(Received for publication, May 5, 1997, and in revised form, June 18, 1997)
From the Department of Biochemistry and Molecular Biology, Wayne State University, School of Medicine, Detroit, Michigan 48201
ArsC encoded by Escherichia coli
plasmid R773 catalyzes the reduction of arsenate to arsenite. The
enzymatic reaction requires reduced glutathione and glutaredoxin. In
this study a direct association between ArsC and glutaredoxin was
demonstrated. An arsC gene with six histidine codons added
to the 5 end of the gene was constructed, and the resulting ArsC
enyzme was shown to be functional. Interaction of the histidine-tagged
ArsC and glutaredoxin was examined by Ni2+ affinity
chromatography. The association required the presence of reduced
glutathione and either the substrate arsenate or a competitive
inhibitor, phosphate or sulfate. A free thiolate on glutathione was not
required. A tryptophan residue was introduced into ArsC at the 11th
position, immediately adjacent to the active site Cys-12. Trp-11
fluorescence was quenched upon addition of arsenate. Addition of
reduced glutathione after arsenate resulted in a rapid increase in
fluorescence followed by a slower decay of the signal. These
spectroscopic signals were specific for arsenate and reduced
glutathione; neither competitive inhibitors nor non-thiol glutathione
analogs produced this effect. Cys-12 thiolate was also required. Thus
the intrinsic fluorescence of Trp-11 provides a useful probe to
investigate the mechanism of this novel reductase.
ArsC, an arsenate reductase encoded by the Escherichia coli R-factor R773, provides arsenate resistance by reducing As(V) (arsenate) to As(III) (arsenite) (1), which is then rapidly extruded from cells by the ArsA-ArsB anion-translocating ATPase (2). In vitro arsenate reduction requires coupling of ArsC, reduced glutathione (GSH) and glutaredoxin (Grx1)1 (3). Grx1 is a small intracellular protein that functions as a GSH-dependent hydrogen donor for ribonucleotide reductase and other general GSH-disulfide reducing reactions (4-6). Oxidized Grx1 is reduced by GSH. The resulting GSSG is reduced by glutathione reductase, with NADPH serving as the source of reducing potential. In vivo an E. coli mutant lacking grx1 still reduced arsenate to arsenite (1). However, there are multiple genes for glutaredoxins in E. coli (7), and at least one gene product, Grx3, can substitute for Grx1 in vitro,2 likely accounting for the difference between the in vitro and in vivo results.
The reaction scheme for the pathway of arsenate reduction is
proposed to have at least two steps. First, ArsC catalyzes arsenate reduction, becoming itself oxidized. Second, Grx1 reduces oxidized ArsC, regenerating the enzyme. This premise implies physical
interaction of ArsC and Grx1. In this report direct association of the
two proteins was examined by the addition of six histidine codons to
the 5 end of arsC. This recombinant gene encodes an ArsC
with an amino-terminal histidine tag (His6ArsC). His6ArsC bound to a
Ni2+ affinity column, allowing the interaction with Grx1 to
be visualized. In the presence of substrates of the reaction, elution
of Grx1 from the affinity column was retarded, and Grx1 and His6ArsC
coeluted.
The role of GSH in the ArsC-catalyzed reaction was investigated. In E. coli GSH exists predominantly in the reduced form in millimolar concentrations (6, 8), functioning as a redox buffer against oxidants and xenobiotics (8). The question arises as to whether GSH participates directly in the ArsC reaction or is required only for reduction of oxidized Grx1. Native ArsC has no tryptophan residues. To examine this question a tryptophan codon was introduced into arsC, encoding an A11W ArsC. Trp-11 is adjacent to Cys-12, an essential residue postulated to transfer electrons from ArsC to arsenate (9). The intrinsic fluorescence of Trp-11 was used as a fluorescence reporter of ligand interactions. Arsenate quenched the fluorescence of Trp-11. GSH alone had no effect, but adding GSH after arsenate produced a rapid increase in fluorescence which decayed slowly. These results suggest that the first step in the reaction is the binding of arsenate, followed by the interaction of the enzyme-arsenate complex with GSH. A reaction scheme is hypothesized in which the enzyme forms a mixed disulfide between the Cys-12 thiolate of ArsC and GSH. Grx1 would then be required to resolve the mixed disulfide, regenerating reduced ArsC.
Strains and plasmids
used in this study are given in Table I.
For protein production, cells were grown at 37 °C with shaking in
Luria-Bertani medium (10). For phenotypic measurements of arsenate
resistance, E. coli strain AW10
(ars::cam) (11) carrying the indicated plasmids
were grown in a low phosphate medium containing 3 mM sodium
arsenate (1). In this strain the chromosomal ars operon was
disrupted and is hypersensitive to arsenate. Turbidity at 600 nm was
measured after 8-12 h of growth at 37 °C with shaking. Appropriate
antibiotics were added as required.
|
All restriction enzymes and nucleic acid-modifying enzymes were obtained from Life Technologies, Inc. The Altered SitesTM in vitro Mutagenesis System was purchased from Promega. Oligonucleotides were synthesized in the Macromolecular Core Facility of Wayne State University School of Medicine. T4 glutaredoxin, thioredoxin reductase, and anti-Grx1 serum were generous gifts from Dr. Fredrik Åslund, Medical Nobel Institute for Biochemistry, Sweden. All other chemicals were obtained from commercial sources.
DNA ManipulationPlasmid isolation, DNA restriction endonuclease analysis, ligation, and transformation were performed as described (10, 12). Mutations in arsC were constructed by site-directed mutagenesis, as described previously (9). Mutagenic oligonucleotides were designed to change the codons for Ala-11 (GCC) to Trp (TGG), Cys-12 (TGC) to Ser (AGC), and Tyr-7 (TAT) to Trp (TGG). Either the wild type arsC or arsCA11W gene inserted into the multiple cloning site of pAlterTM-1 vector (Promega) was used as a template (9). The identity of the mutation was confirmed by DNA sequencing of the entire mutant gene. DNA was prepared using a Qiagen DNA purification system. Sequencing was performed using the internal DNA labeling (Cy5TM-dATP labeling mix) with the T7 DNA polymerase Cy5TM AutoReadTM Sequencing kit and an ALF-express DNA Sequencer (Pharmacia Biotech Inc.).
Construction of the Gene for His6ArsCAn arsC
with six histidine codons at the 5 end was constructed by polymerase
chain reaction cloning. The oligonucleotide primers used for the
polymerase chain reaction were CGCGGATCCTCATATGAGCAACATC and
GCTTCAAGCTTGGTCGTTGGC, respectively. The resulting arsC
polymerase chain reaction product having NdeI and
HindIII sites at the ends was digested with the restriction
enzymes NdeI and HindIII and cloned into
NdeI/HindIII-digested vector plasmid pET-28a
(Novagen). The construction was confirmed by restriction enzyme
digestion and was transformed into E. coli strain BL21 (DE3)
with selection for kanamycin resistance. An arsC encoding
His6C12S was constructed similarly.
Wild type and substituted ArsCs and
Grx1 were purified as described previously (3). For purification of
His6ArsC, E. coli strain BL 21(DE3) bearing the gene for the
His6ArsC was grown in LB medium. At an OD600 = 0.5-0.6,
isopropyl-1-thio--D-galactopyranoside was added to a
final concentration of 0.4 mM as inducer, and the culture
was shaken for an additional 3 h at 37 °C. The cells were washed once with buffer A (20 mM Tris-HCl, pH 7.5, containing 5 mM imidazole, 0.5 M NaCl, and 5 mM
-mercaptoethanol. The cells were suspended in buffer
A at a ratio of 5 ml of buffer/g of cells and lysed by a single passing
through a French pressure cell at 20,000 p.s.i.
Diisopropylfluorophosphate (2.5 µl/g of wet cells) was added to the
lysate immediately after the lysis. The lysate was centrifuged at
100,000 × g for 60 min at 4 °C, the clear
supernatant solution was loaded at a flow rate of 1 ml/min onto a 5-ml
Ni2+-NTA column preequilibrated with 30 ml of buffer A. The
column was then washed with 15 ml of buffer A at a flow rate of 1 ml/min. His6ArsC was eluted with a linear gradient of 60-300
mM imidazole in buffer A. ArsC-containing fractions were
identified by SDS-polyacrylamide gel electrophoresis (13), pooled, and
concentrated using a Millipore Ultrafree-15 BIOMAX-5K centrifugal
filter (Millipore) at 2,000 × g. Purified ArsC was
stored at
80 °C. The concentrations of ArsC enzymes containing
Trp-7 and Trp-11 were determined from the absorbance at 280 nm using a
molar extinction coefficients of 7,953 and 9,150, respectively (14).
The concentrations of other proteins were determined either with a
micromodification of the method of Lowry et al. (15) or
using a Coomassie protein assay (Bio-Rad). The reductase assay for ArsC
was performed as described previously (3). When T4 glutaredoxin was
used in place of E. coli glutaredoxin, the reaction
contained 25 mM MOPS-NaOH, pH 6.6, 0.2 mM
NADPH, 30 µg each of thioredoxin reductase and T4 glutaredoxin, 10 µg of ArsC, and 20 mM sodium arsenate. GSH was added as
indicated.
Purified His6ArsC was dialyzed against buffer A adjusted to pH 6.6. Ni2+-NTA spin columns (Qiagen) of 0.1 ml of equilibrated with the same buffer were loaded with 0.1 mg of His6ArsC protein in a volume of 0.1 ml. The column was centrifuged at 700 × g for 2 min. Purified Grx1 (50 µg) in 0.1 ml was allowed to soak into the resin, and the column was centrifuged again. The column was then washed with 0.6 ml of the buffer A containing 10 mM imidazole to remove nonspecifically bound protein. Finally 0.1 ml of buffer A containing 0.5 M imidazole, pH 8.0, was applied to the column to elute bound His6ArsC. The proteins in the eluates were separated by SDS-polyacrylamide gel electrophoresis on 16% acrylamide gels and analyzed by staining with Coomassie Blue. In the parallel assays His6ArsC was preincubated for 5 min at 37 °C with 40 mM sodium arsenate, 40 mM sodium phosphate, 40 mM sodium sulfate or 10 mM sodium arsenite. Grx1 was preincubated for 5 min at 37 °C with 30 mM GSH or its non-thiol derivatives, S-methylglutathione or S-hexylglutathione, as indicated.
Measurements of FluorescenceFluorescence measurements were performed using an SLM 8000 spectrofluorometer using an excitation wavelength of 295 nm. The slit widths for excitation and emission were 4 mm. Reagents were added with a microliter syringe through a light-protected port to the cuvette containing 2-ml samples, and the solutions were stirred continuously during the measurement. Dilution effects were negligible. Tryptophan fluorescence intensity changes were recorded at 350 nm for A11W and A11W/C12S, 354 nm for free tryptophan, or 330 nm for Y7W. All spectra were corrected for background and Raman scattering by subtracting the buffer signal.
Histidine-tagged ArsC was constructed to contain an additional 20 residues at the amino terminus of the protein, including six histidines and a thrombin recognition site. The modified enzyme is approximately 2 kDa larger than the wild type. It exhibited a wild type rate of arsenate reductase activity, with approximately the same Km and Vmax (data not shown). Cells expressing the mutated gene on a plasmid in trans with the arsAB genes showed normal levels of arsenate resistance (data not shown).
Interaction of His6ArsC and Grx1Approximately 0.3 M imidazole was required to displace His6ArsC from a
Ni2+-NTA affinity column (data not shown). When Grx1 was
loaded onto the column after binding of His6ArsC, none of the Grx1 was
retained (Fig. 1A, lane
3), indicating that there was no interaction between the two
proteins. If Grx1 was preincubated with 30 mM GSH before loading onto a column with bound His6ArsC, no binding of Grx1 was
observed (Fig. 1A, lane 4). Similarly, if
His6ArsC was preincubated with 40 mM sodium arsenate before
loading on the column, no binding of Grx1 was observed (Fig.
1A, lane 5). However, when His6ArsC was
preincubated with sodium arsenate and Grx1 preincubated with GSH, Grx1
coeluted with His6ArsC (Fig. 1A, lane 6). Thus
both arsenate and GSH were required for protein-protein interaction. Immunoblotting with anti-Grx1 serum confirmed that the lower band was
Grx1 (data not shown). Similar results were obtained when the
competitive inhibitor sodium phosphate (Fig. 1A, lane
8) or sodium sulfate (Fig. 1B, lane 12) or
the product, sodium arsenite (Fig. 1B, lane 15),
was substituted for arsenate. In contrast, there was no apparent
interaction when sodium nitrate was substituted for sodium arsenate
(data not shown). These results suggest that filling of an anion site
on ArsC by the substrate arsenate, competitive inhibitors phosphate and
sulfate, or the product arsenite in combination with GSH stabilizes
ArsC-Grx1 interaction.
When non-thiol derivative S-methylglutathione or S-hexylglutathione were substituted for GSH, binding of Grx1 to His6ArsC was decreased (Fig. 1B, lanes 3 and 6). However, the fact that binding still occurred with non-thiol glutathione derivates indicates that the glutathione thiolate is not required for ArsC to interact with Grx1. The ability of the histidine-tagged derivative of the enzymatically inactive C12S ArsC to interact with Grx1 was examined. When both sodium arsenate and GSH were present, binding of Grx1 to His6C12S ArsC was observed (Fig. 1B, lane 9). Thus, interaction of ArsC and Grx1 does not require catalytic activity, nor is the Cys-12 thiolate essential for Grx1 binding.
Role of GSHGSH is required to reduce oxidized Grx1 (5). From
previous results it was not clear whether GSH also played a direct role in the ArsC-catalyzed reaction (3, 9). To investigate this question, T4
glutaredoxin was substituted for E. coli glutaredoxin. Whereas Grx1 requires GSH for reduction, T4 Grx can be reduced directly
by thioredoxin reductase using NADPH as reductant. Using T4 Grx there
was no reduction of arsenate by ArsC (Fig.
2A). However, the addition of
GSH stimulated arsenate reduction (Fig. 2A, curve 1), indicating that GSH participates directly in the ArsC
reaction. Neither dithiothreitol (Fig. 2A, curve
2) nor -mercaptoethanol (Fig. 2A, curve
3) could replace GSH.
If GSH is directly required for the reductase reaction, it may form a mixed disulfide with ArsC. This question was examined using a Grx1 with a C14S substitution (16). This altered glutaredoxin retains Cys-11, the critical residue in catalyzing the reduction of mixed disulfides of glutathione. However, it can no longer serve as a hydrogen donor for ribonucleotide reductase, which requires reduction of intramolecular protein disulfides. The rates of ArsC-catalyzed arsenate reduction were the same with wild type Grx1 (Fig. 2B, curve 1) and C14S Grx1 (Fig. 2B, curve 2). This result is consistent with the function of Grx1 being to reduce a mixed disulfide rather than as a direct hydrogen donor in arsenate reduction. The most likely mixed disulfide would be between the thiolate of Cys-12 in ArsC and glutathione.
Biochemical Properties of Altered ArsC ReductasesArsC contains no tryptophans. To develop potential probes of the mechanism of ArsC, the seventh and eleventh codons of arsC were individually mutated from tyrosine and alanine codons, respectively, to a tryptophan codon. These locations were chosen because of their proximity to the active site Cys-12 residue. E. coli strain AW10 expressing either arsCA11W or arsCY7W in trans with wild type arsA and arsB genes retained wild type arsenate resistance (data not shown). Both A11W and Y7W proteins were produced in amounts similar to that of the wild type, and both were purified. Y7W ArsC had the same Km and Vmax as wild type (data not shown). Purified A11W had the same Km for arsenate, with a Vmax approximately half that of the wild type enzyme (data not shown). The A11W substitution was combined with a C12S substitution. The double-substituted protein was, as expected, catalytically inactive.
Fluorescent Properties of Tryptophan-containing ArsC ReductasesThe emission spectra of both A11W (Fig.
3, curve 1) and A11W/C12S
(Fig. 3, curve 2) had a maximum emission wavelength at 350 nm, whereas the max of Y7W was 330 nm (Fig. 3,
curve 3). Upon denaturation with 6 M guanidine
hydrochloride, the
max of each protein was 354 nm, the
same as that of free tryptophan (Fig. 3, curves 4-7). These
results indicate that Trp-11 is located in a relatively solvent exposed
environment, whereas Trp-7 is in a relatively less polar environment.
A decrease in intrinsic fluorescence was observed upon addition of
sodium arsenate to A11W (Fig.
4A, curve 1). In
contrast, the fluorescence of neither Y7W nor free tryptophan was
affected by addition of sodium arsenate (Fig. 4B,
curves 3 and 4). The addition of a competitive
inhibitor, either sulfate or phosphate, resulted in a slight increase
in A11W fluorescence (Fig. 4A, curves 2 and
3). Neither arsenite, the product of the reaction, nor
chloride, which is neither a substrate nor competitive inhibitor, had a substantial effect on A11W fluorescence (Fig. 4B,
curves 1 and 2).
The addition of GSH alone had no effect on A11W fluorescence (Fig.
5, curve 1). However, if GSH
was added after arsenate, there was a rapid fluorescence enhancement
followed by a decay of the signal (Fig. 4, curve 1, and Fig.
5, curve 2). This synergistic effect of arsenate and GSH was
not observed with Y7W (Fig. 4B, curve 3), nor was
there an effect of GSH after the addition of sulfate (Fig.
4A, curve 2) or phosphate (Fig. 4A,
curve 3). If arsenate was added after GSH, fluorescence
quenching was 10-fold less rapid than when arsenate was added in the
absence of GSH (Fig. 5, curve 1). This reduction in arsenate
effect may be the result of an additive effect of arsenate quenching
and GSH enhancement of fluorescence. It appears, then, that GSH affects
A11W fluorescence only after the substrate arsenate is bound to the
enzyme.
On the other hand, neither the monothiol -mercaptoethanol nor the
dithiol dithiothreitol was able to replace GSH (Fig.
6A). Similarly, non-thiol
analogs and derivatives of glutathione, including ophthalmic acid,
S-methylglutathione, and oxidized GSSG, could not substitute
for GSH (Fig. 6B). These results suggest that the rapid
fluorescence enhancement represents a transient conformational change
in the enzyme that is produced specifically by the combination of
arsenate and the reduced form of glutathione.
The role of the Cys-12 thiolate in the arsenate and glutathione responses was examined using the noncatalytic A11W/C12S enzyme. The intrinsic fluorescence of this double-substituted ArsC was increased slightly by arsenate (Fig. 4C, curve 1) and phosphate (Fig. 4C, curve 2). This response is similar to that of the active A11W enzyme produced by the competitive inhibitor phosphate (Fig. 4A, curve 3). This result may suggest that the initial event reported by Trp-11 is binding of the oxyanion to a site on the enzyme. Because the noncatalytic A11W/C12S does not exhibit fluorescence quenching upon arsenate addition, the decrease in Trp-11 fluorescence may reflect a conformation of the enzyme in a subsequent step of the reaction, perhaps formation of a thiol-As bond. Subsequent addition of GSH after arsenate had a negligible effect on the fluorescence of the inactive enzyme. Thus the glutathione-dependent increase in fluorescence requires prior binding of the oxyanionic substrate, a free thiol on glutathione, and a free thiol on the enzyme, suggesting that the fluorescent signals are reporting events during catalysis. However, since no covalent intermediates have been isolated, there may be alternate explanations for these data.
ArsC encoded by the arsC gene of the ars operon E. coli plasmid R773 is an enzyme that reduces arsenate (As(V)) to arsenite (As(III)). The product, arsenite, is the substrate of the ArsA-ArsB efflux pump that renders the cells resistant to arsenicals and antimonials. We have shown previously that a single cysteine residue, Cys-12, is required for catalysis (9). The reaction also requires glutathione and glutaredoxin, but their exact roles were not known (3). We postulated a model in which the dithiol pair of glutaredoxin was oxidized to a disulfide during arsenate reduction. GSH was assumed to be required for reduction of oxidized Grx1 rather than playing a direct role in ArsC catalysis. In this report we demonstrate 1) direct interaction of His6ArsC and Grx1; 2) that GSH is required for the ArsC-catalyzed reaction as well as for Grx1 reduction; and 3) that Grx1 does not have to be oxidized to the disulfide form and may function as a reductase of a mixed disulfide between the Cys-12 thiol of ArsC and GSH.
Glutaredoxin is a small thioredoxin-like protein that has a broad range of specificity. Grx1 catalyzes electron transfer to ribonucleotide reductase (4) and is involved in the reduction of sulfate (17, 18), ascorbate (19), tellurite (20), and arsenate (3). During oxidative stress protein glutathionylation and deglutathionylation, a process also referred to as S-glutathionylation, is catalyzed by Grx1 (21). The data in this paper are the first biochemical evidence of physical interaction between ArsC and Grx1. A polyHis-ArsC chimeric protein was bound to an immobilized Ni2+ column. Grx1 alone did not bind to the metal column but was retained by association with the His6ArsC. The association of the two proteins depended on the presence of both GSH and the substrate, arsenate. Arsenate could be replaced with a competitive inhibitor, either phosphate or sulfate, and non-thiol derivatives of GSH could substitute for that thiol. Thus catalytic conditions are not required for the protein-protein interaction. An ArsC derivative with a C12S substitution was capable of interacting with Grx1. Although the C12S derivative still has Cys-106, the thiol of this nonessential cysteine residue is buried within the enzyme and does not exhibit reactivity with thiol reagents (9). Thus it is unlikely that ArsC and Grx1 are held together by disulfide bonds.
ArsC has low affinity for arsenate and glutathione, making direct binding measurements difficult. To examine ligand binding to ArsC intrinsic protein fluorescence was used. Native ArsC has no tryptophan. This allowed placement of tryptophan residues in locations of potential utility by site-directed mutagenesis of arsC. Tryptophan codons were introduced at positions 7 and 11, both of which are close to the essential residue Cys-12 (9). Both proteins were purified, and the effect of ligands on intrinsic tryptophan fluorescence was examined. The Y7W enzyme had normal activity, and there was no effect of substrate, product, competitive inhibitors, or thiol reagents on its fluorescence. Although this protein was of no value in defining ligand interactions, it served as a useful control for the specific ligand-induced changes observed with the A11W enzyme.
Although the data and the fluorescence results described above could be interpreted in a variety of ways, they are consistent with the following reaction scheme.
![]() |
(Reaction 1) |
![]() |
(Reaction 2) |
![]() |
(Reaction 3) |
![]() |
(Reaction 4) |
![]() |
(Reaction 5) |
![]() |
(Reaction 6) |
We thank Dr. Fredrik Åslund, Medical Nobel Institute for Biochemistry, Sweden, for gifts of glutaredoxin proteins and antiserum and for valuable discussions. We appreciate greatly the advice and suggestions of Dr. Bharati Mitra.