The rhdA gene of Azotobacter vinelandii codes for RhdA, a rhodanese-domain protein with an active-site loop structure which has not currently been found in proteins of the rhodanese-homology superfamily. Considering the lack of information on the functional role of the ubiquitous rhodaneses, in the present study we examined the in vivo functions of RhdA by using an A. vinelandii mutant strain (MV474), in which the rhdA gene was disrupted by deletion. Preliminary phenotypic characterization of the rhdA mutant suggested that RhdA could exert protection over Fe–S enzymes, which are easy targets for oxidative damage. To highlight the role of RhdA in preserving sensitive Fe–S clusters, in the present study we analysed the defects of the rhdA-null strain by exploiting growth conditions which resulted in enhancing the catalytic deficiency of enzymes with vulnerable Fe–S clusters. We found that a lack of RhdA impaired A. vinelandii growth in the presence of gluconate, a carbon source that activates the Entner–Doudoroff pathway in which the first enzyme, 6-phosphogluconate dehydratase, employs a 4Fe–4S cluster as an active-site catalyst. By combining proteomics, enzymatic profiles and model systems to generate oxidative stress, evidence is provided that to rescue the effects of a lack of RhdA, A. vinelandii needed to activate defensive activities against oxidative damage. The possible functionality of RhdA as a redox switch which helps A. vinelandii in maintaining the cellular redox balance was investigated by using an in vitro model system that demonstrated reversible chemical modifications in the highly reactive RhdA Cys230 thiol.

INTRODUCTION

The ubiquitous rhodanese-domain proteins catalyse in vitro the transfer of a sulfur atom from a suitable sulfur donor (thiosulfate for rhodaneses and 3-mercaptopyruvate for 3-mercaptopyruvate sulfurtransferases) to cyanide, with concurrent formation of thiocyanate [1]. In the majority of organisms, the proteins of this homology superfamily (Pfam accession number PF00581; http://www.sanger.ac.uk/Software/Pfam/) are present as paralogues and they are found as tandem rhodanese repeats, single-domain proteins or in combination with distinct protein domains [2]. The wide variability in the amino acids of the active-site loop of rhodanese-domain proteins [2] makes it conceivable that substrate recognition and biological interactions are driven by the specific active-site structure. From the findings that bovine rhodanese [3], Escherichia coli GlpE [4] and the 3-mercaptopyruvate sulfurtransferase from Leishmania [5] have a higher affinity for reduced thioredoxins than for cyanide, the concept that the abundant rhodanese-like proteins could play roles in managing the processes of stress tolerance is emerging, although in vivo experimental evidence is still lacking. The relevance of specific rhodanese-like proteins in the maintenance of redox homoeostasis has been suggested by in vitro studies using rat mercaptopyruvate sulfurtransferase [6,7]. As for bacterial rhodanese-domain proteins, possible involvement in physiological processes related to xenobiotic-induced oxidative stress and detoxification was indicated by proteomic analyses [8,9]. Recently, we have found evidence that rhodanese-like enzymes favoured PAH (polycyclic aromatic hydrocarbon) degradation in selected Mycobacterium strains [10]. This experimental evidence suggests that rhodanese domains might function as regulatory devices in specific physiological situations, but the redundancy of these proteins in the same organism makes it difficult to unravel their cellular roles.

Among the redundant rhodanese-homology proteins of Azobacter vinelandii, the tandem domain protein RhdA [11] contains an active-site motif (HCQTHHR) which is not commonly found in rhodanese-domain proteins. Crystallographic investigations [12] showed that the RhdA catalytic centre is surrounded by a strong positive electrostatic field that originates from the peculiar active-site loop structure, and from the vicinity of positively charged groups. Considering the lack of general information on the biological role of rhodaneses, we undertook a study of the in vivo function of RhdA by using an A. vinelandii mutant strain (MV474), in which the rhdA gene was disrupted by deletion [11]. Initial characterization of MV474 grown in a sucrose medium revealed that the lack of RhdA enhanced the expression of enzymes of the PHB (polyhydroxybutyrate) biosynthetic operon, giving rise to early accumulation of PHB, and negatively affected the activity of tricarboxylic acid cycle Fe–S enzymes [13]. The effect on aconitate hydratase was dramatic, in spite of the comparable expression of aconitate hydratase polypeptides in both wild-type and rhdA mutant strains. These results [13] led to the proposal that RhdA provides protection against oxidative events that in the aerobe A. vinelandii lead to the inactivation of enzymes containing labile Fe–S [14]. To highlight the role of RhdA in preserving sensitive Fe–S clusters, in the present study, we analysed the defects of the rhdA-null strain MV474 by exploiting growth conditions that enhance the catalytic deficiency of enzymes with vulnerable Fe–S clusters [15]. We found that a lack of RhdA impaired A. vinelandii growth in the presence of gluconate, a carbon source that activates the ED (Entner–Doudoroff) pathway [16], in which the first enzyme, 6-phosphogluconate dehydratase, employs a vulnerable 4Fe–4S cluster as an active-site catalyst [17]. The use of gluconate as a carbon source proved to be an useful growth condition in which to make an in-depth study into the initially hypothesized [13] role of RhdA in protecting cells from oxidative damage. By combining proteomics, enzymatic profiles and model systems to generate oxidative-stress, in the present report we present evidence to show that to rescue the effects of the lack of RhdA, A. vinelandii needed to activate defensive activities against oxidative damage. The possible functionality of RhdA as a redox switch which helps A. vinelandii in maintaining the redox balance was investigated by using an in vitro model system to demonstrate the occurrence of chemical modifications in the highly reactive RhdA Cys230 thiol, the only one present in the molecule.

EXPERIMENTAL

Bacterial strains and growth conditions

The A. vinelandii strains used in this study were UW136 and a derivative of UW136 (MV474), in which disruption of the rhdA gene was achieved by the insertion of a KIXX (kanamycin resistance) cassette, following deletion of 584 bp as described previously [11]. Western blot analysis with MV474 [13] clearly showed that no protein corresponding to RhdA is expressed in this mutant. Cells were grown for 24 h at 30 °C under aerobic conditions in BGN (Burk's gluconate nitrogen) medium (Burk's medium supplemented with 0.2% gluconate and 15 mM ammonium acetate). When the attenuance (D) at 600 nm reached 2.0, the cells were centrifuged at 3800 g for 10 min and stored at −80 °C. For oxidative-stress sensitivity assays, cells were grown in the above conditions up to D600=0.8, then the cultures of either UW136 or MV474 strains were divided into two equal samples, one of which was treated with PMS (phenazine methosulfate) (final concentration of 15 μM).

Sample preparation and gel electrophoresis

Cells were harvested by centrifugation at 2000 g for 30 min and suspended in a 10 mM Tris/HCl (pH 8.0) buffer containing 1.5 mM MgCl2, 10 mM KCl and 0.1% SDS. A buffer volume which was approximately equal to the packed cell volume was used. Before analysis, 100 μl of the sample was solubilized in 1 ml of lysis buffer [7 M urea, 2 M thiourea, 4% (w/v) CHAPS, 65 mM DTT (dithiothreitol) and 2% (v/v) IPG buffer (pH 4–7) (Amersham Biosciences)] and incubated for 20 min on ice. Protein concentrations were estimated using a 2-DE (two-dimensional PAGE) Quant Kit (Amersham Biosciences) following the manufacturer's instructions before adding 0.01% Bromophenol Blue. Proteins were separated by 2-DE using Immobiline Dry Strips (pH 4–7 linear, 14 cm long; Amersham Biosciences). The sample (180 μg of protein) was loaded on the immobilized pH gradient strip after overnight in-gel re-swelling procedure performed following the manufacturer's instructions. The isoelectric focusing was carried out at 20 °C using a Multiphor II electrophoresis unit (Amersham Biosciences) with the following programme: 75 V for 1 h, 300 V for 1 h, 300–1000 V for 4 h, 1000–8000 V for 6 h, 8000 V for 2 h and 50 V for 1 h. After focusing, Immobiline Dry Strips were equilibrated for 15 min at room temperature (19 °C) in equilibration buffer [50 mM Tris/HCl (pH 8.8), 6 M urea, 30% (v/v) glycerol and 2% (w/v) SDS] containing 1% DTT, followed by a second incubation step for 15 min at room temperature in the equilibration buffer in the presence of 2.5% (w/v) iodoacetamide. Separation in the second dimension (SDS/PAGE) was carried out on homogeneous polyacrylamide gels (12% gels) at 10 mA per gel for 30 min, and at 25 mA for 4 h in a Hoefer SE600 Electrophoresis Unit. Gels were silver stained as described previously [18]. In order to corroborate the reproducibility of the results, at least three independent experiments for each strain were performed, and three gels were run for each sample. Silver-stained gels were digitized using an Epson Expression 1680 Pro scanner. ImageMaster 2D Platinum software (Amersham Biosciences) was used for gel-image analysis, including quantification of the spot intensities, which is performed on a volume basis [i.e. values were calculated from the integration of spot optical intensity over the spot area, and is referred to as the %V (relative volume)]. %V is the ratio of the volume of a spot and the sum of the volume of all the spots detected in the same gel. Gels from independent analyses were matched together, and the %V for each spot in the two gel sets (UW 136 and MV474 A. vinelandii strains) was compared. Significant changes in the abundance of selected proteins (Table 1) were estimated by Student's t test (P<0.05). The average expression ratio between the MV474 and UW136 strains was calculated from the averages of %V of five replicates.

Table 1
Proteins identified by MALDI-TOF MS analysis

The relative abundance (%V) of 6-phosphogluconate dehydratase (1, Q4ITN0), aconitate hydratase 1 (2, Q4IW32), aconitate hydratase 2 (3, Q4J5V0), acetoacetyl-CoA reductase (4, Q93QF0) and β-oxothiolase (5, Q9KH97) was determined and used to calculate the protein ratio of MV474 and UW136 [results are means±S.E.M. (n=5)]. Accession numbers refer to the Swiss-Prot/TrEMBL database. n.s.d., not significantly different.

Spot number Protein name Accession number Sequence coverage (%) Molecular mass (Da) pI Protein ratio (MV474/UW136) 
6-Phosphogluconate dehydratase Q4ITN0 32 65348 6.0 n.s.d. 
Aconitate hydratase 1 Q4IW32 25 97314 5.3 1.37±0.02 
Aconitate hydratase 2 Q4J5V0 22 93347 5.5 −0.69±0.02 
Acetoacetyl-CoA reductase Q93QF0 63 26726 6.2 n.s.d. 
β-Oxothiolase Q9KH97 50 40875 6.7 1.23±0.07 
Spot number Protein name Accession number Sequence coverage (%) Molecular mass (Da) pI Protein ratio (MV474/UW136) 
6-Phosphogluconate dehydratase Q4ITN0 32 65348 6.0 n.s.d. 
Aconitate hydratase 1 Q4IW32 25 97314 5.3 1.37±0.02 
Aconitate hydratase 2 Q4J5V0 22 93347 5.5 −0.69±0.02 
Acetoacetyl-CoA reductase Q93QF0 63 26726 6.2 n.s.d. 
β-Oxothiolase Q9KH97 50 40875 6.7 1.23±0.07 

MS analysis and identification of protein spots

In-gel enzymatic digestion of the spots and peptide extraction was carried out as described previously [19]. MALDI-TOF (matrix-assisted laser-desorption ionization–time-of-flight) MS analysis was performed in a Voyager DE-Pro spectrometer (PerSeptive BioSystems). External mass calibration was performed with low-mass peptide standards, and the mass-measurement accuracy was ±0.3 Da. The mass spectra were acquired in reflector mode using Delay Extraction technology. Raw data were elaborated using Data Explorer 5.0 software provided by the manufacturer (PerSeptive BioSystems). The software package Protein Prospector MS-Fit (http://prospector.ucsf.edu) was used to identify protein spots from the NCBI (National Center for Biotechnology Information) protein- and gene-sequence databases. A number of top candidates with high scores from the peptide-matching analysis were further evaluated by comparing their calculated pI and molecular masses using the experimental values obtained from the 2-DE gel. These two parameters were used with large tolerances (deviation in pI=1 and deviation in molecular mass=±15% molecular mass) as filters to exclude false-positive candidates from the output lists.

Enzymatic activity assays

Enzymatic activities were determined in samples (150–300 μg of protein) of cell-free extracts prepared anaerobically by sonication (five 30 s pulses with intermitted 1 min cooling periods) in lysis buffer A [50 mM Tris/HCl (pH 8.0), 100 mM NaCl and 1 mM DTT].

Aconitate hydratase and 6-phosphogluconate dehydratase activities were tested as described by Varghese et al. [20] and Outten et al. [21] respectively. Catalase activity was measured using the method described previously [22], where 1 unit of enzyme is defined as the amount of enzyme that decomposes 1 μmol of H2O2 per min at 37 °C.

For re-activation experiments, before the activity tests, cell-free extracts were anaerobically incubated for 10 min at room temperature in the presence of 1 mM FeSO4 and 5 mM DTT.

Rhodanese activity was tested by a discontinuous method that quantifies the product thiocyanate [23] in the presence of thiosulfate as a sulfur donor [TST (thiosulfate–cyanide sulfurtransferase) activity]. The assays lasted 1 or 2 min, and 1 unit of enzyme is defined as the amount of enzyme that produces 1 μmol of thiocyanate per min at 37 °C.

For re-activation experiments of PMS-treated purified RhdA (1:1 molar ratio of RhdA/PMS), the mixtures after gel filtration (10 μM RhdA) were incubated at 4 °C in the presence of 10 μM E. coli reduced thioredoxin (Sigma), 0.2 μM rat liver thioredoxin reductase (Sigma) and 50 μM NADPH in 20 mM phosphate buffer (pH 7.4) (final volume of 100 μl). In the case of PMS-treated A. vinelandii UW136, the extracts (200 μg of protein) were incubated in the presence of 50 μM thioredoxin, 0.2 μM rat liver thioredoxin reductase and 100 μM NADPH. After 60 min incubation, TST activity was measured as stated above.

In vitro study of overexpressed RhdA

Overexpressed His-tagged RhdA, purified as described previously [24], was used to test the in vitro effects of the oxidative agent PMS. RhdA (100 μM) in 50 mM Tris/HCl (pH 8.0) and 100 mM NaCl was incubated aerobically with PMS (final concentration of 1 mM) at room temperature. At fixed incubation times, sample aliquots were gel-filtered by using Sephadex G-25 columns to remove PMS. The gel-filtered samples were subjected to the rhodanese activity test [23], non-reducing SDS/PAGE [25] and MALDI-TOF MS analyses. Densitometric analyses of the electrophoretic pattern were performed on Coomassie-Blue-stained gels using ImageMaster 1D Elite software (Amersham Biosciences). The MALDI-TOF MS analysis was performed by using a Bruker Daltonics Reflex IV instrument equipped with a nitrogen laser (337 nm) and operated in a positive linear mode with a matrix of α-cyano-4-hydroxycinnamic acid. External standards were used for calibration (Bruker protein II calibration standards) and the mass measurement accuracy was ±1 Da. Each spectrum was accumulated for at least 200 laser shots.

Analytical procedures

The protein concentration was determined using the Bradford assay [26], with BSA used as a standard. PHB content was determined by spectrophotometry as described previously [27].

RESULTS

The lack of RhdA affects A. vinelandii growth on gluconate as a carbon source

In the sucrose medium, which we employed for our initial experiments, the rhdA mutant MV474 grew as well as the wild-type UW136 strain, but the metabolic imbalance of the MV474 strain suggested that the absence of RhdA favoured oxidative events in A. vinelandii [13]. To elucidate better the protective cellular roles of RhdA, more critical growth conditions were exploited, and we found that the growth of A. vinelandii MV474 was impaired in the presence of gluconate as a carbon source. Utilization of gluconate via the ED pathway [16] involves the activity of 6-phosphogluconate dehydratase, an enzyme containing a 4Fe–4S cluster which is an easy target for oxidative damage [17]. Under this growth condition (Figure 1), the lack of RhdA caused a significant inhibition of growth at early stages, and, after 24 h of growth, MV474 reached a final culture density of approx. 90% of that of the wild-type strain. At this stage, the accumulation of PHB was 8-fold higher in the rhdA-null mutant than in UW136. PHB contentration (as determined in three separate growth experiments) was 163±10 μg per mg of protein in UW136, and 1320±25 μg per mg of protein in the mutant MV474 strain, confirming that the metabolic profile of the mutant MV474 strain did favour the synthesis and accumulation of this polymer. Greater accumulation of PHB in the rhdA-null mutant than in UW136 was, indeed, also observed when sucrose was the carbon source provided [13]. PHB accumulation is generally considered a ‘cell-survival mechanism’ [28] that endows bacteria with enhanced stress tolerance [29], thus indicating that PHB accumulation in MV474 represents a response to the physiological imbalance inferred by the lack of RhdA.

The lack of RhdA affects A. vinelandii growth on gluconate as a carbon source

Figure 1
The lack of RhdA affects A. vinelandii growth on gluconate as a carbon source

Growth of A. vinelandii wild-type UW136 (●) and rhdA mutant MV474 (○) in BGN (Burk's gluconate nitrogen) medium was determined by measuring D600. Three independent experiments were carried out, and one typical set is shown here.

Figure 1
The lack of RhdA affects A. vinelandii growth on gluconate as a carbon source

Growth of A. vinelandii wild-type UW136 (●) and rhdA mutant MV474 (○) in BGN (Burk's gluconate nitrogen) medium was determined by measuring D600. Three independent experiments were carried out, and one typical set is shown here.

Protein expression in A. vinelandii wild-type and rhdA-null mutant strains

Comparative proteomics of gluconate-cultured UW136 and MV474 A. vinelandii strains were performed with the aim of identifying the relative abundance of specific enzymes strictly related to the metabolic picture evidenced in the absence of RhdA. The protein spots listed in Table 1 were identified by peptide mass fingerprinting, and their relative abundance (in terms of %V of each spot) was accurately calculated from independent gels with five experimental repeats for each strain (rhdA-null MV474 mutant and UW136 A. vinelandii strains). Through this comparative analysis, we found that the 6-phosphogluconate dehydratase polypeptide (spot 1 in Figure 2) was present in both strains, and that its expression level in MV474 and in UW136 was not significantly different (Table 1). Further analyses of the 2-DE protein profiles were focused to compare the expression of enzymes differently regulated in MV474 under growth conditions in the presence of sucrose [13]. The enzyme of the PHB biosynthetic operon [30] β-oxothiolase displayed increased expression with statistical significance in MV474 (Figure 2 and Table 1), thus explaining the 8-fold higher production of PHB in the rhdA-null mutant strain than in the wild-type strain. Significantly different expression levels of aconitate hydratase polypeptides (spots 2 and 3 in Figure 2) between the MV474 and UW136 strains (Table 1) were detected. Aconitate hydratase 2 was present in higher amounts in UW136 than in MV474, whereas aconitate hydratase 1 was overexpressed in MV474. In E. coli, both aconitate hydratase enzymes process the substrate citrate, and it was found that aconitate hydratase 1 (the product of the acnA gene) was induced by oxidative stress [31]. Evidence that a lack of RhdA results in altered protection from oxidative events, thus giving rise to a loss of the functionality of enzymes containing labile Fe–S clusters [15], was provided by enzymatic analyses (Table 2). The activity of 6-phosphogluconate dehydratase in extracts from MV474 was approx. 60% of that found in UW136, although the polypeptide expression in both A. vinelandii strains did not show statistically significant differences, and aconitate hydratase activity in MV474 was approx. 40% of that of UW136. Evident in vitro re-activation (approx. 1.5-fold differences from untreated samples) of both dehydratases was obtained in extracts from MV474 (Table 2) after treatment with ferrous iron and DTT. Re-activation of Fe–S enzymes by treatment with ferrous iron and DTT is considered to be an indication that damage to the clusters stems from an interaction with reactive oxygen species that induces the loss of iron [15]. Moreover, a probable imbalance of oxidative events in the rhdA-null mutant was suggested by the finding that the activity of the hydrogen peroxide scavenger catalase [3234] was higher in MV474 than in UW136 (30±2 units/mg and 50±2 units/mg in UW136 and MV474 respectively).

Representative 2-DE gel images of A. vinelandii UW136 and MV474 strains

Figure 2
Representative 2-DE gel images of A. vinelandii UW136 and MV474 strains

Whole-cell extracts from A. vinelandii wild-type UW136 and rhdA mutant MV474 strains grown aerobically up to an attenuance (D) at 600 nm of 2.0 in BGN (Burk's gluconate nitrogen) medium were separated by 2-DE as described in the Experimental section. The silver-stained gels of UW136 and MV474 are shown, and the numbers represent the spots identified by MALDI-TOF MS listed in Table 1. Molecular-mass markers are shown on the left-hand side (in kDa).

Figure 2
Representative 2-DE gel images of A. vinelandii UW136 and MV474 strains

Whole-cell extracts from A. vinelandii wild-type UW136 and rhdA mutant MV474 strains grown aerobically up to an attenuance (D) at 600 nm of 2.0 in BGN (Burk's gluconate nitrogen) medium were separated by 2-DE as described in the Experimental section. The silver-stained gels of UW136 and MV474 are shown, and the numbers represent the spots identified by MALDI-TOF MS listed in Table 1. Molecular-mass markers are shown on the left-hand side (in kDa).

Table 2
Activity of dehydratase enzymes in A. vinelandii wild-type and rhdA mutant strains

A. vinelandii UW136 and MV474 strains were grown aerobically in BGN medium up to an attenuance (D) at 600 nm of 2.0. The activities of 6-phosphogluconate dehydratase and aconitate hydratase were determined in samples (150–300 μg of protein) of cell-free extracts prepared anaerobically. Aconitate hydratase and 6-phosphogluconate dehydratase activities were expressed as μmol of cis-aconitate formed/min per mg of protein and μmol of pyruvate formed/min per mg of protein respectively. For re-activation experiments, before the activity tests, the cell-free extracts were anaerobically incubated for 10 min at room temperature in the presence of 1 mM FeSO4 and 5 mM DTT. Results are means±S.D. (n=6).

 UW136 MV474 
Basal activity   
 6-Phosphogluconate dehydratase 0.620±0.012 0.39±0.007 
 Aconitate hydratase 2.05±0.07 0.84±0.05 
Activity after treatment with FeSO4 and DTT   
 6-Phosphogluconate dehydratase 0.700±0.006 0.560±0.007 
 Aconitate hydratase 2.31±0.20 1.27±0.21 
 UW136 MV474 
Basal activity   
 6-Phosphogluconate dehydratase 0.620±0.012 0.39±0.007 
 Aconitate hydratase 2.05±0.07 0.84±0.05 
Activity after treatment with FeSO4 and DTT   
 6-Phosphogluconate dehydratase 0.700±0.006 0.560±0.007 
 Aconitate hydratase 2.31±0.20 1.27±0.21 

Defensive action of the rhdA gene product (RhdA) against superoxide stress

To demonstrate the involvement of RhdA in triggering protective processes against oxidative events, we analysed the response of A. vinelandii MV474 and UW136 strains to induced oxidative stress. The resistance of A. vinelandii to exposure to the oxidative agent PMS was severely affected by the absence of RhdA (Figure 3A). With regard to the untreated control cultures, the growth of the rhdA mutant was less than 40%, whereas the wild-type strain was not significantly affected by PMS treatment. Upon exposure to PMS, the decrease in aconitate hydratase activity, with regard to that measured in untreated extracts, was more pronounced in the rhdA mutant strain (5-fold) than in the wild-type strain (1.6-fold). As for 6-phosphogluconate dehydratase activity, 60% of the activity of untreated extracts was found in the wild-type strain, and 45% in the rhdA mutant.

Effects of the exposure to PMS on A. vinelandii wild-type and rhdA mutant strains

Figure 3
Effects of the exposure to PMS on A. vinelandii wild-type and rhdA mutant strains

A. vinelandii UW136 and MV474 strains were grown aerobically in BGN (Burk's gluconate nitrogen) medium (supplemented with 0.2% gluconate) up to D600=0.8, after which the cultures of either UW136 or MV474 strains were divided into two equal samples, one of which was treated with PMS (final concentration of 15 μM). The cultures were grown for 16 h, and 2-DE was performed as described in the legend of Figure 2. Three independent experiments were performed, and the D600 after 16 h exposure to PMS and the value recorded for PMS-untreated cultures are shown (A). (B) Enlargement of a 2-DE gel of the region showing the AhpC (Q4IS62) protein spot (arrow) in untreated UW136, untreated MV474, PMS-treated UW136 and PMS-treated MV474 (from left- to right-hand side).

Figure 3
Effects of the exposure to PMS on A. vinelandii wild-type and rhdA mutant strains

A. vinelandii UW136 and MV474 strains were grown aerobically in BGN (Burk's gluconate nitrogen) medium (supplemented with 0.2% gluconate) up to D600=0.8, after which the cultures of either UW136 or MV474 strains were divided into two equal samples, one of which was treated with PMS (final concentration of 15 μM). The cultures were grown for 16 h, and 2-DE was performed as described in the legend of Figure 2. Three independent experiments were performed, and the D600 after 16 h exposure to PMS and the value recorded for PMS-untreated cultures are shown (A). (B) Enlargement of a 2-DE gel of the region showing the AhpC (Q4IS62) protein spot (arrow) in untreated UW136, untreated MV474, PMS-treated UW136 and PMS-treated MV474 (from left- to right-hand side).

To investigate whether treatment of A. vinelandii with the superoxide generator PMS could prime different defensive processes in the wild-type and the rhdA mutant strains, the 2-DE protein profiles of both strains were compared under exposure to PMS. Accurate matching of these protein profiles was performed with the purpose of detecting protein spots which underlie the different responses of MV474 to PMS treatment compared with UW136. In PMS-treated MV474, a protein spot (molecular mass≈20 kDa, pI≈5.2) that was not detected in untreated cells or in the PMS-treated wild-type strain was revealed (Figure 3B). MALDI-TOF MS analysis identified it as Q4IS62_AZOVI, defined as AhpC (alkyl hydroperoxide reductase), a member of the OxyR regulon [3235]. Regulated adaptive responses of micro-organisms to oxidative stress have been extensively studied using E. coli as a model organism [32,36,37], and it is generally recognized that catalase and AhpC activities are critical to oxidative stress survival.

Even though AphC was overexpressed in the rhdA-null mutant as a probable defence mechanism, it did not endow A. vinelandii MV474 with resistance to PMS exposure compared with that of the wild-type strain (Figure 3A). This evidence indicated a role of the rhdA gene product in defending A. vinelandii against oxidative stress, and prompted us to explore whether the RhdA protein underwent structural modifications involving sulfur redox chemistry [3739]. RhdA dimer formation was demonstrated by RhdA immunodetection in UW136 cell extracts exposed to the stressor PMS (Figure 4). Considering that the RhdA Cys230 catalytic thiol, the only one present in the protein [12,40], is mandatory for sulfane/sulfur transfer from thiosulfate to cyanide with concurrent formation of thiocyanate (TST activity) [1], thiocyanate production [23] represents a measure of the availability of the Cys230 thiol. When the wild-type strain UW136 was exposed to PMS, the production of thiocyanate in extracts was 19% of that measured in untreated extracts. Restoration of thiocyanate production to up to 98% of that of untreated cells was obtained by incubating the extracts from UW136 PMS-treated cells in the presence of the thioredoxin/thioredoxin reductase system. These results indicated that reversible modifications of the RhdA Cys230 thiol occurred in A. vinelandii under oxidative conditions.

SDS/PAGE immunoblot

Figure 4
SDS/PAGE immunoblot

Proteins from cell extracts (40 μg of protein) either from UW136 after 16 h exposure to the stressor PMS (+PMS) or from untreated UW136 (−PMS) were resolved by non-reducing SDS/PAGE and subjected to Western blot analysis with an anti-RhdA antibody [13]. Asterisks indicate the dimeric (*) and monomeric (**) forms of RhdA respectively. The molecular masses are indicated on the left-hand side (in kDa).

Figure 4
SDS/PAGE immunoblot

Proteins from cell extracts (40 μg of protein) either from UW136 after 16 h exposure to the stressor PMS (+PMS) or from untreated UW136 (−PMS) were resolved by non-reducing SDS/PAGE and subjected to Western blot analysis with an anti-RhdA antibody [13]. Asterisks indicate the dimeric (*) and monomeric (**) forms of RhdA respectively. The molecular masses are indicated on the left-hand side (in kDa).

In order to investigate how RhdA itself could behave as a redox switch, we undertook an in vitro study to demonstrate RhdA molecular modifications. Incubation of purified RhdA overexpressed in E. coli [24] with PMS resulted in a time-dependent loss of its ability to produce thiocyanate (Figure 5A).

Oxidative events on RhdA

Figure 5
Oxidative events on RhdA

Purified RhdA overexpressed in E. coli (100 μM) in 50 mM Tris/HCl (pH 8.0) and 100 mM NaCl was incubated aerobically at room temperature with PMS (final concentration of 1 mM). At fixed times of PMS exposure, samples were withdrawn and analysed in terms of their ability to produce thiocyanate (TST activity), and by electrophoresis and MS. (A) Time-dependent inactivation of RhdA by PMS, indicated as the percentage activity in relation to untreated RhdA (0 min). (B) Non-reducing SDS/PAGE of the samples at the indicated times of exposure to PMS. Asterisks indicate the dimeric (*) and monomeric (**) forms of RhdA respectively. Molecular masses are indicated on the left-hand side (in kDa). (C and D) Dimer to monomer ratios of RhdA before (0 min) and after PMS treatment for various incubation times determined either by densitometric analyses of electrophoretic patterns (grey bars) or MALDI-TOF MS analyses (black bars). In (D), the values are normalized against those before treatment (0 min).

Figure 5
Oxidative events on RhdA

Purified RhdA overexpressed in E. coli (100 μM) in 50 mM Tris/HCl (pH 8.0) and 100 mM NaCl was incubated aerobically at room temperature with PMS (final concentration of 1 mM). At fixed times of PMS exposure, samples were withdrawn and analysed in terms of their ability to produce thiocyanate (TST activity), and by electrophoresis and MS. (A) Time-dependent inactivation of RhdA by PMS, indicated as the percentage activity in relation to untreated RhdA (0 min). (B) Non-reducing SDS/PAGE of the samples at the indicated times of exposure to PMS. Asterisks indicate the dimeric (*) and monomeric (**) forms of RhdA respectively. Molecular masses are indicated on the left-hand side (in kDa). (C and D) Dimer to monomer ratios of RhdA before (0 min) and after PMS treatment for various incubation times determined either by densitometric analyses of electrophoretic patterns (grey bars) or MALDI-TOF MS analyses (black bars). In (D), the values are normalized against those before treatment (0 min).

The effects of exposure to PMS on RhdA were concurrently analysed by non-reducing SDS/PAGE and MALDI-TOF MS. Non-reducing SDS/PAGE (Figure 5B), followed by accurate densitometric analysis of the resulting protein pattern (Figures 5C and 5D, grey bars), revealed that exposure to PMS for up to 20 min increased the dimer to monomer ratio of RhdA with regard to the ratio in untreated samples. As shown in Figures 5(B) and 5(C), the presence of the RhdA dimeric form decreased when PMS treatment was prolonged, thus indicating that RhdA inactivation during PMS treatment (Figure 5A) could not only be ascribed to dimer formation. Further details on RhdA modifications that affect Cys230 thiol availability were provided by MALDI-TOF MS. As shown in Figure 5(C) and 5(D) (black bars), MALDI-TOF MS analysis of the dimer to monomer ratios of RhdA during PMS treatment confirmed that formation of RhdA dimer was inhibited by prolonged exposure to PMS. A molecular mass of 31024.5 m/z, corresponding to the protein containing unmodified Cys230 (calculated molecular mass of 31027.9 Da), was measured by MALDI-TOF MS analyses of untreated RhdA (Figure 6A). When treated with PMS, a second peak with a molecular mass of 31056.7 m/z was detected (Figure 6B), suggesting the formation of a sulfinic acid (Cys230–SOOH) (+32 Da) [38]. When the experiments were repeated in the presence of an equimolar ratio of RhdA/PMS, MALDI-TOF MS analysis of the sample after 60 min exposure to PMS led to the detection of the peak corresponding to the unmodified protein, and a second peak with an increase in the molecular mass of 16 Da (results not shown), suggesting the presence of sulfenic acid (Cys230–SOH). This latter peak disappeared after treatment with 2-mercaptoethanol as a reductant, and the TST activity of PMS-treated RhdA was recovered by up to 60% of that of the untreated sample by incubation in the presence of the thioredoxin/thioredoxin reductase system.

MALDI-TOF MS analysis of RhdA before and after oxidation by PMS

Figure 6
MALDI-TOF MS analysis of RhdA before and after oxidation by PMS

Mass spectra of RhdA before (A) and after 20 min treatment with PMS (1:10 molar ratio of RhdA/PMS) (B). Spectra were recorded in positive linear mode. a.u., absorbance unit; Intens., intensity.

Figure 6
MALDI-TOF MS analysis of RhdA before and after oxidation by PMS

Mass spectra of RhdA before (A) and after 20 min treatment with PMS (1:10 molar ratio of RhdA/PMS) (B). Spectra were recorded in positive linear mode. a.u., absorbance unit; Intens., intensity.

DISCUSSION

Among the ten ORFs (open reading frames) of A. vinelandii coding for rhodanese-domain proteins displaying different architecture and active-site motifs (Pfam accession number PF00581, Azotobacter vinelandii AvOP), the rhdA gene [11] codes for a tandem-domain rhodanese (RhdA) with an active-site loop [12] which provides a strong positive electrostatic field around the catalytic residue Cys230. Preliminary phenotypic characterization of the A. vinelandii strain disrupted in the rhdA gene [13] suggested that RhdA could exert some protection that allowed the cell enzymes containing Fe–S clusters, which are prone to oxidation, to function [15]. Consistent with this hypothesis, in the present study we found that the phenotypic features resulting from the absence of RhdA were elicited using growth conditions that strictly requires enzymes containing labile Fe–S clusters as catalysts. In the presence of gluconate as a carbon source, which implies the functioning of 4Fe–4S 6-phosphogluconate dehydratase [16,17], A. vinelandii growth was impaired only in the rhdA-null mutant.

Under this growth condition, we found that in the rhdA-null mutant aconitate hydratase 1, which in E. coli is induced during oxidative stress [31], was overexpressed, and that the activity of the hydrogen peroxide scavenger catalase was enhanced. The presence in A. vinelandii of catalase enzymes which are inducible by reactive oxygen intermediates produced during respiration was reported recently [41]. As already found in a sucrose-cultured rhdA-null mutant strain [13], PHB accumulation, generally considered to be a cell-survival mechanism [28,29], was increased in the rhdA mutant strain. The overall picture confirmed that deletion of the rhdA gene makes A. vinelandii more prone to stress, and the increased vulnerability of dehydratase Fe–S clusters in the rhdA mutant was further evidence proving that in the absence of RhdA, protection from oxidative events was impaired. There is a general consensus that oxidants convert the dehydratase [4Fe–4S]2+ form into an unstable [4Fe–4S]3+ state, which releases the catalytic iron molecule [15,42], an event that eventually induces the inactivation of the enzymes. Re-activation of the damaged enzymes in vitro by treatment with iron and DTT proved that inactivation involves damage to the cluster [42]. The results shown in Table 2, combined with the analyses of polypeptide expression, indicated that the low catalytic efficiency of dehydratases in the rhdA-null mutant was the result of oxidative damage of the Fe–S cluster. To overcome oxidative damage resulting from the oxidizing potential of O2, eukaryotic and prokaryotic organisms have evolved strategies to remove reactive oxygen species and repair damage [36,37]. The aerobe A. vinelandii was the focus of studies devoted to investigate the mechanisms of protection of the nitrogenase complex from oxygen inactivation [43], but knowledge of the physiological responses to the generation of oxidative species under non-diazotrophic conditions (as examined in the present study) is lacking. When oxidative stress was provoked by the external addition of chemical oxidants (the superoxide generator PMS), we found that the sensitivity of A. vinelandii was particularly pronounced only in the rhdA mutant strain. Overexpression of AhpC as an adaptive response to oxidative stress [3237] did not endow the rhdA mutant with PMS resistance comparable with that of A. vinelandii wild-type. In E. coli and in other bacteria, AhpC is a member of the OxyR regulon, which up-regulates peroxide defences [34,36]. In E. coli, the model organism of studies on OxyR-mediated adaptive responses, it was found that OxyR can respond to disulfide stress resulting from defects in the systems that maintain an intracellular reducing environment [44], and that the OxyR-regulated ahpC transcript was induced by PMS [45]. Although OxyR-regulated responses in A. vinelandii are unknown, the presence in A. vinelandii chromosome of a gene coding for a protein homologous with OxyR (gi67155513, http://www.ncbi.nlm.nih.gov) could be taken as an indication that similar adaptive mechanisms exist in A. vinelandii. The picture that has emerged from the present study, which aimed to define the physiological functions of the rhodanese-like RhdA, indicated that this protein might function as an antioxidant that helps A. vinelandii in ‘scavenging’ oxidative species. Elimination of oxidative species does imply that RhdA can function as redox switch via thiol modification, an ability that might be inferred by the structural peculiarity of the environment of its catalytic cysteine residue [12] that can promote ionization, even at neutral pH, to the thiolate anion (Cys–S). It is generally recognized [3739] that thiolate anions are easily oxidized to the unstable sulfenic acid form (Cys–SOH), which may react with an accessible thiol to form a disulfide (S–S) bond. The decreased availability of the RhdA Cys230 thiol in extracts from PMS-treated A. vinelandii UW136, proved by RhdA activity tests, and the reversibility of the PMS effects in the presence of the thioredoxin/thioredoxin reductase reducing system, provided evidence that RhdA Cys230 thiol modifications occurred in vivo. Molecular details of RhdA Cys230 thiol modifications (the only one present in the molecule) were explored by using an artificial situation. Formation of the RhdA dimeric form upon exposure for a short period to PMS, and further oxidation producing the stable sulfinic acid species [38,39] was demonstrated by combining activity tests, electrophoresis and MALDI-TOF MS analyses. Also, in the artificial situation of these in vitro studies, in the presence of the thioredoxin/thioredoxin reductase reducing system, RhdA Cys230 thiol modifications were reduced back to the active thiol form, which is mandatory for thiocyanate production. Although to study oxidative modifications of RhdA we used an in vitro approach, the RhdA functionality as a cellular switch is supported by the in vivo evidence that, in the absence of RhdA, A. vinelandii needed to activate defensive activities generally recognized as adaptive responses to a cellular oxidative imbalance [36,37].

We thank Professor Pasquale Ferranti (Dipartimento di Scienza degli Alimenti, Università di Napoli “Federico II”, Portici, Napoli, Italy) for his invaluable advice on the identification of protein spots, and Dr Simona Nonnis (Department of Animal Pathology, Hygiene and Veterinary Public Health, Sezione di Biochimica, Università di Milano, Milano, Italy) for providing skilful assistance in performing MALDI-TOF MS analyses.

Abbreviations

     
  • AhpC

    alkyl hydroperoxide reductase

  •  
  • 2-DE

    two-dimensional PAGE

  •  
  • DTT

    dithiothreitol

  •  
  • ED

    Entner–Doudoroff

  •  
  • MALDI-TOF

    matrix-assisted laser-desorption ionization–time-of-flight

  •  
  • PHB

    polyhydroxybutyrate

  •  
  • PMS

    phenazine methosulfate

  •  
  • TST

    thiosulfate–cyanide sulfurtransferase

  •  
  • %V

    relative volume

FUNDING

This work was supported by the Università degli Studi di Milano [grant number Unimi FIRST 2006, 2007] (to S. P.).

References

References
1
Westley
J.
Adler
H.
Westley
I.
Nishida
C.
The sulfurtransferases
Fundam. Appl. Toxicol.
1983
, vol. 
3
 (pg. 
377
-
382
)
2
Bordo
D.
Bork
P.
The rhodanese/Cdc25 phosphatase superfamily
EMBO Rep.
2002
, vol. 
3
 (pg. 
741
-
746
)
3
Nandi
L. D.
Westley
J.
Reduced thioredoxin as a sulfur-acceptor substrate for rhodanese
Int. J. Biochem. Cell Biol.
1998
, vol. 
30
 (pg. 
973
-
977
)
4
Ray
W. K.
Zeng
G.
Potters
M. B.
Mansuri
A. M.
Larson
T. J.
Characterization of a 12-kilodalton rhodanese encoded by glpE of Escherichia coli and its interaction with thioredoxin
J. Bacteriol.
2000
, vol. 
182
 (pg. 
2277
-
2284
)
5
Williams
R. A. M.
Kelly
S. M.
Mottram
J. C.
Coombs
G. H.
3-Mercaptopyruvate sulfurtransferase of Leishmania contains an unusual C-terminal extension and is involved in thioredoxin and antioxidant metabolism
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
1480
-
1486
)
6
Nagahara
N.
Katayama
A.
Post-translational regulation of mercaptopyruvate sulfurtransferase via a low redox potential cysteine-sulfenate in the maintenance of redox homeostasis
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
34569
-
34576
)
7
Nagahara
N.
Yoshii
T.
Abe
Y.
Matsumura
T.
Thioredoxin-dependent enzymatic activation of mercaptopyruvate sulfurtransferase: an intersubunit disulfide bond serves as redox switch for activation
J. Biol. Chem.
2007
, vol. 
282
 (pg. 
1561
-
1569
)
8
Santos
P. M.
Benndorf
D.
Sá-Correia
I.
Insights into Pseudomonas putida KT2440 response to phenol-induced stress by quantitative proteomics
Proteomics
2004
, vol. 
4
 (pg. 
2640
-
2652
)
9
Krivobok
S.
Kuony
S.
Meyer
C.
Louwagie
M.
Willison
J. C.
Juoanneau
Y.
Identification of pyrene-induced proteins in Mycobacterium sp. strain 6PY1: evidence for two ring-hydroxylating dioxygenases
J. Bacteriol.
2003
, vol. 
185
 (pg. 
3828
-
3841
)
10
Cavalca
L.
Guerrieri
N.
Colombo
M.
Pagani
S.
Andreoni
V.
Enzymatic and genetic profiles in environmental strains grown on polycyclic aromatic hydrocarbons
Antonie Van Leeuwenhoek
2007
, vol. 
91
 (pg. 
315
-
325
)
11
Colnaghi
R.
Pagani
S.
Kennedy
C.
Drummond
M.
Cloning, sequence analysis and overexpression of the rhodanese gene of Azotobacter vinelandii
Eur. J. Biochem.
1996
, vol. 
236
 (pg. 
240
-
248
)
12
Bordo
D.
Deriu
D.
Colnaghi
R.
Carpen
A.
Pagani
S.
Bolognesi
M.
The crystal structure of a sulfurtransferase from Azotobacter vinelandii highlights the evolutionary relationship between rhodanese and phosphatase enzymes family
J. Mol. Biol.
2000
, vol. 
298
 (pg. 
691
-
704
)
13
Cereda
A.
Carpen
A.
Picariello
G.
Iriti
M.
Faoro
F.
Ferranti
P.
Pagani
S.
Effects of the deficiency of the rhodanese-like protein RhdA in Azotobacter vinelandii
FEBS Lett.
2007
, vol. 
581
 (pg. 
1625
-
1630
)
14
Keyer
K.
Imlay
J. A.
Inactivation of dehydratase 4Fe-4S clusters and disruption of iron homeostasis upon cell exposure to peroxynitrite
J. Biol. Chem.
1997
, vol. 
272
 (pg. 
27652
-
27659
)
15
Imlay
J. A.
Iron-sulphur clusters and the problem with oxygen
Mol. Microbiol.
2006
, vol. 
59
 (pg. 
1073
-
1082
)
16
Peekhaus
N.
Conway
T.
What's for dinner?: Entner-Doudoroff metabolism in Escherichia coli
J. Bacteriol.
1998
, vol. 
180
 (pg. 
3485
-
3502
)
17
Gardner
P. R.
Fridovich
I.
Superoxide sensitivity of the Escherichia coli 6-phosphogluconate dehydratase
J. Biol. Chem.
1991
, vol. 
266
 (pg. 
1478
-
1483
)
18
Blum
H.
Beler
H.
Gross
H. J.
Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels
Electrophoresis
1987
, vol. 
8
 (pg. 
93
-
99
)
19
Shevchenko
A.
Wolm
M.
Vorm
O.
Mann
M.
Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels
Anal. Chem.
1996
, vol. 
68
 (pg. 
850
-
858
)
20
Varghese
S.
Tang
Y.
Imlay
J. A.
Contrasting sensitivities of Escherichia coli aconitases A and B to oxidation and iron depletion
J. Bacteriol.
2003
, vol. 
185
 (pg. 
221
-
230
)
21
Outten
F. W.
Djaman
O.
Stortz
G.
A suf operon requirement for Fe-S cluster assembly during iron starvation in Escherichia coli
Mol. Microbiol.
2004
, vol. 
52
 (pg. 
861
-
872
)
22
Beers
R. F.
Sizer
I. W.
A spectrophotometric method for measuring the breakdown of hydrogen peroxide by catalase
J. Biol. Chem.
1952
, vol. 
195
 (pg. 
133
-
140
)
23
Sorbo
B. H.
Crystalline rhodanese. Purification and physiochemical examination
Acta Chem. Scand.
1953
, vol. 
7
 (pg. 
1129
-
1136
)
24
Forlani
F.
Carpen
A.
Pagani
S.
Evidence that elongation of the catalytic loop of the Azotobacter vinelandii rhodanese changed selectivity from sulfur- to phosphate-containing substrates
Protein Eng.
2003
, vol. 
16
 (pg. 
515
-
519
)
25
Laemmli
U. K.
Cleavage of structural proteins during the assembly of the head of bacteriophage T4
Nature
1970
, vol. 
227
 (pg. 
680
-
685
)
26
Bradford
M. M.
A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding
Anal. Biochem.
1976
, vol. 
72
 (pg. 
248
-
254
)
27
Law
J. H.
Slepecky
R. A.
Assay of poly-β-hydroxybutyric acid
J. Bacteriol.
1961
, vol. 
82
 (pg. 
33
-
36
)
28
Anderson
A. J.
Dawes
E. A.
Occurrence, metabolism, metabolic role, and industrial uses of bacterial polyhydroxyalkanoates
Microbiol. Rev.
1990
, vol. 
54
 (pg. 
450
-
472
)
29
Ayub
N. D.
Pettinari
M. J.
Ruiz
J. A.
Lopez
N. I.
A polyhydroxybutyrate-producing Pseudomonas sp. isolated from Antarctic environments with high stress resistance
Curr. Microbiol.
2004
, vol. 
49
 (pg. 
170
-
174
)
30
Peralta-Gil
M.
Segura
D.
Guzman
J.
Servin-Gonzalez
L.
Espin
G.
Expression of the Azotobacter vinelandii poly-β-hydroxybutyrate biosynthetic phbBAC operon is driven by two overlapping promoters and is dependent on the transcriptional activator PhbR
J. Bacteriol.
2002
, vol. 
184
 (pg. 
5672
-
5677
)
31
Cunningham
L.
Gruer
M. J.
Guest
J. R.
Transcriptional regulation of the aconitase genes (acnA and acnB) of Escherichia coli
Microbiology
1997
, vol. 
143
 (pg. 
3795
-
3805
)
32
Zheng
M.
Wamg
X.
Templeton
L. J.
Smulski
D. R.
LaRossa
R. A.
Storz
G.
DNA microarray-mediated transcriptional profiling of the Escherichia coli response to hydrogen peroxide
J. Bacteriol.
2001
, vol. 
183
 (pg. 
4562
-
4570
)
33
Ochsner
U. A.
Maj
M.
Nicholls
P.
Loewen
P.
Role of the Pseudomonas aeruginosa oxyR-recG operon in oxidative stress defense and DNA repair: OxyR-dependent regulation of the katB-ankB,ahpb and ahpC-ahpF
J. Bacteriol.
2000
, vol. 
182
 (pg. 
4533
-
4544
)
34
Hishinuma
S.
Yuki
M.
Fujimura
M.
Fukumori
F.
OxyR regulated the expression of two major catalases, KatA and KatB, along with peroxiredoxin, AhpC in Pseudomonas putida
Environ. Microbiol.
2006
, vol. 
8
 (pg. 
2115
-
2124
)
35
Costa Seaver
L.
Imlay
J. A.
Alkyl hydroperoxide reductase is the primary scavenger of endogenous peroxide in Escherichia coli
J. Bacteriol.
2001
, vol. 
183
 (pg. 
7173
-
7181
)
36
Pomposiello
P. J.
Demple
B.
Redox-operated genetic switches: the SoxR and OxyR transcription factors
Trends Biotechnol.
2001
, vol. 
19
 (pg. 
109
-
114
)
37
Kiley
P. J.
Storz
G.
Exploiting thiol modifications
PLoS Biol.
2004
, vol. 
2
 pg. 
e400
 
38
Jacob
C.
Knight
I.
Winyard
P. G.
Aspects of the biological redox chemistry of cysteine: from simple redox responses to sophisticated signalling pathways
Biol. Chem.
2006
, vol. 
387
 (pg. 
1385
-
1397
)
39
Giles
G. I.
Jacob
C.
Reactive sulphur species: an emerging concept in oxidative stress
Biol. Chem.
2002
, vol. 
383
 (pg. 
375
-
388
)
40
Bordo
D.
Forlani
F.
Spallarossa
A.
Colnaghi
R.
Carpen
A.
Bolognesi
M.
Pagani
S.
A persulfurated cysteine promotes active-site reactivity in Azotobacter vinelandii rhodanese
Biol. Chem.
2001
, vol. 
382
 (pg. 
1245
-
1252
)
41
Sandercock
J. R.
Page
W. J.
RpoS expression and the general stress response in Azotobacter vinelandii during carbon and nitrogen diauxic shifts
J. Bacteriol.
2008
, vol. 
190
 (pg. 
946
-
953
)
42
Flint
D. H.
Tuminello
J. F.
Emptage
M. H.
The inactivation of Fe-S cluster containing hydro-lyases by superoxide
J. Biol. Chem.
1993
, vol. 
268
 (pg. 
22369
-
22376
)
43
Oelze
J.
Respiratory protection of nitrogenase in Azotobacter species: is a widely held hypothesis unequivocally supported by experimental evidence?
FEMS Microbiol. Rev.
2000
, vol. 
24
 (pg. 
321
-
333
)
44
Aslund
F.
Zheng
M.
Beckwith
J.
Storz
G.
Regulation of the OxyR transcription factor by hydrogen peroxide and the cellular thiol-disulfide status
Proc. Natl. Acad. Sci. U.S.A.
1999
, vol. 
96
 (pg. 
6161
-
6165
)
45
Lee
J. H.
Yeo
W. S.
Roe
J. H.
Induction of the SufA operon encoding Fe-S assembly proteins by superoxide generators and hydrogen peroxide: involvement of OxyR, IHF and an unidentified oxidant-responsive factor
Mol. Microbiol.
2004
, vol. 
51
 (pg. 
1745
-
1755
)