both these mechanisms require specific genetic regulation. The reduction in DNA binding of Rv3488 in the presence of cadmium is akin to the repressor role of Fur family of proteins in the apo-form and their derepression upon binding to metals. The M. smegmatis viability assays directly indicate that Rv3488 induces tolerance towards sublethal concentrations of Cd2+ ions. This suggests that Rv3488 is possibly involved in the sensing of Cd2+ in the cytosol of mycobacterial cell, in a manner similar to the activity of the CmtR protein [49]. Like the CmtR protein, Rv3488 has higher affinity for Cd2+. Therefore, discriminatory sensing of Cd2+ is wholly correlated with the relative affinity of the protein. However, unlike CmtR, there is no potential P1-type ATPase-encoding gene in the vicinity of the rv3488 operon. Of related interest are the transcriptomics studies, which report that upon either manganese or zinc stress on M. tuberculosis, there is no induction or repression of the rv3488 gene [50,51].

The absence of Rv3488 homolog in M. smegmatis prompted us to use the organism as a surrogate model. The intracellular expression of rv3488 gene in M. smegmatis was confirmed by Western blotting. The expression of Rv3488 in M. smegmatis did not affect its growth. However, Rv3488-expressing M. smegmatis showed enhanced intracellular survival in J774A.1 macrophages up to 12 h. Macrophages employ several strategies for killing of the phagocytosed pathogen, one of which is toxic metal stress (e.g. Cu2+). It is possible that the promiscuous metal-binding ability of Rv3488 aids in the intracellular survival of mycobacteria. However, at 24 h, a comparable killing of bacilli was observed. The dynamics of intracellular killing of M. smegmatis depends on several factors including the cells lines, multiplicity of infection, and cell culture, and the survival dynamics of M. smegmatis in J774A.1 consists of phases of rapid killing and growth [32,33,52]. Therefore, the sudden decrease in CFU at 24 h after an initial growth induction is not surprising.

Overall, our studies establish Rv3488 as a member of the PadR-s2 subfamily, with a biologically relevant dimeric structure comprised of a wHTH NTD and a long C-terminal helix α4, which along with helix α1 is responsible for dimerization. The members of this family have conserved DNA-binding residues but as yet unidentified effectors and effector-binding residues. Rv3488 has been uniquely identified among the PadR family members for being able to bind to metal ions, with highest affinity for cadmium. It binds to an 18-bp palindromic sequence in a region upstream to its own gene, and may, therefore, have autoregulatory function. The DNA and metal binding may correspond to the repression and derepression processes. Through gene regulation, and through metal detoxification, it may aid in enhanced intracellular survival of the pathogen. Further gene knockout and gene expression studies are needed to better understand the functional role of this protein.

Conclusion

We have characterized the structure and function of Rv3488 from M. tuberculosis H37Rv, which is a member of the PadR-s2 subfamily of transcriptional regulators. The structure consists of a dimer with two wHTH motifs, which is common for DNA-binding transcription regulators. Rv3488 displays strong and specific affinity for cadmium ions. The crystal structure of Rv3488–Cd has one Cd2+ ion bound to each subunit of Rv3488, which is octahedrally co-ordinated by side-chain atoms of histidines H16, H34, H101, glutamic acid E30, and water molecules. Rv3488 binds to its promoter DNA, which is a 30-bp DNA sequence located upstream of rv3488 gene from −41 to −12-bp, containing an 18-bp palindromic sequence. In addition to self-regulation, it may be involved in the regulation of lipF, rv1999c, and possibly other genes. Recombinant M. smegmatis strain-expressing M. tuberculosis Rv3488 gene exhibited increased intracellular survival inside the macrophage J774A.1 cells, and more tolerance towards sublethal concentrations of toxic heavy metal Cd2+ ions. Therefore, in addition to transcriptional regulation, Rv3488 may be enhancing the intracellular survival of mycobacterium by countering toxic metal stress.

Abbreviations

     
  • DSC

    differential scanning calorimetry

  •  
  • EMSA

    electrophoretic mobility shift assays

  •  
  • IPTG

    isopropyl-β-d-1-thiogalactopyranoside

  •  
  • ITC

    isothermal titration calorimetry

  •  
  • m.o.i.

    multiplicity of infection

  •  
  • MABA

    microplate alamar blue assay

  •  
  • MarR

    multiple antibiotic resistance

  •  
  • MDR

    multiple drug resistance

  •  
  • Ni2+-NTA

    nickel–nitrilotriacetic acid

  •  
  • Pad

    phenolic acid decarboxylase

  •  
  • PadR

    phenolic acid decarboxylase repressor

  •  
  • PadR-s1

    PadR-like proteins of subfamily 1

  •  
  • PadR-s2

    subfamily 2 PadR-like proteins

  •  
  • RMSD

    root-mean-square deviation

  •  
  • SEC

    size-exclusion chromatography

  •  
  • wHTH

    winged-helix-turn-helix

Author Contribution

M.K. performed the cloning, expression purification, EMSA experiments, and protein crystallization. R.K.P. and B.K.B. performed X-ray data collection and structure determination. S.T. performed ITC and DSC experiments. A.K.M. and K.K.S. contributed to growth, cell viability, and survival experiments. A.A. conceived, designed, and directed the study, analyzed the results, and wrote the manuscript.

Funding

This work was supported by grants from DST 2016/0535 to A.A.

Acknowledgments

We are thankful to the X-ray diffraction facility of the National Institute of Immunology (NII), New Delhi, India for X-ray data collection. M.K. is a recipient of research fellowship from the Council of Scientific and Industrial Research (CSIR), New Delhi, India. M.K. is grateful to AcSIR (Academy of Scientific and Innovative Research) for PhD registration. This is communication number 9745 from CSIR-CDRI.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
The World Health Organization
. (
2018
)
Global Tuberculosis Report
. https://www.who.int/tb/publications/global_report/en/
2
Piddock
,
L.J.V.
(
2006
)
Clinically relevant chromosomally encoded multidrug resistance efflux pumps in bacteria
.
Clin. Microbiol. Rev.
19
,
382
402
3
Agustiandari
,
H.
,
Peeters
,
E.
,
de Wit
,
J.G.
,
Charlier
,
D.
and
Driessen
,
A.J.M.
(
2011
)
LmrR-mediated gene regulation of multidrug resistance in Lactococcus lactis
.
Microbiology
157
,
1519
1530
4
Lomovskaya
,
O.
,
Lewis
,
K.
and
Matin
,
A.
(
1995
)
Emrr is a negative regulator of the Escherichia coli multidrug resistance pump emrAB
.
J. Bacteriol.
177
,
2328
2334
5
Adewoye
,
L.
,
Sutherland
,
A.
,
Srikumar
,
R.
and
Poole
,
K.
(
2002
)
The MexR repressor of the mexAB-oprM multidrug efflux operon in Pseudomonas aeruginosa: characterization of mutations compromising activity
.
J. Bacteriol.
184
,
4308
4312
6
Alekshun
,
M.N.
,
Levy
,
S.B.
,
Mealy
,
T.R.
,
Seaton
,
B.A.
and
Head
,
J.F.
(
2001
)
The crystal structure of MarR, a regulator of multiple antibiotic resistance, at 2.3 Å resolution
.
Nat. Struct. Biol.
8
,
710
714
7
Heravi
,
K.M.
,
Lange
,
J.
,
Watzlawick
,
H.
,
Kalinowski
,
J.
and
Altenbuchner
,
J.
(
2015
)
Transcriptional regulation of the vanillate utilization genes (vanABK operon) of Corynebacterium glutamicum by VanR, a PadR-like repressor
.
J. Bacteriol.
197
,
959
972
8
Park
,
S.C.
,
Kwak
,
Y.M.
,
Song
,
W.S.
,
Hong
,
M.
and
Yoon
,
S.
(
2017
)
Structural basis of effector and operator recognition by the phenolic acid-responsive transcriptional regulator PadR
.
Nucleic Acids Res.
45
,
13080
13093
9
Madoori
,
P.K.
,
Agustiandari
,
H.
,
Driessen
,
A.J.M.
and
Thunnissen
,
A.-M.W.H.
(
2009
)
Structure of the transcriptional regulator LmrR and its mechanism of multidrug recognition
.
EMBO J.
28
,
156
166
10
Fibriansah
,
G.
,
Kovács
,
Á.T
,
Pool
,
T.J.
,
Boonstra
,
M.
,
Kuipers
,
O.P.
and
Thunnissen
,
A.M.W.H.
(
2012
)
Crystal structures of two transcriptional regulators from Bacillus cereus define the conserved structural features of a PadR subfamily
.
PLoS ONE
7
,
e48015
11
Lee
,
C.
,
Kim
,
M.I.
and
Hong
,
M.
(
2017
)
Structural and functional analysis of BF2549, a PadR-like transcription factor from Bacteroides fragilis
.
Biochem. Biophys. Res. Commun.
483
,
264
270
12
Barthelmebs
,
L.
,
Lecomte
,
B.
,
Divies
,
C.
and
Cavin
,
J.F.
(
2000
)
Inducible metabolism of phenolic acids in Pediococcus pentosaceus is encoded by an autoregulated operon which involves a new class of negative transcriptional regulator
.
J. Bacteriol.
182
,
6724
6731
13
Barthelmebs
,
L.
,
Divies
,
C.
and
Cavin
,
J.F.
(
2000
)
Knockout of the p-coumarate decarboxylase gene from Lactobacillus plantarum reveals the existence of two other inducible enzymatic activities involved in phenolic acid metabolism
.
Appl. Environ. Microbiol.
66
,
3368
3375
14
Gury
,
J.
,
Barthelmebs
,
L.
,
Tran
,
N.P.
,
Diviès
,
C.
and
Cavin
,
J.F.
(
2004
)
Cloning, deletion, and characterization of PadR, the transcriptional repressor of the phenolic acid decarboxylase-encoding padA gene of Lactobacillus plantarum
.
Appl. Environ. Microbiol.
70
,
2146
2153
15
Ngoc
,
P.T.
,
Gury
,
J.
,
Dartois
,
V.
,
Nguyen
,
T.K.C.
,
Seraut
,
H.
,
Barthelmebs
,
L.
et al. 
(
2008
)
Phenolic acid-mediated regulation of the padC gene, encoding the phenolic acid decarboxylase of Bacillus subtilisa
.
J. Bacteriol.
190
,
3213
3224
16
Huillet
,
E.
,
Velge
,
P.
,
Vallaeys
,
T.
and
Pardon
,
P.
(
2006
)
Ladr, a new PadR-related transcriptional regulator from Listeria monocytogenes, negatively regulates the expression of the multidrug effux pump MdrL
.
FEMS Microbiol. Lett.
254
,
87
94
17
De Silva
,
R.S.
,
Kovacikova
,
G.
,
Lin
,
W.
,
Taylor
,
R.K.
,
Skorupski
,
K.
and
Kull
,
F.J.
(
2005
)
Crystal structure of the virulence gene activator AphA from Vibrio cholerae reveals it is a novel member of the winged helix transcription factor superfamily
.
J. Biol. Chem.
280
,
13779
13783
18
Florez
,
A.B.
,
Alvarez
,
S.
,
Zabala
,
D.
,
Brana
,
A.F.
,
Salas
,
J.A.
and
Mendez
,
C.
(
2015
)
Transcriptional regulation of mithramycin biosynthesis in Streptomyces argillaceus: dual role as activator and repressor of the PadR-like regulator MtrY
.
Microbiology
161
,
272
284
19
Arita
,
K.
,
Hashimoto
,
H.
,
Igari
,
K.
,
Akaboshi
,
M.
,
Kutsuna
,
S.
,
Sato
,
M.
et al. 
(
2007
)
Structural and biochemical characterization of a cyanobacterium circadian clock-modifier protein
.
J. Biol. Chem.
282
,
1128
1135
20
Isom
,
C.E.
,
Menon
,
S.K.
,
Thomas
,
L.M.
,
West
,
A.H.
,
Richter-Addo
,
G.B.
and
Karr
,
E.A.
(
2016
)
Crystal structure and DNA binding activity of a PadR family transcription regulator from hypervirulent Clostridium difficile R20291
.
BMC Microbiol.
16
,
231
21
Kwak
,
Y.M.
,
Park
,
S.C.
,
Na
,
H.
,
Kang
,
S.G.
,
Lee
,
G.-S.
,
Ko
,
H.-J.
et al. 
(
2018
)
Crystal structure of the VanR transcription factor and the role of its unique α-helix in effector recognition
.
FEBS J
.
285
,
3786
3800
22
Richter
,
L.
and
Saviola
,
B.
(
2009
)
The lipF promoter of Mycobacterium tuberculosis is upregulated specifically by acidic pH but not by other stress conditions
.
Microbiol. Res.
164
,
228
232
23
Richter
,
L.
,
Tai
,
W.
,
Felton
,
J.
and
Saviola
,
B.
(
2007
)
Determination of the minimal acid-inducible promoter region of the lipF gene from Mycobacterium tuberculosis
.
Gene
395
,
22
28
24
Otwinowski
,
Z.
and
Minor
,
W.
(
1997
)
Processing of X-ray diffraction data collected in oscillation mode
.
Methods Enzymol.
276
,
307
326
25
Vagin
,
A.
and
Teplyakov
,
A.
(
1997
)
MOLREP: an automated program for molecular replacement
.
J. Appl. Cryst.
30
,
1022
1025
26
Emsley
,
P.
and
Cowtan
,
K.
(
2004
)
Coot: model-building tools for molecular graphics
.
Acta Crystallogr. D Biol. Crystallogr.
60
,
2126
2132
27
Winn
,
M.D.
,
Ballard
,
C.C.
,
Cowtan
,
K.D.
,
Dodson
,
E.J.
,
Emsley
,
P.
,
Evans
,
P.R.
et al. 
(
2011
)
Overview of the CCP4 suite and current developments
.
Acta Crystallogr. D Biol. Crystallogr.
67
,
235
242
28
Hancock
,
D.C.
and
Reilly
,
N.J.O.
(
2005
)
Production of polyclonal antibodies in rabbits
.
Methods Mol. Biol.
295
,
27
40
PMID:
[PubMed]
29
Goude
,
R.
,
Roberts
,
D.M.
and
Parish
,
T.
(
2015
)
Electroporation of mycobacteria
.
Methods Mol. Biol.
1285
,
117
130
30
Ni
,
D.
,
Xu
,
P.
and
Gallagher
,
S.
(
2016
)
Immunoblotting and immunodetection
.
Curr. Protoc. Mol. Biol.
114
,
10.8.1
10.8.37
31
Singh
,
N.
,
Tiwari
,
S.
,
Srivastava
,
K.K.
and
Siddiqi
,
M.I.
(
2015
)
Identification of novel inhibitors of Mycobacterium tuberculosis PknG using pharmacophore based virtual screening, docking, molecular dynamics simulation, and their biological evaluation
.
J. Chem. Inf. Model.
55
,
1120
1129
32
Anes
,
E.
,
Peyron
,
P.
,
Staali
,
L.
,
Jordao
,
L.
,
Gutierrez
,
M.G.
,
Kress
,
H.
et al. 
(
2006
)
Dynamic life and death interactions between Mycobacterium smegmatis and J774 macrophages
.
Cell. Microbiol.
8
,
939
960
33
Walburger
,
A.
,
Koul
,
A.
,
Ferrari
,
G.
,
Nguyen
,
L.
,
Prescianotto-Baschong
,
C.
,
Huygen
,
K.
et al. 
(
2004
)
Protein kinase G from pathogenic mycobacteria promotes survival within macrophages
.
Science
304
,
1800
1804
34
Simonson
,
T.
and
Calimet
,
N.
(
2002
)
CysxHisy Zn2+ interactions: thiol vs. thiolate coordination
.
Proteins Struct. Funct. Genet.
49
,
37
48
35
Dudev
,
T.
and
Lim
,
C.
(
2003
)
Principles governing Mg, Ca, and Zn binding and selectivity in proteins
.
Chem. Rev.
103
,
773
787
36
Lee
,
Y.M.
and
Lim
,
C.
(
2008
)
Physical basis of structural and catalytic Zn-binding sites in proteins
.
J. Mol. Biol.
379
,
545
553
37
Wang
,
C.
,
Vernon
,
R.
,
Lange
,
O.
,
Tyka
,
M.
and
Baker
,
D.
(
2010
)
Prediction of structures of zinc-binding proteins through explicit modeling of metal coordination geometry
.
Protein Sci.
19
,
494
506
38
Torrance
,
J.W.
,
MacArthur
,
M.W.
and
Thornton
,
J.M.
(
2008
)
Evolution of binding sites for zinc and calcium ions playing structural roles
.
Proteins Struct. Funct. Genet.
71
,
813
830
39
Touw
,
W.G.
,
van Beusekom
,
B.
,
Evers
,
J.M.G.
,
Vriend
,
G.
and
Joosten
,
R.P.
(
2016
)
Validation and correction of Zn–Cys x His y complexes
.
Acta Crystallogr. D Struct. Biol.
72
,
1110
1118
40
Richter
,
L.
and
Saviola
,
B.
(
2012
)
The Mycobacterium tuberculosis DNA binding protein Rv3488 binds to the intergenic region between Rv3488 and lipF
.
Gene Ther. Mol. Biol.
14
,
24
29
http://gtmb.org/volumes/Vol14/4.saviola_et_al_24-29.pdf
41
Banci
,
L.
,
Bertini
,
I.
,
Cantini
,
F.
,
Ciofi-Baffoni
,
S.
,
Cavet
,
J.S.
,
Dennison
,
C.
et al. 
(
2007
)
NMR structural analysis of cadmium sensing by winged helix repressor CmtR
.
J. Biol. Chem.
282
,
30181
30188
42
Cavet
,
J.S.
,
Graham
,
A.I.
,
Meng
,
W.
and
Robinson
,
N.J.
(
2003
)
A cadmium-lead-sensing ArsR-SmtB repressor with novel sensory sites. Complementary metal discrimination by NmtR and CmtR in a common cytosol
.
J. Biol. Chem.
278
,
44560
44566
43
Ma
,
Z.
,
Gabriel
,
S.E.
and
Helmann
,
J.D.
(
2011
)
Sequential binding and sensing of Zn(II) by Bacillus subtilis Zur
.
Nucleic Acids Res.
39
,
9130
9138
44
Shin
,
J.-H.
,
Jung
,
H.J.
,
An
,
Y.J.
,
Cho
,
Y.-B.
,
Cha
,
S.-S.
and
Roe
,
J.-H.
(
2011
)
Graded expression of zinc-responsive genes through two regulatory zinc-binding sites in Zur
.
Proc. Natl Acad. Sci. U.S.A.
108
,
5045
5050
45
Gilston
,
B.A.
,
Wang
,
S.
,
Marcus
,
M.D.
,
Canalizo-Hernández
,
M.A.
,
Swindell
,
E.P.
,
Xue
,
Y.
et al. 
(
2014
)
Structural and mechanistic basis of zinc regulation across the E. coli Zur regulon
.
PLoS Biol.
12
,
e1001987
46
Maciag
,
A.
,
Dainese
,
E.
,
Rodriguez
,
G.M.
,
Milano
,
A.
,
Provvedi
,
R.
,
Pasca
,
M.R.
et al. 
(
2007
)
Global analysis of the Mycobacterium tuberculosis Zur (FurB) regulon
.
J. Bacteriol.
189
,
730
740
47
Schmitt
,
M.P.
(
2002
)
Analysis of a DtxR-like metalloregulatory protein, MntR, from Corynebacterium diphtheriae that controls expression of an ABC metal transporter by an Mn2+-dependent mechanism
.
J. Bacteriol.
184
,
6882
6892
48
Martin
,
P.
,
Fareh
,
M.
,
Poggi
,
M.C.
,
Boulukos
,
K.E.
and
Pognonec
,
P.
(
2006
)
Manganese is highly effective in protecting cells from cadmium intoxication
.
Biochem. Biophys. Res. Commun.
351
,
294
299
49
Wang
,
Y.
,
Hemmingsen
,
L.
and
Giedroc
,
D.P.
(
2005
)
Structural and functional characterization of Mycobacterium tuberculosis CmtR, a PbII/CdII-sensing SmtB/ArsR metalloregulatory repressor
.
Biochemistry
44
,
8976
8988
50
Botella
,
H.
,
Peyron
,
P.
,
Levillain
,
F.
,
Poincloux
,
R.
,
Poquet
,
Y.
,
Brandli
,
I.
et al. 
(
2011
)
Mycobacterial P 1-type ATPases mediate resistance to zinc poisoning in human macrophages
.
Cell Host Microbe.
10
,
248
259
51
Pandey
,
R.
,
Russo
,
R.
,
Ghanny
,
S.
,
Huang
,
X.
,
Helmann
,
J.
and
Rodriguez
,
G.M.
(
2015
)
Mntr(Rv2788): a transcriptional regulator that controls manganese homeostasis in Mycobacterium tuberculosis
.
Mol. Microbiol.
98
,
1168
1183
52
Jordao
,
L.
,
Bleck
,
C.K.E.
,
Mayorga
,
L.
,
Griffiths
,
G.
and
Anes
,
E.
(
2008
)
On the killing of mycobacteria by macrophages
.
Cell. Microbiol.
10
,
529
548