Free Radical Biology and Medicine 143 (2019) 366–374
Contents lists available at ScienceDirect
Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed
First evidence of glutathione metabolism in Leptospira interrogans Natalia Sasoni, Danisa M.L. Ferrero, Sergio A. Guerrero, Alberto A. Iglesias, Diego G. Arias
T
∗
Instituto de Agrobiotecnología del Litoral (CONICET-UNL), Facultad de Bioquímica y Ciencias Biológicas (FBCB), Centro Científico Tecnológico CONICET Santa Fe, Colectora Ruta Nacional Nº 168 km 0, Santa Fe, 3000, Argentina
ARTICLE INFO
ABSTRACT
Keywords: Pathogenic Leptospira Glutathione synthesis Thiol Redox metabolism
Background: Glutathione (GSH) plays a role as a main antioxidant metabolite in all eukaryotes and many prokaryotes. Most of the organisms synthesize GSH by a pathway involving two enzymatic reactions, each one consuming one molecule of ATP. In a first step mediated by glutamate-cysteine ligase (GCL), the carboxylate of Lglutamic acid reacts with L-cysteine to form the dipeptide γ-glutamylcysteine (γ-GC). The second step involves the addition of glycine to the C-terminal of γ-GC catalyzed by glutathione synthetase (GS). In many bacteria, such as in the pathogen Leptospira interrogans, the main intracellular thiol has not yet been identified and the presence of GSH is not clear. Methods: We performed the molecular cloning of the genes gshA and gshB from L. interrogans; which respectively code for GCL and GS. After heterologous expression of the cloned genes we recombinantly produced the respective proteins with high degree of purity. These enzymes were exhaustively characterized in their biochemical properties. In addition, we determined the contents of GSH and the activity of related enzymes (and proteins) in cell extracts of the bacterium. Results: We functionally characterized GCL and GS, the two enzymes putatively involved in GSH synthesis in L. interrogans serovar Copenhageni. LinGCL showed higher substrate promiscuity (was active in presence of Lglutamic acid, L-cysteine and ATP, and also with GTP, L-aspartic acid and L-serine in lower proportion) unlike LinGS (which was only active with γ-GC, L-glycine and ATP). LinGCL is significantly inhibited by γ-GC and GSH, the respective intermediate and final product of the synthetic pathway. GSH showed inhibitory effect over LinGS but with a lower potency than LinGCL. Going further, we detected the presence of GSH in L. interrogans cells grown under basal conditions and also determined enzymatic activity of several GSH-dependent/related proteins in cell extracts. Conclusions: and General Significance. Our results contribute with novel insights concerning redox metabolism in L. interrogans, mainly supporting that GSH is part of the antioxidant defense in the bacterium.
1. Introduction
the thiol group of GSH is able to donate reducing equivalents to reactive species and neutralize them. Second, GSH is a substrate of enzymes involved in detoxification of reactive species, such as glutathione-Stransferase (GST) and glutathione peroxidase (GPx) [2]. The usual pathway for GSH synthesis in prokaryotes and eukaryotes takes place from L-glutamic acid (Glu), L-cysteine (Cys) and L-glycine (Gly), with the formation of γ-glutamylcysteine (γ-GC) as an intermediate. The pathway consists of the two steps detailed by reactions (1) and (2), which are respectively catalyzed by glutamate cysteine ligase (GCL, EC 6.3.2.2) and glutathione synthetase (GS, EC 6.3.2.3):
Glutathione (GSH, L-glutamyl-L-cysteinylglycine) is involved in multiple cellular processes as: (i) sulfur storage and transport; (ii) maintenance of structure and function in proteins; and (iii) regulation of the activity of enzymes by disulfide reduction or by glutahionylation (disulfide-mixed formation between GSH and proteins) [1]. Nevertheless, the main role of GSH is related with its action as main antioxidant metabolite in many organisms. Under oxidative stress, GSH contributes to maintain cellular redox homeostasis by two ways. First,
Abbreviations: DTT, dithiothreitol; DTNB, 5,5′-dithiobis(2-nitrobenzoic acid); γ-GC, gamma-glutamylcysteine reduced; bis-γ-GC, gamma-glutamylcysteine disulfide; Grx, glutaredoxin; GR, glutathione reductase; GSH, reduced glutathione; GSSG, glutathione disulfide; GST, glutathione-S-transferase; H2O2, hydrogen peroxide; NADPH, reduced nicotinamide adenine dinucleotide phosphate; NADP+, oxidized nicotinamide adenine dinucleotide phosphate; NADH, reduced nicotinamide adenine dinucleotide; NAD+, oxidized nicotinamide adenine dinucleotide ∗ Corresponding author. Instituto de Agrobiotecnología del Litoral (CONICET-UNL), Centro Científico Tecnológico Santa Fe, Colectora Ruta Nacional Nº168 - km 0, Santa Fe, 3000, Argentina. E-mail address: [emailprotected] (D.G. Arias). https://doi.org/10.1016/j.freeradbiomed.2019.08.028 Received 1 May 2019; Received in revised form 8 July 2019; Accepted 24 August 2019 Available online 26 August 2019 0891-5849/ © 2019 Elsevier Inc. All rights reserved.
Free Radical Biology and Medicine 143 (2019) 366–374
N. Sasoni, et al.
L-Glu + L-Cys + ATP → γ-GC + ADP + Pi
(1)
γ-GC + L-Gly + ATP → GSH + ADP + Pi
(2)
coli TOP10 cells. The sequences of the amplified genes were confirmed by automated sequencing of two different clones (Macrogen). The lingshA and linCOG-DdlA sequences were subcloned into pMAL-C-TEV and pCold-I expression vectors, respectively.
In this pathway, both enzymes require of a divalent cation (mainly Mg2+) and the first reaction is the rate limiting step. In most organisms synthesizing GSH there are two independent genes gshA and gshB, respectively coding for GCL and GS [3]. Still, in a few number of organisms a single gshF gene codes for a bifunctional enzyme [1,4–6]. Conversely, in most Gram-positive bacteria de novo synthesis of GSH is missing; instead, it can be taken from the environment [7]. In this regard, GSH transporters were identified, not only in Gram-positive but also in Gram-negative bacteria [8,9]. Although GSH is the most ubiquitous low molecular mass thiol, some organisms have another sulfur compound, (cysteine, γ-GC, bacillithiol, mycothiol, trypanothione, ovothiol, coenzyme A, or ergothionine) being the key metabolite involved in keeping the cellular redox balance. In addition, in many organisms the thiol acting as the main antioxidant metabolite has not been clearly identified at the present [9]. Among these are bacteria from the Spirochetes phylum [10,11], such as Leptospira interrogans, the causative agent of leptospirosis in humans. Herein, we present the recombinant production and functional characterization of GCL and GS from L. interrogans. Results support that these enzymes would be responsible for GSH synthesis in this pathogenic bacterium. This was reinforced by the identification of enzymatic activities and detection of GSH in cell extracts of the microorganism. To the best of our knowledge this is the first exhaustive analysis of the redox scenario in Leptospira spp., giving evidence on the presence of GSH in these microorganisms.
2.4. Protein expression and purification The constructs [pMAL-C-TEV/LinGCL] or [pCold-I/LinGS] were used to transform E. coli BL21 (DE3) cells. A single colony of E. coli BL21 (DE3) transformed with the appropriate recombinant plasmid was selected. Overnight cultures were diluted 1/100 in 500 ml of Lysogeny broth (LB) medium supplemented with 100 mg ml−1 ampicillin and grown under identical conditions to exponential phase, OD600 ~0.6. The induction was performed with 0.25 mM IPTG at 25 °C for 16 h. After, the cells were harvested at 5000×g at 4 °C for 15 min and stored at −20 °C until the posterior usage. Purification of each recombinant protein was performed using IMAC with a 1 ml HiTrap™ chelating HP column (GE Healthcare). Briefly, the bacterial pellet was resuspended in binding buffer (20 mM Tris–HCl pH 7.5, 400 mM NaCl and 10 mM imidazole) and disrupted by sonication (using a high intensity ultrasonic processor Vibra-cell TM VCX-600; Sonics & Materials Inc). The lysate was centrifuged (10000×g for 30 min) to remove cell debris. The resultant crude extract was loaded onto column that had been equilibrated with binding buffer. After being washed with 10 bead volumes of the same buffer, the recombinant protein was eluted with elution buffer (20 mM Tris–HCl, pH 7.5, 400 mM NaCl, 300 mM imidazole). In addition, LinGCL obtained from the IMAC chromatography (as MBP-fusion protein) was treated with TEV protease (at ratio of 1 μg protease to 100 μg of recombinant protein at 8 °C for 16 h in 100 mM Tris-HCl pH 8.0 and 1 mM DTT). Subsequently, the LinGCL protein was isolated of MBP by affinity chromatography (using 10 ml amylose column, New England Biolabs), according to the manufacturer's instructions. All active fractions containing pure protein were pooled, concentrated and frozen with 20% (v/v) glycerol at −80 °C. Under these conditions, the proteins were stable and remained active for at least 6 months.
2. Materials and methods 2.1. Materials Bacteriological media were purchased in Britania Laboratories and BD Biosciences. All other reagents and chemicals were of the highest quality commercially available.
2.5. Protein methods
2.2. Bacteria and plasmids
Protein electrophoresis was carried out under denaturing conditions (SDS– PAGE) as described elsewhere [13]. Protein concentration was determined following the procedure described by Bradford [14] using bovine serum albumin (BSA) as standard. The molecular mass of recombinant LinGCL and LinGS at their native functional states was determined by gel filtration chromatography using a Superdex 200 column (GE Healthcare) equilibrated with 50 mM HEPES buffer pH 8.0 containing 100 mM NaCl and 0.1 mM EDTA, at a flow rate of 0.2 ml⋅min−1. A calibration curve was constructed by plotting the log10 of the molecular mass vs. the distribution coefficient (Kav) measured from standard proteins (thyroglobulin, 669 kDa; ferritin, 440 kDa; aldolase, 158 kDa; conalbumin, 75 kDa; ovalbumin, 43 kDa; carbonic anhydrase, 29 kDa; ribonuclease A, 13.7 kDa; and aprotinin 6.5 kDa).
Escherichia coli TOP 10 (Invitrogen) and BL21 (DE3) cells were utilized in routine plasmid construction and expression experiments. The vector pGEM-T Easy (Promega) was selected for cloning and sequencing purposes. The expression vectors were pMAL-C-TEV (kindly provided by Dr. Wulf Blankenfeldt from the Helmholtz Center for Infection Research, Germany) and pCold-I (Takara Bio Inc.). Genomic DNA from L. interrogans was obtained using Wizard Genomic DNA Purification Kit (Promega). DNA manipulation, E. coli cultures, and transformations were performed according to standard protocols [12]. 2.3. Molecular cloning The gene encoding sequences for putative LinGCL (lingshA) and LinGS (linCOG-DdlA) proteins were obtained from available information about L. interrogans serovar Copenhageni genome (Microbes online, http://www.microbesonline.org/and L. interrogans genome project, http://bioinfo03.ibi.unicamp.br/leptospira/and LeptoDB, https:// www.leptonet.org.in/). To amplify these genes by PCR we designed the primers LinGCL-Fow (CATATGTTGAAAACAAAAGAACTTACT), LinGCL-Rev (CTCGAGTCAATGATTGCATAGTTTAAC), LinGS-Fow ( CATATGATGCAATCATTGAAATTGAA) and LinGS-Rev (CTCGAGTTAA AAACCAAGTAAGTCTA). The restriction sites are in bold. PCR amplifications were performed under the following conditions: 95 °C for 10 min; 30 cycles of 95 °C for 1 min, 45 °C for 1 min, 72 °C for 1 min, and 72 °C for 10 min. Then, DNA fragments were cloned into the pGEMT Easy vector and the resulting constructions were used to transform E.
2.6. In vitro culture of Leptospira interrogans L. interrogans serovar Copenhageni strain Fiocruz L1-130 cells were grown in EMJH medium (Ellinghausen-McCullough-Johnson-Harris), a BSA-tween 80 medium (pH 7.4) in static culture for 3–5 days at 28 °C [15,16]. The cell density was recorded at 420 nm in an S-26 spectrophotometer (Boeco-Germany). The bacteria number was estimated according the relation established for Louvel et al. [17] where OD420 nm of 0.35 represent 7.5 × 108 bacteria ml−1. Cell extracts were obtained from biomass pools of three biological replicates (obtained from exponential phase cultures of L. interrogans in EMJH medium). The cellular pellet was suspended with 1 ml of 100 mM 367
Free Radical Biology and Medicine 143 (2019) 366–374
N. Sasoni, et al.
potassium phosphate buffer pH 7.0, 2 mM EDTA. Cell lysis was performed through three cycles of freezing (at −80 °C) and thawing (at 37 °C), and it was completed by sonication. The cell lysate was centrifuged at 21000×g for 30 min at 4 °C. The soluble protein-extract (supernatant) obtained was used to enzymatic activity determination, as mentioned above.
coefficient of 9.6 mM−1 cm−1 [22]). The reaction mixture was composed of 100 mM potassium phosphate buffer pH 7.0, 1 mM CDNB, 2 mM EDTA and different concentrations of cell extract (from 0.6 μg μl−1). The reaction started with the addition of 1 mM GSH. One unit (U) of activity is defined as the amount of enzyme catalyzing the formation of 1 μmol of glutathione-conjugate per min.
2.7. Enzyme activity assays
2.8. GSH quantification in bacterial cells
Both GCL and GS enzymatic activities were determined following the ortho-phosphate (Pi) formation [see equations (1) and (2)] by a colorimetric method using Malachite Green, adapted from Ref. [18]. LinGCL activity determination was performed using a reaction mixture containing 100 mM HEPES-NaOH pH 8.0, 1 mM Cys, 10 mM Glu, 10 mM MgCl2 and 1 mM ATP (or GTP), and the appropriate enzyme dilution. For LinGS enzymatic activity the reaction mixture contained 100 mM HEPES-NaOH pH 8.0, 1 mM γ-GC, 200 mM Gly, 10 mM MgCl2 and 1 mM ATP (or GTP), and the appropriate enzyme dilution. In all cases, reaction mixtures were incubated at 37 °C for 10 min. After that, the reactions were stopped by Malachite Green reactive addition. The enzyme dilution was performed in a buffer composed of 0.2 mg ml−1 BSA and 100 mM HEPES-NaOH pH 8.0. Negative controls were performed by omitting the enzyme dilution or amino acid substrates. Alternatively, we evaluated different metabolites as putative inhibitors over enzymatic activities: 1 mM of ADP, cystine, GSSG, γ-GC or bis-γ-GC. GSH was probed in a range 0.1–6 mM. Finally, DTT and H2O2 were employed at different ratios with the enzyme (1:1, 1:100 and 1:1000), incubating for 15 min at 37 °C. The complex formed between the Malachite Green and the Pi released from the enzymatic reaction was measured at 630 nm in a Multiskan Ascent microplate photometer reader (Thermo Scientific). Sodium ortho-phosphate was used as standard for the whole procedure. One unit (U) of activity is defined as the amount of enzyme catalyzing the formation of 1 μmol of Pi per min, under the conditions above described. Saturation curves were performed using different concentrations of one substrate, and maintaining at saturating level the other substrates. Experimental data were plotted as the turnover number (min−1) versus substrate concentration (mM or μM). The kinetic constants were acquired by fitting the data with Michaelis-Menten equation. Kinetic constants were the means of at least three independent sets of data. In inhibition studies, IC50 refers to the inhibitor concentration that gives 50% of the initial activity. The Ki values were acquired by fitting the data with Cheng-Prusoff equations [19]. Glutathione reductase (GR) activity was determined by following NADPH oxidation spectrophotometrically at 340 nm and 30 °C. The reaction mixture contained 100 mM potassium phosphate buffer pH 7.0, 200 μM NADPH, 2 mM EDTA and different concentrations of the cell extract (from 0.6 μg μl−1). The reaction started by the addition of 1 mM GSSG [20]. One unit (U) of activity is defined as the amount of enzyme catalyzing the formation of 1 μmol of NADP+ per min. Glutaredoxin (Grx) activity was evaluated by reduction of 2-hydroxyethyl disulfide (HEDS) following the NADPH oxidation at 340 nm and 30 °C, as described in Ref. [21]. The reaction mixture contained 100 mM potassium phosphate buffer pH 7.0, 2 mM EDTA, 3 mM GSH, 200 μM NADPH, 1 U ml−1 Saccharomyces cerevisiae glutathione reductase and different concentrations of the cell extract (from 0.6 μg μl−1). The reaction was started with the addition of HEDS at a final concentration of 1 mM. One unit (U) of activity is defined as the amount of enzyme catalyzing the formation of 1 μmol of NADP+ per min. To determine glutathione-S-transferase (GST) activity, we monitored the increase in absorbance at 340 nm and 30 °C. This increase in absorbance results from the conjugation of the thiol group of GSH with 1-chloro-2,4-dinitrobenzene (CDNB), and the amount of the adduct is directly proportional to the absorbance recorded (molar extinction
L. interrogans serovar Copenhageni cells were grown in EMJH media (three biological replicates of 50 ml each) to exponential (~5 days) or stationary phase (~10 days). The cells (~5.1010) were harvested by centrifugation and washed with PBS buffer (three times). After that, the cells were resuspended in 1 ml of 50 mM Tris-HCl pH 8.0 and then, it lysed by sonication. The cell suspension was centrifuged at 21000×g for 15 min and 4 °C. The proteins present in the supernatant were precipitated with 80% cold acetone (in a relation 3:1) for 10 min. The sample was centrifuged under similar conditions as before and the new supernatant was treated with 10 mM H2O2 for 1 h at 30 °C. The resultant metabolites extract was lyophilized in Thermo Savant SPD121P-115 speed-vac concentrator. Finally, the lyophilized material was dissolved in a minimum volume of 50 mM Tris-HCl pH 8.0, to quantify the GSH content. E. coli DH5α cells processed in identical conditions were used as a positive biological control, after reports on the presence of GSH in this bacterium [23]. To refer the bacterial total GSH content for L. interrogans we used the relation OD420 nm of 0.35 represent 7.5 × 108 bacteria ml−1 [15,16], while for E. coli it was used the relation OD630 8 −1 [12]. nm of 1 represent 7 × 10 bacteria ml Total GSH quantification was performed by monitoring the DTNB reduction spectrophotometrically at 405 nm (ε = 13.6 mM−1 cm−1). The reaction mixture (50 μl final volume) contained 100 mM potassium phosphate buffer, pH 7.0; 0.05 U ml−1 S. cerevisiae GR; 2 mM EDTA; 1 mM DTNB; 200 M NADPH and 10 μl of each free-protein bacterial extracts. For measure GSH concentration in biological samples, a calibration curve was constructed (vi vs [GSSG], where vi is the enzyme initial velocity), considering that each mol of GSSG generates 2 mol of GSH. 3. Results 3.1. First evidence of GSH metabolism in L. interrogans Until now, the occurrence of any low molecular mass thiol (including the potential presence of GSH) in Treponema pallidum and Leptospira spp. remains unsolved. To advance in the subject, we explored for the putative existence of GSH metabolism in L. interrogans, performing assays for activity of key related enzymes on crude extracts obtained from the bacterium. As shown in Fig. 1, we were able to assay activity of many classic GSH-dependent enzymes and related proteins; such as GR, GST, Grx, as well as the two enzymes (GCL and GS) involved in GSH-synthesis. It is worthy of mention that the low level of GST activity (~0.05 mU mg−1) detected is compatible with the fact that CDNB is a substrate for a specific type of GST (soluble -type), and that it is used with different efficiency by different GSTs, either from prokaryotic or eukaryotic organisms [24,25]. Materials depicted in Fig. 1 support, a priori, a potential presence of GSH-dependent metabolism in L. interrogans, which would be in agreement with reported data on the genomic project of this bacterium (http://meta.microbesonline.org/). This background prompted us to further explore in the subject by: (i) detecting the specific redox compound, and (ii) cloning the related genes to produce enzymes putatively involved in the GSH synthetic pathway. In order to confirm the existence of GSH in L. interrogans, we evaluated its content in cell extracts from the bacterium. The quantification was performed using a calibration curve ([GSH] from 0.3 to 3 μM, Supplemental data - Fig. 1), following the DTNB reduction of its oxidized form GSSG, as detailed in 368
Free Radical Biology and Medicine 143 (2019) 366–374
N. Sasoni, et al.
These genes encode for proteins of 520 (LinGCL) and 343 (LinGS) amino acids, with theoretical molecular masses of 61 kDa and 39 kDa, respectively. The sequence coding for LinGCL was sub-cloned into pMALC-TEV, while the LinGS gene was sub-cloned into the pCold-I vector. LinGCL was produced as a fusion product His6-MBP-protein, whereas LinGS was produced as a fusion protein containing an N-terminus His6tag. Both recombinant proteins were purified by IMAC from the soluble fraction obtained after disruption of E. coli BL21 (DE3) cells transformed with the respective plasmid, to reach a purity > 95% according to SDS-PAGE (Fig. 2A). The purified LinGCL eluted on gel filtration chromatography as a monomeric protein of 61 kDa; whilst LinGS exhibited the elution profile of a dimeric protein of 82 kDa (Fig. 2B). Fig. 1. Detection of GSH-dependent enzyme activities in Leptospira interrogans. Soluble protein-extract from L. interrogans in growth exponential phase was analyzed. Glutathione reductase (GR), glutaredoxin (Grx), and glutathione Stransferase (GST) were determined at 30 °C and pH 7.0. The glutamyl-cysteine ligase (GCL) and glutathione synthetase (GS) activities were determined at 37 °C and pH 8.0.
3.3. Kinetic properties of LinGCL and LinGS, and in vitro GSH synthesis The recombinant proteins, LinGCL and LinGS, were first evaluated for activity in presence of their respective canonical substrates: L-Glu, LCys and ATP, or γ-GC, L-Gly and ATP. Under these reaction conditions both enzymes were functional exhibiting specific activities of 1.2 ± 0.2 U mg−1 (LinGCL) and 240 ± 10 U mg−1 (LinGS), which are in agreement with values reported for homologous enzymes characterized from other organisms [26,27]. In addition, the activity of LinGCL was higher at pH 7.5–8.5; while LinGS showed more active at pH 8.0 (Supplemental data - Fig. 2), all results that are similar to data reported elsewhere [28]. Table 2 illustrates about the steady-state parameters of LinGCL for L-Glu, L-Cys and ATP (see also Supplemental data - Fig. 3). The enzyme exhibited similar apparent affinity for L-Cys and ATP, whereas the calculated Kmapp for L-Glu was one order of magnitude higher than those corresponding to the other two canonical substrates. Similar values were reported for GCL from other organisms, both eukaryotic and prokaryotic, including the enzyme from rat, human [29], and E. coli [30]. Based on the previous studies [30,31] and in order to make a comparative analysis, we assayed putative alternative substrates for
Material and Methods. Although with values lower than in E. coli and other bacteria, Table 1 shows that we were able to detect significant levels of GSH in L. interrogans extracts. The obtained results suggest that there are no differences in the GSH content depending on the culture phase of this bacterium. Data depicted in Fig. 1 and Table 1 suggests that GSH would be a key redox metabolite in the antioxidant response in L. interrogans. 3.2. Purification of recombinant proteins that are involved in the pathway for GSH synthesis in L. interrogans We identified sequences putatively encoding for LinGCL (lingshA, LIC11812) and LinGS (linCOG-DdlA, LIC11811) in the genome project of L. interrogans (Microbes online, http://www.microbesonline.org/).
Table 1 Total GSH content in bacteria cells. The free-protein cellular extracts (from three independent biological samples) were treated with 10 mM H2O2 for 1 h at 30 °C. This cell suspension was lyophilized and then resuspended in 50 mM Tris-HCl pH 8.0. The total GSH quantify was performed with GR-dependent DTNB reduction assay (see Material and methods). E. coli cells were used as positive biological control. Bacterium
GSH content
Reference
Leptospira interrogans
60 ± 4 fmol (107 bacteria)−1 (exponential phase) 87 ± 9 fmol (107 bacteria)−1 (stationary phase) 1300 ± 80 fmol (107 bacteria)−1 (exponential phase) 27 ± 2 μmol g−1 (dry weight) 1.6 ± 0.2 μmol g−1 (dry weight) 0.5 ± 0.2 μmol g−1 (dry weight)
This work This work This work [52] [52] [52]
Escherichia coli Pseudomonas fluorescens Staphylococcus aureus
Fig. 2. Purified recombinant LinGCL and LinGS. (A) SDS-PAGE (12%) was loaded with 2 μg of respective recombinant protein and stained with Coomassie Brillant blue. Lane 1, LinGS purified by IMAC; lane 2, LinGCL purified after IMAC, TEV-protease cleavage and amylose affinity chromatography; and lane 3, molecular mass marker. (B) Native molecular mass determination of purified LinGCL (○) or LinGS (Δ) by gel filtration chromatography using Superdex G200 columm. The column was calibrated with commercial molecular mass markers (□).
369
Free Radical Biology and Medicine 143 (2019) 366–374
N. Sasoni, et al.
Table 2 Apparent kinetic parameters for recombinant LinGCL and LinGS. The measurements were performed at pH 8.0 and 37 °C. Substrate
Km
LinGCL
1 mM L-Cys 10 mM L-Glu
ATP
0.26 ± 0.06
120 ± 10
7.9⋅103
1 mM ATP 10 mM L-Glu
L-Cys
0.11 ± 0.03
123 ± 9
1.9⋅104
1 mM ATP 1 mM L-Cys
L-Glu
3.3 ± 0.5
170 ± 10
8.5⋅102
1 mM L-Cys 10 mM L-Glu
GTP
0.22 ± 0.07
57 ± 6
4.3⋅103
1 mM GTP 10 mM L-Glu
L-Cys
0.038 ± 0.004
36.9 ± 0.7
9.7⋅105
1 mM GTP 1 mM L-Cys
L-Glu
0.9 ± 0.1
42 ± 2
7.8⋅102
1 mM γ-GC 1 mM ATP
L-Gly
27 ± 5
9600 ± 500
5.9⋅103
200 mM L-Gly 1 mM ATP
γ-GC
0.26 ± 0.02
10500 ± 300
6.7⋅105
200 mM L-Gly 1 mM γ-GC
ATP
0.26 ± 0.06
12200 ± 900
7.9⋅105
LinGS
LinGCL. In this way, the activity of the enzyme was evaluated using LAsp (as L-Glu analogue), L-Ser and L-cystine (instead of L-Cys) or GTP (in ATP replacement). The values of activity thus obtained were expressed as relative respect to those exhibited by the enzyme assayed with the canonical substrates (L-Glu, L-Cys, ATP and Mg2+). In presence of GTP the enzyme activity (and Vmapp values) was 50% lower than with ATP despite the apparent affinity of LinGCL for GTP was very similar to ATP (Table 2 and Supplemental data - Fig. 3). Furthermore, when GTP is the nucleotide substrate the Kmapp values determined for L-Cys and L-Glu decreased by about 3-fold with respect to the values determined in the presence of ATP (Table 2 and Supplemental data Fig. 3). In addition, the analogues to the amino acid canonical substrates were tested in the presence of both nucleotides. As seen in Fig. 3, LinGCL has the ability to use L-Asp, L-Ser and L-cystine (all with less capacity) as alternative substrates. Subsequently, LinGCL was evaluated as a putative bifunctional
kcat
kcat Km−1 (M−1 s−1)
Co-substrate
app
(mM)
(min−1)
Enzyme
app
enzyme (both GCL and GS activities, the latter with γ-GC and L-Gly). Nevertheless, recombinant LinGCL did not exhibit classic GS activity under our evaluated conditions. Instead, the enzyme showed a minor activity when the reaction mixture contained γ-GC, L-Glu and ATP (or GTP). The lack of classic GS activity correlates with the absence of a DAla-D-Ala ligase domain (ATP-Grasp) at C-terminus (present in modular enzymes) [5,32]. The ATP-Grasp domain is present in bifunctional enzymes (that exhibit both GCL and GS activities) such as those found in Gram-positive organisms, among them Streptococcus agalactiae, Streptococcus thermophilus, Listeria monocytogenes and Pasteurella multocida [5,6,32,33]. The behavior of the LinGCL was similar to that reported for the GCL of Proteus mirabilis [34] where the enzyme exhibited a slight promiscuity of substrates. We also evaluated the LinGS promiscuity for the use of alternative substrates, specifically L-Glu or L-Ser (instead of L-Gly), L-Cys (as analogue of γ-GC), and GTP (replacing ATP). Otherwise, we evaluated the ability of the enzyme to catalyze GSH synthesis from each single amino acid (L-Glu, L-Cys and L-Gly) and ATP. Our results showed that the enzyme exhibited activity only with the canonical substrates (see kinetic parameters in Table 2 and Supplemental data - Fig. 4), being inactive in presence of any of the alternative substrates (data not shown). These results are in agreement with previous reports on the properties of GS from other organisms [35]. We determined that the products generated in each of the steps of GSH synthesis (both γ-GC, GSH and ADP) behaved as reversible inhibitors of LinGCL (Fig. 4A–C). Interestingly, ADP and Pi (up to 10 mM) not generated significant inhibitory effects on LinGCL activity in the assayed reaction conditions. Conversely, γ-GC exhibited a linear competitive effect with respect to L-Cys and L-Glu with Ki values of 30 ± 6 μM and 140 ± 40 μM, respectively; but the dipeptide did not generate inhibition respect to ATP. On the other hand, GSH produced: (i) a linear mixed inhibitory effect with respect to ATP (Kic of 1200 ± 200 μM and Kiu of 3300 ± 300 μM, Fig. 4A); (ii) a linear competitive inhibition with respect to L-Glu (Ki of 180 ± 50 μM, Fig. 4B); and (iii) a linear non-competitive inhibition regarding to L-Cys (Ki of 1800 ± 400 μM, Fig. 4C). Similar inhibition profiles were observed for GCL form Streptococcus agalactiae [32]. Though, bis-γ-GC and GSSG caused no effect on the enzyme activity (Supplemental data -
Fig. 3. Evaluation of LinGCL substrates promiscuity. The assays were performed at pH 8.0 and 37 °C, in the presence of 1 mM of L-Glu, L-Cys, L-Ser, L-Asp, Lcystine or γ-GC, and 1 mM of GTP or ATP, and 10 mM MgCl2. The activities were informed as relative activities to the LinGCL activity with L-Glu, L-Cys and ATP. 370
Free Radical Biology and Medicine 143 (2019) 366–374
N. Sasoni, et al.
Fig. 4. GSH inhibition on LinGCL or LinGS activity. The LinGCL and LinGS activities were measured in presence of 0.5–6 mM GSH or 0.5–40 mM GSH, respectively, at pH 8.0 and 37 °C. In each reaction, one substrate was used at variable levels (it shows in each plot), while the co-substrates were at fixed and saturating concentrations. Ki values were determined from IC50 plots vs variable substrate concentration using Cheng-Prusoff equations [19].
Fig. 5). Finally, we observed a slight inhibitory effect on GCL activity in the presence of NADH, NAD+, NADPH and NADP+ (at 200 μM, Supplemental data - Fig. 6). All inhibition behaviors were similar to those found for the Xenopus laevis GCL in the presence of these coenzymes [36]. We also evaluated if the reaction products of LinGS (ADP and GSH) have any inhibitory effect on the enzyme activity. In this way, ADP and Pi showed a very slight inhibitory effect with an estimated Ki value higher than 10 mM respect to γ-GC, L-Gly or ATP (data no shown). Considering GSH, it produced a reversible linear inhibition effect of the type: (i) mixed respect to ATP (Kic of 1000 ± 200 μM and Kiu of 10000 ± 1000 μM, Fig. 4D); (ii) noncompetitive regarding L-Gly (Ki of 9100 ± 900 μM, Fig. 4E); and (iii) competitive respect to γ-GC (Ki of 3200 ± 400 μM, Fig. 4F). In further studies, we found that neither DTT nor H2O2 exerted an effect on the activity of LinGCL or LinGS when analyzed as putative reversible or irreversible inhibitors (data not shown). In addition, no shift in the electrophoretic mobility of the enzymes was observed under non-reducing condition (without sample treatment with DTT) respect to
reducing condition (with sample treatment with DTT) in SDS-PAGE analysis (data not shown). These results suggest the lack of formation of intra and/or intermolecular disulfide bonds when the protein is subjected to treatments with redox agents. To reinforce the hypothesis that LinGCL and LinGS can operate in combination to synthesize GSH, we reconstituted such a synthetic pathway. Formation of GSH was detected as described in Material and Methods, using a glutathione reductase-dependent enzymatic recycling method [5,29]. Results in Fig. 5 confirm that LinGCL has not GS activity. Also shown is that LinGS alone is not able to synthetize GSH from L-Cys, L-Gly and L-Glu and ATP and that both enzymes are necessary for such a purpose. These results indicate that both enzymes are operative together, supporting the potential existence of the pathway for GSH synthesis in L. interrogans. 4. Discussion The coordinated action of proteins and antioxidant metabolites, such as low molecular mass thiols, is critical to cope with potential 371
Free Radical Biology and Medicine 143 (2019) 366–374
N. Sasoni, et al.
most Gram positive bacteria [7] as well as in some genera of spirochetes [10,11]. For these latter organisms, it has been described that in Treponema spp. and Borrelia spp. coenzyme A (CoASH) is the primary alternative low mass thiol acting as GSH [11]. Still, no information is available about which is the main low molecular mass thiol that is functional in other spirochetes such as Leptospira spp. As stated above, we evaluated the potential existence of a GSHdependent metabolism in L. interrogans, first detecting the activity of various GSH-dependent enzymes: GCL, GS, GR, Grx and GST in cell extracts. Except for GR, all the detected activities agree with the available genetic information from the microorganism. Indeed, to find that L. interrogans serovar Copenhageni Genome Project (http:// bioinfo03.ibi.unicamp.br/leptospira/) there are already annotated sequences coding for three isoforms of GST (LIC1298, LIC11363 and LIC10807), one dithiol-Grx (LIC11810), one monothiol-Grx (LIC11809), one GCL (LIC11812), and two glutathione peroxidases (LIC12648, butE; LIC13442, gpo). Conversely, the GR activity determined in the crude extracts cannot be paralleled with genomic data, where no gene coding for GR is identified. An explanation for this could be that measurement of GR activity is not because of the existence of a specific protein but due to the action of the thioredoxin system (thioredoxin reductase plus thioredoxin) in the crude extract. The ability of L. interrogans thioredoxin system for in vitro bis-γ-GC or GSSG reduction was demonstrated by us previously [39]. A summary of the above indicates that it was possible to measure enzyme activities and also identifies genes putatively arranging a complete GSH-metabolism in L. interrogans, supporting that this metabolite could constitute a key redox component of this pathogen bacterium. As above stated, the genomic data of L. interrogans give no information about a gene encoding the classic GS (the gshB gene); this being contrary to that found in Leptospira biflexa, which possesses a gene coding for this enzyme (Microbes online, http://www. microbesonline.org/). On the other hand, it was reported that some bacteria contain the gshF gen [4], which encodes a bifunctional enzyme (with GCL and GS activity) [4,6,32,33]. The N-terminal domain of this protein exhibits moderate identity with the GCL proteins. The C-terminal protein portion, contains an ATP-Grasp domain (homologous to the
Fig. 5. In vitro GSH enzymatic synthesis by LinGCL and LinGS. Each reaction mixture (detailed in the table) was incubated for 1 h at 30 °C in 100 mM MOPSNaOH pH 8.0, and then it was stopped by incubation at 60 °C for 10 min. The GSH content was determined by following the enzymatic GR-dependent DTNB reduction at 405 nm (as is described in Materials and methods).
damages derived from the generation of ROS and RNS in different cells. GSH is the most abundant thiol in nearly all eukaryotes as well as in Gram negative bacteria and cyanobacteria organisms [37]. Involvement of GSH as a reducer in different processes generates its oxidized form GSSG, which is reduced back to the former reduced state by the reaction mediated by glutathione reductase (GR) in most of the organisms having these redox components [38]. Available bioinformatic data state that the classic GSH biosynthetic machinery (GCL and GS) is lacking in
Fig. 6. Schematic summary of putative antioxidant GSH-dependent metabolism in L. interrogans. In grey box with white edges are the enzymes characterized in this work. The dashed line highlighted the inhibition by γ-GC and GSH, over LinGCL activity. The spot line showed a least inhibitory effect of ADP and GSH, over LinGS activity. The recycle from GSSG and bis-γ-GC, to GSH and γ-GC, is catalyzed by the Trxsystem. The last system also has the ability to maintain the cysteine pool when the cystine accumulates. In white boxs with dashed edges are the enzymes that have been identified in the bacterium genome, but its function has not been studied yet. Abbreviations used in this figure: 1CGrx, monothiolGrx; 2CGrx, dithiol-Grx; AhpC, alkyl hydroperoxide reductase subunit C (2Cys peroxiredoxin); γ-GC, γglutamylcysteine; bis-γ-GC, γ-glutamylcysteine disulfide; γ-GCSNO, S-nitroso-γ-glutamylcysteine; GCL, γ-glutamylcysteine ligase; Gpx, glutathione peroxidase; GS, glutathione synthetase; GSH, reduced glutathione; GSSG, glutathione disulfide; GST, glutathione-S-transferase; H2O2, hydrogen peroxide; NADPH, reduced nicotinamide adenine dinucleotides phosphate; NADP+, oxidized nicotinamide adenine dinucleotides phosphate; Prot-SH, thiolprotein; Prot-SSG, glutathionylated protein; ROOH, organic hydroperoxides; Trx, thioredoxin; TrxR, thioredoxin reductase. Ox: oxidized; Red: reduced.
372
Free Radical Biology and Medicine 143 (2019) 366–374
N. Sasoni, et al.
D-Ala-D-Ala ligase), which presents negligible similarity with the classic GS but is considered to be a domain involved in the synthesis of GSH [4–6,33]. Gopal et al. (2005) also conducted a bioinformatics analysis to findi that L. interrogans, serovar Lai 56601, contains separated (but adjacent) ORFs coding for a GCL (LA2106) and one protein wth ATP-Grasp domain (LA2107, encoding by COG-DdlA gene) similar to C-terminal of bifunctional enzymes. From the background detailed above, we performed a similar search (with a predictor of operons, http://www.microbesonline.org/operons/ ) in order to know if the gene context of the coding sequence for the GCL of L. interrogans serovar Copenhageni is similar to the reported for the serovar Lai. Our analysis indicated that the gene contexts for GCL and the protein with ATP-Grasp (a putative non classic GS enzyme) are similar in both serovars (Supplemental data - Fig. 7). Also, amino acid sequences of the putative non classic GS in L. interrogans serovar Copenhageni and its orthologue present in the serovar Lai 56601 (LA2107) show 100% identity (data non shown). The low sequence identity of LinGS with the classic GS is clear after the sequences alignment shown Supplemental data - Fig. 8; whereas the higher sequence identity of LinGS with the ATP-Grasp domain responsible for GS activity of bifunctional proteins it evidenced in the sequences alignment depicted in Supplemental data – Fig. 9. From these in silico analyses we could infer the existence of a putative GS-like protein in L.interrogans, which would explain the activity detected in cellular extracts of the bacterium. On other hand, the sequences alignment performed (Supplemental data - Fig. 10) point out that LinGCL is more conserved with the homologous proteins from organisms that contain a single gene (gshA) than with modular GCL-GS proteins (bifunctional enzymes encoded by gshF gene). The identification of sequences putatively encoding two enzymes (GCL and GS) involved in GSH-synthesis in L. interrogans prompted us to obtain both proteins in a recombinant form, with high purity, to characterize them at the kinetic and structural level. Some properties exhibited by LinGCL are similar to the homologous enzyme from other organisms [26,30] [26,40,41]. Thus, the enzyme presented a monomeric structure [26,30], which was not affected by redox conditions in the environment [40] and the pH optimum for activity was in the range 7.5–8.5 [26,41]. On the other hand, LinGCL exhibited some degree of promiscuity, as exemplified by: (i) the capacity to catalyze the addition of L-Asp or L-Ser on γ-GC as alternative reaction, but in lesser magnitude to conventional GCL activity; (ii) the ability to use GTP with a catalytic efficiency ~50% lower respect to ATP (in presence of L-Glu and L-Cys as co-substrates); and (iii) the greater affinity for L-Cys than for L-Glu. In the literature there are discordant reports on the characteristics of usage of alternative substrates by GCL from different sources [30,31,42,43]. Indeed, the GCL from E. coli shows wide ability to bind substrate analogues to L-Glu and L-Cys [31]. On the contrary, the human enzyme is inactive with L-Asp (as analogues of L-Glu) but it accepts a variety of analogues of L-Cys (such as L-Ser) [31,42]. Curiously, although the GCL from rat is structurally different to that from E. coli they have similar kinetic properties [43]. Considering the observed higher affinity of LinGCL for L-Cys, this could be related to a low intracellular level of this amino acid in L. interrogans. Although at the present the levels of amino acids in the bacterium are unknown, it is tempting to speculate that the high affinity of the enzyme toward Cys would priorize the intracellular use of this metabolite for synthesis of γGC rather than for reactions involving other enzymes. A characteristic observed in LinGCL is that it exhibits a marked product inhibition by γ-GC. Until now, very few cases of this product inhibition were reported for GCL from other organisms. Similar behavior to LinGCL was observed for GCL from Streptococcus agalactiae [32]. We determined that this product exhibited a competitive inhibitory effect respect to L-Glu and L-Cys, but not in relation to ATP. These results suggest that the product would affect the binding of both amino acid substrates at the binding site. The low Ki determined for γ-GC
supports a potential negative regulation of the enzyme activity, which could control the use of amino acids (mainly L-Cys) and nucleotides (ATP or GTP) according to redox and energetic conditions in the cell. Besides, the fact that the enzyme was insensitive to bis-γ-GC might be analyzed as that the inhibition of LinGCL by γ-GC could maintain a balance between the synthesis and degradation of this latter thiol. In this context, under oxidative stress conditions part of γ-GC is oxidized to bis-γ-GC, which would lead to a lower inhibition of the enzyme. This latter would allow further synthesis of γ-GC, thus contributing to redox homeostasis. The GSH inhibitory effect is similar to observed for γ-GC. These results could indicate that the synthesis of γ-GC, is mainly regulated by the availability of this end-product metabolite via negative feedback. On the other, the low ability of LinGCL to catalyze the formation of a γ-GC/L-Glu conjugate (analogous to GSH, where L-Glu replaces L-Gly) could be compared with similar compounds that are generated by plant GCL [40]. These conjugates have been considered as precursors of long chain peptides involved in the detoxification of heavy metals [44]. Concerning the characterization of recombinant LinGS (an ATPGrasp protein with classic GS activity), the properties we determined can be compared in good agreement with those reported for the homologous enzyme from other organisms [3,28,35,45]. Point examples of such comparisons are: (i) we found that LinGS exhibited specific for its canonical substrates, performing greater catalytic capacity than LinGCL [3,35]; (ii) the optimum pH for the GS activity was determined to be 8.0 [28,45]; and (iii) LinGS quaternary structure was found homodimeric [45,46]. In addition, the native structure and functional activity of LinGS were not modified by incubation with redox agents such as DTT or H2O2. LinGS exhibited high apparent affinity for γ-GC when compared to other substrates. This would guarantee the non-inhibition of the first enzyme (LinGCL) of the pathway (by accumulation of γ-GC) and, consequently, a continuous flow of GSH synthesis. Our data indicate that GSH inhibit the synthetase activity by competition with the binding of γ-GC and, strikingly, this inhibitory effect is much lower than for LinGCL. This suggests that LinGS functional activity could be controlled mainly by the availability of its substrates. The latter is consistent with that reported for homologous enzymes present in other organisms [47]. The GSH level was quantified in L. interrogans cells. The presence of this metabolite in the bacterium causing leptospirosis could be related to its survival at high oxidative stress conditions, such as informed in other pathogens [48]. In addition, GSH, besides contributing to the maintenance of redox homeostasis, could be involved in mechanisms of post-translational modification of proteins, such as S-glutathionylation [48,49]. The S-glutahtionylation might serve to prevent the irreversible oxidation of protein Cys residues, so that they can be reduced to their native state when the organism has been removed from the oxidative environment [37]. In fact, Grxs are very efficient in the reduction of the mixed disulfides in glutathionylated proteins (process called de-glutathionylation) [37]. Thus far, the identification of S-glutathionylation targets in L. interrogans as well as the knowledge about Grx functions remains scarce, being important subjects for future studies. As a main conclusion, Fig. 6 schematizes a metabolic redox scenario taking place in L. interrogans. Such a picture includes the enzymes characterized in this work as well as the enzymes and proteins that are GSH-dependent (the functions of which have not yet been studied). As we point out in the paragraph above, the Grx could be involved in Sglutathionylation process, as well as the GPxs that could be important in the peroxide detoxification (working together with catalase [50]) and AhpC [51]). Thus far, the results obtained in this work establish a first biochemical characterization of a GSH-dependent metabolism in L. interrogans. Further studies will be relevant to determine a complete GSH metabolism in this bacterium, which is critical for the understanding of physiology and pathogenicity of the organism.
373
Free Radical Biology and Medicine 143 (2019) 366–374
N. Sasoni, et al.
Acknowledgements
[25] H. Daiyasu, K. Watanabe, H. Toh, Recruitment of thioredoxin-like domains into prostaglandin synthases, Biochem. Biophys. Res. Commun. 369 (2) (2008) 281–286. [26] J.J. Abbott, J.L. Ford, M.A. Phillips, Substrate binding determinants of Trypanosoma brucei gamma-glutamylcysteine synthetase, Biochemistry 41 (8) (2002) 2741–2750. [27] T. Hara, T. Tanaka, H. Kato, T. Nishioka, J. Oda, Site-directed mutagenesis of glutathione synthetase from Escherichia coli B: mapping of the gamma-L-glutamylL-cysteine-binding site, Protein Eng. 8 (7) (1995) 711–716. [28] S.K. Sharma, H.S. Banyal, Characterization of Plasmodium berghei glutathione synthetase, Parasitol. Int. 60 (3) (2011) 321–323. [29] O.W. Griffith, R.T. Mulcahy, The enzymes of glutathione synthesis: gamma-glutamylcysteine synthetase, Adv. Enzymol. Relat. Area Mol. Biol. 73 (1999) 209–267 (xii). [30] T. Hibi, H. Hisada, T. Nakatsu, H. Kato, J. Oda, Escherichia coli B gamma-glutamylcysteine synthetase: modification, purification, crystallization and preliminary crystallographic analysis, Acta. Crystallogr. D Biol. Crystallogr. 58 (Pt 2) (2002) 316–318. [31] B.S. Kelly, W.E. Antholine, O.W. Griffith, Escherichia coli gamma-glutamylcysteine synthetase. Two active site metal ions affect substrate and inhibitor binding, J. Biol. Chem. 277 (1) (2002) 50–58. [32] B.E. Janowiak, M.A. Hayward, F.C. Peterson, B.F. Volkman, O.W. Griffith, Gammaglutamylcysteine synthetase-glutathione synthetase: domain structure and identification of residues important in substrate and glutathione binding, Biochemistry 45 (35) (2006) 10461–10473. [33] W. Li, Z. Li, J. Yang, Q. Ye, Production of glutathione using a bifunctional enzyme encoded by gshF from Streptococcus thermophilus expressed in Escherichia coli, J. Biotechnoly (2011) 261–268. [34] H. Kumagai, R. Nakayama, T. Tochikura, γ-Glutamylcysteine synthetase from Proteus mirabilis, Agric. Biol. Chem. 46 (5) (1982) 1301–1309. [35] K. Kino, S. Kuratsu, A. Noguchi, M. Kokubo, Y. Nakazawa, T. Arai, M. Yagasaki, K. Kirimura, Novel substrate specificity of glutathione synthesis enzymes from Streptococcus agalactiae and Clostridium acetobutylicum, Biochem. Biophys. Res. Commun. 352 (2) (2007) 351–359. [36] J.S. Davis, J.B. Balinsky, J.S. Harington, J.B. Shepherd, Assay, purification, properties and mechanism of action of gamma-glutamylcysteine synthetase from the liver of the rat and Xenopus laevis, Biochem. J. 133 (4) (1973) 667–678. [37] L. Masip, K. Veeravalli, G. Georgiou, The many faces of glutathione in bacteria, Antioxidants Redox Signal. 8 (5–6) (2006) 753–762. [38] D.A. Dickinson, H.J. Forman, Cellular glutathione and thiols metabolism, Biochem. Pharmacol. 64 (5–6) (2002) 1019–1026. [39] N. Sasoni, A.A. Iglesias, S.A. Guerrero, D.G. Arias, Functional thioredoxin reductase from pathogenic and free-living Leptospira spp, Free Radic. Biol. Med. 97 (2016) 1–13. [40] R. Gromes, M. Hothorn, E.D. Lenherr, V. Rybin, K. Scheffzek, T. Rausch, The redox switch of gamma-glutamylcysteine ligase via a reversible monomer-dimer transition is a mechanism unique to plants, Plant J. 54 (6) (2008) 1063–1075. [41] J.M. Jez, R.E. Cahoon, S. Chen, Arabidopsis thaliana glutamate-cysteine ligase: functional properties, kinetic mechanism, and regulation of activity, J. Biol. Chem. 279 (32) (2004) 33463–33470. [42] T. Hibi, H. Nii, T. Nakatsu, A. Kimura, H. Kato, J. Hiratake, J. Oda, Crystal structure of gamma-glutamylcysteine synthetase: insights into the mechanism of catalysis by a key enzyme for glutathione homeostasis, Proc. Natl. Acad. Sci. U. S. A. 101 (42) (2004) 15052–15057. [43] C.S. Hang, W.R. Moore, A. Meister, On the active site thiol of y-glutamylcysteine synthetase: relationships to catalysis, inhibition, and regulation, Proc. Natl. Acad. Sci. U. S. A. 85 (1987) 2464–2468. [44] P. Meuwly, P. Thibault, W.E. Rauser, gamma-Glutamylcysteinylglutamic acid - a new homologue of glutathione in maize seedlings exposed to cadmium, FEBS Lett. 336 (3) (1993) 472–476. [45] J.L. Luo, C.S. Huang, K. Babaoglu, M.E. Anderson, Novel kinetics of mammalian glutathione synthetase: characterization of gamma-glutamyl substrate cooperative binding, Biochem. Biophys. Res. Commun. 275 (2) (2000) 577–581. [46] D. Gudrun, K. Maria-Regina, Purification and evaluation of the glutathione-synthesizing enzymes from Candida boidinii for cell-free synthesis of glutathione, J. Biotechnol. 4 (3) (1986) 143–158. [47] Y. Liu, A.S. Hyde, M.A. Simpson, J.J. Barycki, Emerging regulatory paradigms in glutathione metabolism, Adv. Cancer Res. 122 (2014) 69–101. [48] V.V. Loi, M. Rossius, H. Antelmann, Redox regulation by reversible protein Sthiolation in bacteria, Front. Microbiol. 6 (2015) 187. [49] I. Dalle-Donne, R. Rossi, G. Colombo, D. Giustarini, A. Milzani, Protein S-glutathionylation: a regulatory device from bacteria to humans, Trends Biochem. Sci. 34 (2) (2009) 85–96. [50] A. Eshghi, K. Lourdault, G.L. Murray, T. Bartpho, R.W. Sermswan, M. Picardeau, B. Adler, B. Snarr, R.L. Zuerner, C.E. Cameron, Leptospira interrogans catalase is required for resistance to H2O2 and for virulence, Infect. Immun. 80 (11) (2012) 3892–3899. [51] D.G. Arias, A. Reinoso, N. Sasoni, M.D. Hartman, A.A. Iglesias, S.A. Guerrero, Kinetic and structural characterization of a typical two-cysteine peroxiredoxin from Leptospira interrogans exhibiting redox sensitivity, Free Radic. Biol. Med. 77 (2014) 30–40. [52] R.C. Fahey, W.C. Brown, W.B. Adams, M.B. Worsham, Occurrence of glutathione in bacteria, J. Bacteriol. 133 (3) (1978) 1126–1129.
This work was supported by grants from ANPCyT (PICT2013-0253, PICT2014-2103, PICT2016-1778 and PICT2017-2268). NS and FDML are fellows from CONICET. DGA, AAI and SAG are investigator career members from CONICET. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.freeradbiomed.2019.08.028. References [1] H. Zhang, H.J. Forman, Glutathione synthesis and its role in redox signaling, Semin. Cell Dev. Biol. 23 (7) (2012) 722–728. [2] E.V. Kalinina, N.N. Chernov, M.D. Novichkova, Role of glutathione, glutathione transferase, and glutaredoxin in regulation of redox-dependent processes, Biochemistry (Mosc.) 79 (13) (2014) 1562–1583. [3] M.D. Shelly, C. Lu, Glutathione synthesis, Biochim. Biophys. Acta 1830 (5) (2013) 3143–3153. [4] S. Gopal, I. Borovok, A. Ofer, M. Yanku, G. Cohen, W. Goebel, J. Kreft, Y. Aharonowitz, A multidomain fusion protein in Listeria monocytogenes catalyzes the two primary activities for glutathione biosynthesis, J. Bacteriol. 187 (11) (2005) 3839–3847. [5] B.E. Janowiak, O.W. Griffith, Glutathione synthesis in Streptococcus agalactiae. One protein accounts for gamma-glutamylcysteine synthetase and glutathione synthetase activities, J. Biol. Chem. 280 (12) (2005) 11829–11839. [6] B. Vergauwen, D. De Vos, J.J. Van Beeumen, Characterization of the bifunctional gamma-glutamate-cysteine ligase/glutathione synthetase (GshF) of Pasteurella multocida, J. Biol. Chem. 281 (7) (2006) 4380–4394. [7] R.C. Fahey, Glutathione analogs in prokaryotes, Biochim. Biophys. Acta 1830 (5) (2013) 3182–3198. [8] B. Vergauwen, J. Elegheert, A. Dansercoer, B. Devreese, S.N. Savvides, Glutathione import in Haemophilus influenzae Rd is primed by the periplasmic heme-binding protein HbpA, Proc. Natl. Acad. Sci. U. S. A. 107 (30) (2010) 13270–13275. [9] B. Vergauwen, K. Verstraete, D.B. Senadheera, A. Dansercoer, D.G. Cvitkovitch, E. Guedon, S.N. Savvides, Molecular and structural basis of glutathione import in Gram-positive bacteria via GshT and the cystine ABC importer TcyBC of Streptococcus mutans, Mol. Microbiol. 89 (2) (2013) 288–303. [10] D. Parsonage, D.C. Desrosiers, K.R. Hazlett, Y. Sun, K.J. Nelson, D.L. Cox, J.D. Radolf, L.B. Poole, Broad specificity AhpC-like peroxiredoxin and its thioredoxin reductant in the sparse antioxidant defense system of Treponema pallidum, Proc. Natl. Acad. Sci. U. S. A. 107 (14) (2010) 6240–6245. [11] J.A. Boylan, C.S. Hummel, S. Benoit, J. Garcia-Lara, J. Treglown-Downey, E.J. Crane 3rd, F.C. Gherardini, Borrelia burgdorferi bb0728 encodes a coenzyme A disulphide reductase whose function suggests a role in intracellular redox and the oxidative stress response, Mol. Microbiol. 59 (2) (2006) 475–486. [12] J. Sambrook, D.W. Russell, Molecular Cloning: A Laboratory manual., United States, (2001). [13] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (5259) (1970) 680–685. [14] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248–254. [15] H.C. Ellinghausen, W.G. McCullough, Nutrition of Leptospira pomona and growth of 13 other serotypes: fractionation of oleic albumin complex and a medium of bovine albumin and polysorbate 80, Am. J. Vet. Res. 26 (1965) 45–51. [16] R.C. Johnson, V.G. Harris, Differentiation of pathogenic and saprophytic letospires. I. Growth at low temperatures, J. Bacteriol. 94 (1) (1967) 27–31. [17] H. Louvel, M. Picardeau, Genetic manipulation of Leptospira biflexa, Curr. Protoc. Microbiol. Chapter 12 (2007) Unit 12E 4. [18] C. Fusari, A.M. Demonte, C.M. Figueroa, M. Aleanzi, A.A. Iglesias, A colorimetric method for the assay of ADP-glucose pyrophosphorylase, Anal. Biochem. 352 (1) (2006) 145–147. [19] Y.C. Cheng, W.H. Prusoff, Mouse ascites sarcoma 180 thymidylate kinase. General properties, kinetic analysis, and inhibition studies, Biochemistry 12 (14) (1973) 2612–2619. [20] B. Mannervik, Measurement of glutathione reductase activity, Curr. Protoc Toxicol. Chapter 7 (2001) Unit7 2. [21] V.E. Marquez, D.G. Arias, C.V. Piattoni, C. Robello, A.A. Iglesias, S.A. Guerrero, Cloning, expression, and characterization of a dithiol glutaredoxin from Trypanosoma cruzi, Antioxidants Redox Signal. 12 (6) (2010) 787–792. [22] W.H. Habig, M.J. Pabst, W.B. Jakoby, Glutathione S-transferases. The first enzymatic step in mercapturic acid formation, J. Biol. Chem. 249 (22) (1974) 7130–7139. [23] G. Smirnova, N. Muzyka, O. Oktyabrsky, Transmembrane glutathione cycling in growing Escherichia coli cells, Microbiol. Res. 167 (3) (2012) 166–172. [24] N. Allocati, L. Federici, M. Masulli, C. Di Ilio, Glutathione transferases in bacteria, FEBS J. 276 (1) (2009) 58–75.
374
