Functional Characterization of Corynebacterium glutamicum Mycothiol S-Conjugate Amidase

The present study focuses on the genetic and biochemical characterization of mycothiol S-conjugate amidase (Mca) of Corynebacterium glutamicum. Recombinant C. glutamicum Mca was heterologously expressed in Escherichia coli and purified to apparent homogeneity. The molecular weight of native Mca protein determined by gel filtration chromatography was 35 kDa, indicating that Mca exists as monomers in the purification condition. Mca showed amidase activity with mycothiol S-conjugate of monobromobimane (MSmB) in vivo while mca mutant lost the ability to cleave MSmB. In addition, Mca showed limited deacetylase activity with N-acetyl-D-glucosamine (GlcNAc) as substrate. Optimum pH for amidase activity was between 7.5 and 8.5, while the highest activity in the presence of Zn2+ confirmed Mca as a zinc metalloprotein. Amino acid residues conserved among Mca family members were located in C. glutamicum Mca and site-directed mutagenesis of these residues indicated that Asp14, Tyr137, His139 and Asp141 were important for activity. The mca deletion mutant showed decreased resistance to antibiotics, alkylating agents, oxidants and heavy metals, and these sensitive phenotypes were recovered in the complementary strain to a great extent. The physiological roles of Mca in resistance to various toxins were further supported by the induced expression of Mca in C. glutamicum under various stress conditions, directly under the control of the stress-responsive extracytoplasmic function-sigma (ECF-σ) factor SigH.


Introduction
Microbes have evolved a number of strategies to face the challenges of changing environments and survive under conditions of stress. One of these strategies is to synthesize low-molecular-weight (LMW) thiols that act as the most important cellular antioxidants and detoxifiers to protect cells from toxic xenobiotics [1]. Eukaryotes and Gram-negative bacteria produce the tripeptide glutathione (GSH; c-L-glutamyl-L-cysteinylglycine) as the LMW thiol redox buffer, whereas some Gram-positive bacteria, such as members of Corynebacterium, Mycobacterium, Rhodococcus and Streptomyces, cannot produce GSH but instead synthesize its functional equivalent, mycothiol (MSH; AcCys-GlcN-Ins) [2,3]. Like GSH, MSH plays a key role in protecting the cell against environmental stresses such as antibiotics, alkylating agents, oxidants, heavy metals and extreme pH [4][5][6]. Therefore, MSH is considered as a general protection agent for improving the robustness and the survival of cells facing environmental stress conditions. Mycothiol S-conjugate amidase (Mca) is a key enzyme involved in MSHdependent detoxification, which was first isolated from the acid-fast bacterium Mycobacterium smegmatis [7]. In detoxification process, mycothiol directly reacts with electrophilic compounds by its thiol moiety, forming MSH S-conjugates (MSRs) [8]. MSR is then cleaved by Mca to release N-acetylcysteine (AcCys) Sconjugate (a mercapturic acid) into the medium and retain glucosaminyl inositol (GlcN-Ins) in the cell to regenerate MSH [8]. Mca has been reported to have extensive substrate specificity towards alkylating agents including S-conjugates of monobromobimane (mBBr), iodoacetamide (IAM) and N-ethylmalemide (NEM), and antibiotics such as cerulenin and rifampin [8]. Any compound, whether electrophilic or non-polar, that contains aromatic rings, a, b-unsaturated ketones, epoxide rings, arene oxides, or quinines structure, is considered capable to conjugate with MSH [9,10]. These chemical structures also exist in some antibiotics. Interestingly, some antibiotic biosynthetic operons, such as erythromycin and lincomycin, also contain Mca homologs [11]. Diverse mercapturic acids of antibiotics have been detected in the broth of some antibiotic-producing actinomycetes [12]. These findings have led to the proposal that these antibiotics and electrophilic compounds can be conjugated to MSH and are detoxified by Mca. Recently, plenty of marine natural products inhibiting Mca have been identified; some of them are lethal to M. smegmatis and play potential roles in chemical synthesis of drugs directed against Mca [13][14][15].
Recently, we have reported that C. glutamicum mutants lacking MSH are more sensitive to some oxidative agents, electrophiles, and antibiotics, but MSH cannot directly remove these substrates [16]. A search of GenBank sequence database revealed that C. glutamicum Mca, encoded by an annotated mca gene, has higher amino acid sequence identities with Mca of M. smegmatis and M. tuberculosis (about 30%). To better understand the role of Mca in xenobiotic detoxification, model organism C. glutamicum was used to provide a certain basis for the study of pathogenic microorganisms. In this study, the potential roles of Mca in the survival of C. glutamicum by coping with multiple stresses and detoxifying toxins were investigated. Besides, the enzyme was characterized in terms of its metal ion requirement, substrate specificity and regulation mechanism, thus contributing to a deeper understanding of the important enzymatic detoxifier and drug target.

Bacterial strains and culture conditions
Bacterial strains used in this study are listed in S1 Table. Escherichia coli and C. glutamicum strains were routinely cultured in Luria-Bertani (LB) broth or on LB plates at 37˚C and 30˚C, respectively. For generation of mutants and maintenance of C. glutamicum, brain heart infusion broth with 0.5 M sorbitol (BHIS) was used. The C. glutamicum RES167 strain was the parent of all derivatives used in this study. In-frame deletions were generated by the method as described previously [17]. Cell growth was monitored by measuring absorbance at 600 nm (A 600 ). Antibiotics were added at the following concentrations: kanamycin, 50 mg ml 21 for E. coli and 25 mg ml 21 for C. glutamicum; nalidixic acid, 40 mg ml 21 for C. glutamicum; chloramphenicol, 20 mg ml 21 for E. coli and 10 mg ml 21 for C. glutamicum.

Plasmid construction
The genes encoding C. glutamicum Mca (NCgl0948) and RNA polymerase sigma factor RpoE (SigH, NCgl0733) were amplified by PCR using C. glutamicum RES167 genomic DNA as template with primers listed in S1 Table. The resulting DNA fragments were digested and afterwards subcloned into similar digested pET28a vectors, obtaining plasmids pET28a-mca and pET28a-sigH, respectively. The plasmid pK18mobsacB-Dmca was used to construct the C. glutamicum mca deletion mutant. The 786 bp upstream PCR product and 810 bp downstream PCR product of mca were amplified using primer pairs DMcaF1/DMcaR1 and DMcaF2/DMca-R2, respectively. In the next step, the upstream and downstream products were digested with BamHI and SalI, respectively, and inserted into BamHI/SalI sites of pK18mobsacB to get pK18mobsacB-Dmca. Plasmids pK18mobsacB-DsigH was constructed in a similar approach using primers listed in S1 Table. To complement the mca mutant, primers CMcaF/CMcaR were used to amplify the mca gene fragment from C. glutamicum RES167 genome. The PCR product of mca was digested with BamHI/EcoRI and inserted into the BamHI/ EcoRI sites of pXMJ19 to produce pXMJ19-mca. The pXMJ19-mca plasmid was transformed into relevant C. glutamicum strains by electroporation. Expression in C. glutamicum was induced by addition of 0.5 mM isopropyl b-D-1-thiogalactopyranoside (IPTG). The lacZ fusion reporter plasmid pK18mobsacB-P mca ::lacZ was constructed by fusion of the mca promoter to the lacZY reporter gene via overlap PCR [18]. In the first round of PCR, a 1000 bp mca promoter DNA fragment and the lacZY DNA fragment were amplified with the primer pairs pMca-F1/pMca-R and lacZY-F/lacZY-R, respectively. These fragments were used as template in the second round of PCR with primers pMca-F and lacZY-R. The resulting PCR fragments were digested with SmaI/PstI and inserted into the suicide vector pK18mobsacB to get pK18mobsacB-P mca ::lacZ fusion construct [18]. Site-directed mutagenesis of Mca was carried out by overlap PCR as described [19]. To replace Asp residue at position 14 with Ala residue (D14A), the mutant mca:D14A DNA segment was amplified by two rounds of PCR. Primer pairs DMcaF1/McaD14AR and McaD14AF/DMcaR2 were used to amplify segments 1 and 2, respectively. The second round of PCR was carried out using EMcaF/ EMcaR as primer pair and segments 1 and 2 as templates to get mca:D14A segment, which contained mutation at the 14 D site of Mca. The mca:D14A DNA fragment was digested and cloned into similar digested pET28a to generate pET28a-mca:D14A plasmid. Similarly, mca:H10A, mca:H12A, mca:D15A, mca:E16A, mca:E43A, mca:D132A, mca:Y137A, mca:H139A, mca:D141A and mca:H142A segments were obtained with primers listed in S1 Table. These DNA fragments were cloned into pET28a to generate corresponding plasmids for expression. The fidelity of all the constructs was confirmed by DNA sequencing (Sangon Biotech, Shanghai, China).

Heterologous expression and purification of recombinant proteins
To express and purify His 6 -tagged proteins, recombinant pET28a plasmids were transformed into the E. coli BL21(DE3) host. Recombinant strains were grown at 37˚C in LB broth to A 600 of 0.4, shifted to 22˚C, induced with 0.4 mM isopropyl b-D-1-thiogalactopyranoside (IPTG) and grown for additional 12 h. Harvested cells were disrupted by sonication and then purified with Ni-NTA His?Bind Resin (Novagen, Madison, WI) based on manufacturer's instructions. Protein samples were run on 15% SDS-PAGE and visualized by Coomassie brilliant blue staining. Purified recombinant proteins were dialyzed in phosphate-buffered saline (PBS) overnight at 4˚C and then stored at 280˚C until used. Protein concentrations were determined using the Bradford Protein Assay Kit (Bio-Rad, Hercules, CA) on the manufacturer's instructions, with bovine serum albumin (BSA) as standard.

Molecular weight determination
The native molecular weight of recombinant His 6 -Mca was determined by gel filtration using a high performance liquid chromatography (HPLC) system equipped with a HiLoad 26/600 Superdex 200 GL column (GE Healthcare, Piscataway, NJ). The column was pre-equilibrated with 50 mM potassium phosphate buffer (pH 7.2) containing 0.15 M NaCl. The elution volumes were used to calculate the K av values for each of the standard protein K av 5(V e -V 0 )/(V t -V 0 ), where V 0 is the void volume of the column, V t is the total volume of the column, and V e is the elution volume of the protein). The resulting K av was drawn against the molecular mass of the standard proteins to plot the standard curve. The data were fitted with a linear equation. The following protein standards were used (GE Healthcare): aprotinin (6.5 kDa), ribonuclease A (13.7 kDa), carbonic anhydrase (29 kDa), ovalbumin (44 kDa), conalbumin (75 kDa), and aldolase (158 kDa).

Deacetylase and amidase activity assays
Deacetylase activity of C. glutamicum Mca was analyzed by assessing the formation of glucosamine (GlcN) using the method of Huang and Hernick [20] with minor modifications. The reaction mixture (350 ml) containing 50 mM HEPES, 50 mM NaCl, 1.0 mM tris(2-carboxyethyl)phosphine (TCEP, pH 7.5) and 2.0 mM N-acetyl-D-glucosamine (GlcNAc) was pre-incubated at 30˚C before the addition of Mca protein (1.0 mM) to initiate the reaction. After various time intervals, reaction aliquots (60 ml) were terminated by the addition of trichloroacetic acid (5%). After centrifugation (13,000 rpm, 5 min), the supernatant (30 ml) was collected, diluted with borate solution (0.75 M, pH 9.0) and labeled with fluorescamine (FSA, 2.3 mM). After 30 min (fluorescence signal remained stable for up to 1 h), the resulted fluorescence was measured (excitation 395 nm, emission 485 nm) using a SpectraMax M5 plate reader (Molecular Devices, Sunnyvale, CA). The observed increase in fluorescence ([FU]/min) was converted into mM min 21 according to the GlcN standard curve. To determine steady-state parameters, enzyme activity was measured at eight different concentrations of GlcNAc (0-5.0 mM), and kinetic parameters k cat , K m , and k cat / K m were obtained by fitting the Michaelis-Menten equation to the initial linear velocities using the curve-fitting program Kaleidagraph (Synergy Software), which also calculates the asymptotic standard errors.
Similarly, amidase activity of C. glutamicum Mca was analyzed by assessing the formation of GlcN at 37˚C using mycothiol S-conjugate of mBBr (MSmB) as substrate [20]. The reaction mixture (100 ml) included 50 mM HEPES (pH 7.5), 50 mM NaCl, 0.1 mM dithiothreitol (DTT) and 1.0 mM MSmB, and 2.0 mM Mca. After 60 min, the reaction was terminated by addition of an equal volume of acetonitrile followed by incubation at 60˚C for 10 min. After centrifugation (13,000 rpm, 5 min), the supernatant (30 ml) containing GlcN was analyzed as described above. MSmB was prepared by derivatization with mBBr through the method described previously [7].

Effects of pH on Mca activity
pH-dependence experiments were conducted on C. glutamicum Mca with 50 mM buffer solutions (MES, pH 6-6.8; MOPS, pH 6.5-7.5; HEPES, pH 7.3-8.8; Nbis(2-hydroxyethyl)-glycine, pH 8-9; borate, pH 9-10; carbonate, pH 10-11) containing 1 mM TCEP. The method was the same as deacetylase and amidase activity assays. Steady-state kinetic parameters K m , k cat , and k cat /K m for protein activity were determined by fitting initial velocities to the Michaelis-Menten equation. The following equation was fitted to the pH rate profile.
where V is the observed rate of the reaction, K is pH-dependent rate constant for GlcNAc and MSmB substrates, and K a and K b are the ionization constants of the acid and base species, respectively [21].

Preparation of metal-free Mca and metal reconstitution
Metal-free Mca and reconstituted Mca with the desired metal ions were prepared as previously described [22]. Briefly, purified protein (100 mM) was added to the solution containing 25 mM Tris, 25 mM diethylene triamine pentaacetic acid, and 10% glycerol, at pH 7.5 and put on ice. After 1 h, the protein solution was dialyzed with buffer (25 mM Tris, 10% glycerol, pH 7.5) at 4˚C. The concentration of residual metal ions was determined by atomic absorption spectrometry (ZEEnit 650P, Analytik Jena, Germany). For reconstitution with metal ion, apo-Mca (10 mM) was added to a stoichiometric concentration of the desired divalent-metal ions (Co 2+ , Fe 2+ , Mn 2+ , Ni 2+ and Zn 2+ ) and put on ice for 30 min. To determine the optimal metal/protein ratio, apo-Mca was incubated with various concentrations of the metal ions (0-25 mM) for 30 min on ice prior to activity assays. These solutions were dialyzed again as mentioned above to remove unbound metal ions and then analyzed by atomic absorption spectrophotometry.

Sensitivity assays for antibiotics, heavy metals, alkylating agents and oxidants
Disk diffusion assays were performed for antibiotics, alkylating agents, and oxidative agents according to Rawat et al. [5]. Briefly, cells were grown to the midlog phase and a lawn of cells was plated onto LB plates. After the paper disks were placed into the plates, various amounts of agents (10 ml) were added to the disk. The disks were allowed to dry and the plates were incubated for 2 to 3 days. For the minimal broth dilution assay, antibiotics were serially diluted (0.56) in LB medium (1 ml) and cells (A 600 51.6) were inoculated. After 1 to 2 days of incubation at 30˚C, the tubes were checked for growth. The bacteriostasis growth curve assay was used to determine the sensitivity of C. glutamicum to heavy metal stress. LB-grown strains (50 ml) were transferred to LB broth (5 ml), with and without addition of various concentrations of heavy metals (Cd 2+ , Ni 2+ , Cr 2+ and Cu 2+ ) and incubated at 30˚C. The cellular growth was monitored turbidimetrically (A 600 ) after 24 h. All assays were performed in triplicate.

MSH purification
MSH was purified from C. glutamicum RES167 with thiopropyl sepharose 6B column followed by Sephadex LH-20 (GE Healthcare, Piscataway, NJ) chromatography as described by Feng et al. [23]. The concentration of purified MSH was measured by thiol-specific fluorescent-labeling HPLC method [24] with GSH as thiol standard. The HPLC system was equipped with an Extend-C18 column (25064.6 mm, Beckman Ultrasphere ODS IP column; Beckman Coulter, Brea, CA) operated with acetic acid-methanol gradient elution at the flow rate of 0.9 ml min 21 . The bimane derivative of MSH was eluted at approximately 15 min in this system.

Analysis of amidase activity with mBBr as substrate in vivo
The in vivo amidase activity of Mca with mBBr as substrate was performed as previously described [8]. Briefly, triplicate samples (10 ml each) of exponentially growing C. glutamicum strains were chilled on ice for 20 min followed by addition of mBBr (0.5 mM in acetonitrile) and further incubated on ice for 30 min. Excess amount of 2-mercaptoethanol (1 mM) was added to scavenge unreacted mBBr. Cells were harvested and extracted with 50% acetonitrile at 60˚C for 10 min. After acidifying with methanesulphonic acid, cell debris was removed by centrifugation and the supernatant was analyzed by HPLC. To determine the amount of acetylcysteinyl bimane (AcCysmB) in the medium, the culture supernatant was analyzed by HPLC system equipped with a Beckman Ultrasphere ODS IP column (4.6625 cm; Beckman Coulter, Brea, CA). MSmB and AcCysmB were eluted with gradients of eluent A (0.25% acetic acid, pH 3.6 with NaOH) and eluent B (methanol). The proportion of buffer in continuous gradients was as follows: 10%, 0-5 min; 10-50%, 5-15 min; and 50-100%, 15-30 min. The MSmB standard (2 mM) was prepared as described [7]. The concentration of bimane derivatives was presented as mmol g 21 dry cell weight.

Chromosomal fusion reporter construction and b-Galactosidase assay
The lacZ fusion reporter plasmid pK18mobsacB-P mca ::lacZ was transformed into the WT(pXMJ19), DsigH(pXMJ19), and DsigH(pXMJ19-sigH) strains by electroporation, and the chromosomal pK18mobsacB-P mca ::lacZ fusion reporter strains were selected by plating on LB-kanamycin plates. The resulted strains were grown in LB medium to an optical density at 600 nm (A 600 ) of 0.6-0.7 and then treated with different toxic agents of various concentrations at 30˚C for 30 min. b-Galactosidase activity was assayed with o-nitrophenyl-b-galactoside as substrate [25]. Data were presented as the means of triplicate experiments with error bars representing the standard deviation.

RNA isolation and quantitative real-time PCR (qRT-PCR) analysis
Total RNA was isolated from exponentially growing WT(pXMJ19), DsigH(pXMJ19) and DsigH(pXMJ19-sigH) strains exposed to different toxic agents of indicated concentrations for 30 min using the RNeasy Mini Kit (Qiagen, Hilden, Germany) along with the DNase I Kit (Sigma-Aldrich, Taufkirchen, Germany). Purified RNA was reverse-transcribed with random 9-mer primers and MLV reverse transcriptase (TaKaRa, Dalian, China). Quantitative RT-PCR analysis was performed using 7500 Fast Real-Time PCR System (Applied Biosystems, Foster City, CA) as described previously [26]. The primers used are listed in S1 Table. The relative abundance of the target mRNAs was quantified based on the cycle threshold value. To standardize the results, the relative abundance of 16 S rRNA was used as an internal standard.
Electrophoretic mobility shift assay (EMSA) EMSA was performed through the method described by Zhang et al. [19] with modifications. Briefly, DNA probes (400 bp P mca fragments) were amplified from the mca promoter region of corresponding pk18mobsacB-P mca ::lacZ reporter vectors using primers Pmca-F2 and Pmca-R (S1 Table). The reaction mixture (10 ml) contained 20 mM Tris-HCl (pH 7.4), 4 mM MgCl 2 , 100 mM NaCl, 1 mM dithiothreitol, 10% (v/v) glycerol, 20 ng DNA probes and 0-3.0 mg of purified His 6 -SigH. After incubation for 30 min at room temperature, the binding reaction mixture was subjected to electrophoresis on 6% native polyacrylamide gel containing 5% glycerol in 0.56TBE buffer, and the DNA probe was detected with SYBR Green (Promega, Fitchburg, WI).

Roles of C. glutamicum mca in resistance to alkylating agents and oxidants
It was shown previously that MSH reacts with alkylating agents, such as 1-chloro-2,4-dinitrobenzene (CDNB), mBBr, IAM, NEM and methylglyoxal (MG), to form conjugates as that for Streptomyces coelicolor Mca [8]. Moreover, MSH takes part in the detoxification of alkylating agents in C. glutamicum, as mutants lacking MSH are more susceptible to these agents than the wild type [16]. Thus, the sensitivity of RES167 wild type and the Dmca mutant to alkylating agents was examined. As shown in Table 1, Dmca was more susceptible to mBBr, IAM, NEM, CDNB and MG than the wild type strain, showing 0.60-, 0.92-, 0.59-,1.89-and 1.1-fold increase in the size of growth inhibition zones, respectively, suggesting that MSH-mBBr, MSH-IAM, MSH-NEM, MSH-CDNB and MSH-MG adducts were formed that are substrates of Mca. This assumption was further confirmed by the observation that the sensitivity phenotypes of the mca mutant was fully recovered in the complementary strain Dmca(pXMJ19-mca) ( Table 1). Rawat et al. [11] demonstrated that MSH-deficient M. smegmatis mutants are more sensitive to redox cycling agents, such as menadione (MD), plumbagin and nitrofurantoin. Similarly, C. glutamicum mutants lacking MSH were more sensitive to some oxidants, including H 2 O 2 , formaldehyde and diamide [16]. To investigate whether C. glutamicum detoxifys oxidants via MSH-dependent Mca pathway, the sensitivity of Dmca to various oxidative agents was determined by disk diffusion assays ( Table 2). Results showed that the Dmca was significantly more sensitive than the wild type strain to MD (p,0.01), but there was no difference in its sensitivity to the low concentration of hydrogen peroxide (H 2 O 2 ), cumene hydrogen peroxide (CHP), formaldehyde and diamide (p.0.05).
However, a slight difference in the sensitivity between the wild type and Dmca was observed at high concentrations of H 2 O 2 , CHP and diamide (Table 2). Importantly, all these sensitivity phenotypes were restored in the complementary strain Dmca(pXMJ19-mca). In contrast, there was no difference in sensitivity to either high or low concentration of DTT between the wild type and Dmca ( Table 2). These results consolidate the previous findings in M. smegmatis [11].

Role of C. glutamicum mca in resistance to antibiotics
It was reported that M. smegmatis Mca is involved in antibiotics resistance by cleaving MSH S-conjugates of rifamycin S [27] and cerulenin [28], resulting in the excretion of mercapturic acid [27,28]. To test whether C. glutamicum Mca is also involved in detoxification of antibiotics, sensitivity test was performed on the RES167 wild type, Dmca and the complementary strain Dmca(pXMJ19-mca) against six types of antibiotics with different chemical structures. First, sensitivity of C. glutamicum strains to Macrolides and b-Lactams was tested. As shown in Table 3, Dmca was more sensitive to rifamycin S than the wild type, as indicated by the larger size of growth inhibition zone (p,0.01). This result was further confirmed with the more sensitive minimal broth dilution assay [27], in which the minimum inhibitory concentration (MIC) of rifamycin S for WT(pXMJ19) was 4-fold higher than that for Dmca(pXMJ19) ( Table 4). These observations indicate that the Mca-dependent detoxification pathway indeed plays a major role in the detoxification of rifamycin S. However, no significant difference was found in the size of growth inhibition zones around erythromycin, lincomycin and spectinomycin (all belong to macrolides as rifamycin S) between WT(pXMJ19), Dmca(pXMJ19) and the complementary strain Dmca(pXMJ19mca) ( Table 3; p.0.05). Similar trend was obtained in the MIC data (Table 4). Similarly, for penicillin of b-Lactams, no significant difference among WT(pXMJ19), Dmca(pXMJ19) and Dmca(pXMJ19-mca) was observed in either the size of growth inhibition zones or MIC values (Tables 3&4).
Recent studies have reported that MSH-deficient mutants are more sensitive to vancomycin of glycopeptides [29,30]. To determine whether this sensitivity depends on Mca, the sensitivity of C. glutamicum strains to vancomycin was tested. A clear difference was detected in the size of growth inhibition zone around vancomycin disk (Table 3), with about 2-4 fold difference in the corresponding MIC values between Dmca(pXMJ19) and WT(pXMJ19)/Dmca(pXMJ19-mca) ( Table 4).
As for aminoglycosides, there was insignificant difference in the size of growth inhibition zones around streptomycin and neomycin disks among WT(pXMJ19), Dmca(pXMJ19) and Dmca(pXMJ19-mca) strains (Table 3). However, in the more sensitive minimal broth dilution assay [27], both WT(pXMJ19) and Dmca(pXMJ19-mca) showed significant higher MIC values than Dmca(pXMJ19) for streptomycin and neomycin but not gentamycin (Table 4). Overall, the results of sensitivity assays supported the hypothesis that Mca plays a role in detoxifying a number of antibiotics in C. glutamicum.

Role of C. glutamicum mca in resistance to heavy metals
Heavy metal ions, including Cd 2+ , Ni 2+ , Cr 2+ and Cu 2+ , markedly inhibited growth of the mca mutant relative to the wild type strain, and the growth inhibition phenotype was complemented in the Dmca(pXMJ19-mca) strain by plasmid-borne mca expression (Fig. 1). These results suggested that Mca is also involved in heavy metal resistance in C. glutamicum.

Detoxification of mBBr by Mca in vivo
Since mBBr easily permeates into cells and subsequently converts intracellular MSH to its bimane derivatives (MSmB) [31], mBBr was chosen as substrate to examine whether C. glutamicum Mca could indeed achieve detoxification by cleaving the amide bond between MSH and toxins in vivo.
As expected, the MSmB substrate was almost completely transformed into AcCysmB, the product of the Mca reaction, in the wild type cells. On the contrary, the amount of MSmB remained unchanged in the Dmca cells. Introduction of the mca gene to the mutant through pXMJ19 vector partially restored its ability to transform MSmB into CysmB (Fig. 2), further confirming the amidase activity of Mca. Consistently, significant higher level of AcCysmB leaked into medium was observed for the wild type WT(pXMJ19) and the complementary strain Dmca(pXMJ19-mca) compared to Dmca (Fig. 2). These results indicate that Mca improves C. glutamicum resistance to various toxins by catalyzing the hydrolysis of the cysteinyl-glucosamine amide bond in mycothiol S-conjugates, as similar as the S. coelicolor Mca [8].

Monomeric structure of Mca
The open reading frame of the putative C. glutamicum mca gene (ncgl0948) was cloned into pET28a vector and expressed in E. coli BL21(DE3). The resulted strain induced by IPTG showed protein over-expression at around 35.0 kDa in SDS-PAGE. After cell sonication and protein separation into fractions by centrifugation, Mca was retained in the soluble fraction. The purified recombinant Mca showed a single band in SDS-PAGE gel (S1A Figure). The oligomerization properties of purified Mca were examined by analytical gel-filtration chromatography (S1B&C Figures). In the gel-filtration chromatogram, a sharp peak appeared at the elution time of 5.5 min (S1C Figure). According to the standard curve (S1B Figure), the native molecular mass of Mca was estimated to be 35 kDa, closer to the value deduced from its amino acid sequence (33 KDa). This result indicates that Mca eluted from the column is entirely monomeric. Because a protein could switch between monomer and multimer when binding to different ligands [32,33], so it doesn't completely exclude the possibility that Mca may also exist as oligomers in vivo upon ligand binding.    specificities with Mca [21]. A recent report demonstrated that mshB null mutant of C. glutamicum can still accumulate a certain amount of MSH [16], indicating that some other enzymes possibly have deacetylase activity to catalyze GlcNAc-Ins in C. glutamicum. Thus, C. glutamicum Mca, having higher identity with MshB, may be a reasonable candidate possessing deacetylase activity (S2 Figure). To test this possibility, FSA-based assay was used to measure the steady-state turnover of GlcNAc by Mca. As reported, FSA reacted with amines of GlcN to form a fluorescent product (excitation 395 nm, emission 485 nm) [20] (S3 Figure). If Mca could cleave GlcNAc to form GlcN, there would be fluorescence produced in the reaction mixture. Just as expected, fluorescence production was observed for GlcNAc treated with Mca and the observed rate of amide bond hydrolysis was 0.87¡0.4 mM min 21 (Fig. 3B) measured with the GlcN standard curve (Fig. 3A).  (Fig. 4D) and was moderately activated upon stoichiometric addition of Ni 2+ (Fig. 4C) and Co 2+ (Fig. 4A). However, no deacetylase and amidase activities of Mca were detected in the presence of Mn 2+ (Fig. 4B) and Fe 2+ (Fig. 4E). These results indicate that Mca is a metalloprotein with Zn 2+ as metal ion cofactor.

pH dependence of Mca-catalyzed reaction
The Mca-catalyzed reaction exhibited a bell-shaped dependence on pH (Fig. 4F), indicating that there are two ionizations of importance to the maximal deacetylase and amidase catalytic activity of Zn 2+ -Mca with the observed pK a values of 6.5 and 9.5, respectively. The optimum pH for deacetylase and amidase activity of C.glutamicum Mca activity was between 7.5 and 8.5, similar to that of M. tuberculosis MshB [22].

Mca activity affected by site-directed mutagenesis
Amino acid sequence alignments showed that C. glutamicum Mca shared conserved residues with other reference MshB and Mca proteins (S2 Figure). These conserved residues include Asp14, Asp15, Asp132, Asp141, Glu16, Glu43, Tyr137, His10, His12, His139 and His142. To date, whether mutation of these sites affects Mca activity has not been elucidated. To this end, a combination of mutagenesis and kinetic experiments was carried out. All mutant proteins purified by the Ni-NTA His?Bind Resin are as stable as the wild type Mca protein (S4 Figure). Mutation of Tyr137 and His139 led to an overall decrease in amidase and  (Tables 6&7). However, there was no effect of Asp15, Asp132, Glu16, Glu43, His10 and His12 on amidase and deacetylase activities. These results indicate that Asp14, Tyr137, His139 and Asp141 are important for Mca activity.

Toxins-induced mca expression and its positive regulation by SigH
Since Mca has been shown to promote survival of C. glutamicum in the presence of various toxins, qRT-PCR and lacZ activity profiling were employed to examine whether mca expression responds to these toxic stress inducers at the transcriptional level. The lacZ activity of P mca ::lacZ chromosomal promoter fusion reporter in the RES167 wild type strain was quantitatively measured in bacterial cells either untreated or treated with different toxic agents of various concentrations (Fig. 5A). Concentrations of toxic agents applied were able to reduce the growth rate but under sub-lethal concentrations (S5 Figure). The level of mca expression was increased by approximately 3.37-, 2.39-, 3.03-, and 2.51fold in the RES167 reporter strain treated with 10 mg ml 21 rifamycin S, 75 mM CdCl 2 , 10 mM MD, and 7.5 mM MG, respectively, as compared to untreated samples (Fig. 5A). Further, expression of the P mca ::lacZ fusion displayed a dose- dependent increase in response to these adverse environmental conditions (Fig. 5A). These results clearly demonstrate that environmental stress induces mca expression, which in turn directly contributes to tolerance of C. glutamicum to these stress conditions. A similar dose-dependent pattern of mca expression in response to toxins rifamycin S, CdCl 2 , MD and MG was also observed in qRT-PCR analysis (Fig. 5B). As SigH, the stress-responsive extracytoplasmic function-sigma (ECF-s) factor, was reported to respond to thiol-oxidative stress and regulate the expression of multiple resistance genes [34,35], we examined whether mca expression was subjected to SigH regulation by measuring the transcription of chromosomal P mca ::lacZ fusions. Significant decrease of lacZ activity was observed for exponentially grown DsigH mutant exposed to 10 mg ml 21 rifamycin S, 75 mM CdCl 2 , 10 mM MD, and 7.5 mM MG for 30 min, as compared to the wild type (Fig. 5A). The reduced mca expression in the DsigH mutant was fully recovered in the complementary strain DsigH(pXMJ19-sigH) either under toxic-inducible or non-inducible conditions (Fig. 5A). SigH-depdendent mca activation was also confirmed by qRT-PCR analysis (Fig. 5B). These data suggest that SigH positively regulates the expression of mca.
To further determine whether the mca gene is directly regulated by SigH, in vitro EMSA assay was performed by the interaction of SigH (His 6 -SigH) with the mca promoter region. Incubation of His 6 -SigH with P mca , a 400 bp PCR fragment amplified from the mca promoter, led to retarded mobility of the probe (Fig. 5C), indicating direct binding of this protein to the mca promoter. Furthermore, the DNA-protein complexes increased in response to more His 6 -SigH used in the reactions. A 400 bp control DNA amplified from the mca coding region did not show detectable SigH binding (Fig. 5C, lane 5). Collectively, these results indicate that SigH activates mca expression by directly binding to the mca promoter.

Discussion
Mca is an important amidase involved in detoxification of the MSH S-conjugates formed by MSH reacting with exogenous substrates, such as alkylating agents, antibiotics, and oxidants [8]. During detoxification, the amidase activity of Mca is mainly the cleavage of an amide bond in MSH moiety of the conjugates to yield AcCysmB and GlcN-Ins [7]. We demonstrated here that C. glutamicum Mca is a zinc metalloprotein that has both deacetylase and amidase activities, though the deacetylation rate of GlcNAc-Ins is lower than the deamination rate of MSmB in vitro. Our findings also revealed that Mca in C. glutamicum plays important roles in detoxification of alkylating agents, oxidants, antibiotics and heavy metals. The physiological roles of Mca in resistance to multiple toxins were further supported by the induced expression of Mca in C. glutamicum under various stress conditions, directly under the control of the stress-responsive extracytoplasmic function-sigma (ECF-s) factor SigH.
C. glutamicum Mca exhibits amidase activity toward MSmB in vitro, the best substrate known for Mca and having the highest amidase activity (Table 5). Consistent with these results, when WT(pXMJ19), Dmca(pXMJ19) and Dmca(pXMJ19-mca) strains were incubated on ice with mBBr, the conjugate MSmB was produced in all three strains. While most of the MSmB conjugate was cleaved to yield AcCysmB in the wild type and Dmca(pXMJ19-mca) complementary strain, the conjugate was hardly converted to AcCysmB in the Dmca mutant (Fig. 2). This in vivo result confirmed that C. glutamicum Mca has amidase activity of cleaving MSH S-conjugates to detoxify some exogenous toxins. Interestingly, C. glutamicum Mca was also found closely related to M. tuberculosis MshB (30% sequence identity) (S2 Figure). MshB possesses GlcNAc-Ins deacetylase activity and transfers the acetyl groups from GlcNAc-Ins to form GlcN-Ins, which is an intermediate in the biosynthesis of MSH pathway [36]. Indeed, C. glutamicum Mca exhibited deacetylase activity with GlcNAc-Ins (Table 5), albeit markedly weaker than the amidase activity. In fact, an mshB null mutant in C. glutamicum, which exhibited similar resistance to many toxic reagents as the wild type, still produces low level of MSH [16]. However, whether Mca is actually involved in MSH synthesis in C. glutamicum by acting as an MshB substitute needs to be investigated in the future.
In the present study, detoxification ability of C. glutamicum Mca to various antibiotics was assessed in vivo by growth inhibition zone and MIC assays. As expected, the mca mutant showed increased sensitivity to some antibiotics of different types, including vancomycin, tetracycline, ciprofloxacin, rifamycin S, streptomycin and neomycin (Tables 3&4), consistent with the finding in Streptomyces coelicolor [8]. Examination of the structure of antibiotics found that many antibiotics have aromatic rings, a, b-unsaturated ketones, epoxide rings, arene oxides and quinones chemical structure, which could conjugate with MSH as they do with GSH [9,10]. Furthermore, mercapturic acids of antibiotics have been found in the broth of some antibiotic-producing actinomycetes [12]. Therefore, the results presented here indicate a possible mechanism that the above-mentioned antibiotics first react with MSH, and then their conjugates are detoxified through Mca amidase. However, the mca mutant was not sensitive to some other antibiotics, such as lincomycin, erythromycin, gentamycin, spectinomycin, and penicillin. The reasons for the mca mutant to lose the sensitivity to these antibiotics may be as follows: the formation of the mycothiol S-conjugate suffices to detoxify the compound and the accumulation of the S-conjugate within the cell have little or no adverse consequence, or other detoxification pathways may be involved.
Some oxidants (e.g. plumbagin, napthoquinone and MD) can also rapidly form conjugates with MSH and the mutants disrupted in the mycothiol biosynthetic pathway are more sensitive to plumbagin, napthoquinone, and MD than the control strains [11]. In this study, the disruption of the mca gene in C. glutamicum resulted in significant increase in its sensitivity to MD. However, the mca mutant was not sensitive to low concentration of H 2 O 2 , CHP or diamide, and only slightly sensitive to very high concentration of H 2 O 2 , CHP, and diamide (Table 2), despite the sensitivity to either low or high concentrations of H 2 O 2 , CHP and diamide exhibited by mutants lacking MSH [16]. Thus, oxidants that can form stable S-conjugates (e.g., MD) may be cleaved by Mca to form mercapturic acids that are exported, while the low concentrations of oxidants (H 2 O 2 , CHP and diamide) acting as redox cycling agents may be mainly cleaned via other antioxidant pathway, and Mca may replenish other antioxidant pathway to work together only when cells are exposed to the high concentrations of peroxide, in line with the results of Steffek et al. [27].
Alkylating agents, including mBBr, IAM, NEM, CDNB and MG, having the maleimide ring structures and directly conjugating with MSH, had been reported previously [11]. In M. tuberculosis and M. smegmatis, a MSH-dependent detoxification pathway has been described, in which the MSH-NEM, MSH-mBBr, and MSH-IAM adducts produced in the cell serve as substrates for an amidase. The amidase catalyzes hydrolytic cleavage of the amide bond and converts these MSH S-conjugates to maleamic acid that is secreted into the medium [11]. Here, the C. glutamicum mca mutant was more susceptible to alkylating agents (including mBBr, IAM, NEM, CDNB and MG; Table 1) than the wild type, in line with the results of Rawat et al. reported for Mycobacterium smegmatis Mca [11]. Because a number of electrophilic compounds also have the maleimide ring structures, the MSH-dependent Mca detoxification pathway may protect the cell against electrophilic assault from such alkylating agents.
Important information arising from this study was that Mca is required to protect cells against heavy metals, including Cd 2+ , Cr 2+ , Ni 2+ and Cu 2+ . Metal ions, such as Cd(II) and As(V), can form conjugates with GSH and MSH of thiols [37,38]. These findings indicate that some heavy metal ions could form MSH Sconjugates that are cleaved by Mca to be detoxified. However, many other heavy metals, including Cd(II), have been shown to promote ROS production and subsequently lead to the elevation of intracellular GSSG, lower the GSH redox ratio [39]. At this time, Mca may also perform alternative mechanism to resist heavy metals-induced oxidative stress, i.e. Mca compensates MSH loss by playing deacetylase activity to make MSH-dependent peroxiredoxins function better.
In conclusion, this study showed that Mca, a SigH-dependent zinc metalloprotein that has both deacetylase and amidase activities, plays important roles in the detoxification of various alkylating agents, oxidants, antibiotics and heavy metals in C. glutamicum. Our insights into the versatile protective roles of Mca in C. glutamicum could be applied to enhance the robustness of this scientifically and commercially important bacterium in the future.  Figure. Growth curves of C. glutamicum in response to sub-lethal concentrations of toxins. C. glutamicum wild type was grown in LB medium to an A 600 of 0.6-0.7 and exposed to different toxic agents of various concentrations. The cultures continued to be incubated for 10 h, and the A 600 was measured in 2 h intervals. doi:10.1371/journal.pone.0115075.s005 (TIF) S1 Table. Bacterial strains, plasmids, and primers used in this study. doi:10.1371/journal.pone.0115075.s006 (DOCX)