INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Technetium (⁹⁹Tc) is a fission product of ²³⁵U formed during the generation of nuclear power. The solubility and mobility of Tc are highly dependent upon its redox state. Under oxic conditions, Tc is present in its most stable form, the pertechnetate anion [Tc(VII)O₄]. This form, which is highly soluble and mobile in the environment, can enter the food chain as a sulfate analogue (2, 20, 34). These properties, coupled with its long half-life (2.13 × 10⁵ years), make contamination by Tc one of the major factors in the long-term impact of the nuclear fuel cycle. One approach to remove ⁹⁹Tc from aqueous solution is to reduce the pertechnetate form Tc(VII) into the insoluble, low-valence form Tc(IV). This can be achieved by abiotic (5) or biotic (19, 20) processes.

Abiotic reduction involves electron transfer between Fe(II)-containing minerals and Tc(VII) (3). The most efficient mineral appears to be magnetite, particularly when it is anodically polarized (5). Biotic precipitation of pertechnetate, probably its reduction into a low-valence, insoluble Tc oxide, has been reported for several species of bacteria during the past few years. These include species such as Geobacter metallireducens (14), Geobacter sulfurreducens (17), Escherichia coli (13), Desulfovibrio desulfuricans (16), Shewanella putrefaciens (33), and Deinococcus radiodurans (9). Both indirect (chemical) and direct (enzymatic) reduction processes have been observed, depending on the bacterial growth conditions. Chemical processes have been clearly demonstrated in the case of the sulfate-reducing bacterium D. desulfuricans and in the case of the metal-reducing bacterium G. sulfurreducens (15, 17). Cultures of D. desulfuricans supplied with sulfate and lactate as electron acceptor and donor, respectively, precipitated Tc extracellularly as an insoluble sulfide. In this case, the Tc sulfide results from the chemical reaction between H₂S, formed during reduction of sulfate, and TcO₄ (19). A chemical reduction of Tc(VII) by the Fe(II) is also observed during reduction of Fe(III) by G. sulfurreducens (17). In addition, enzymatic reduction of Tc(VII) has been reported for different bacterial species. There are several lines of evidence indicating that this process involves hydrogenase, an enzyme which reversibly catalyzes the splitting of molecular hydrogen into protons and electrons. Indeed, hydrogen is an effective electron donor for Tc(VII) reduction for E. coli, D. desulfuricans, and S. putrefaciens (13, 15, 16, 33). Similarly, in the case of G. sulfurreducens, Tc(VII) reduction has an exclusive requirement for hydrogen as the electron donor, unlike the reduction of Fe(III), which can be coupled to the oxidation of different organic electron donors (17). The most convincing evidence has come from the work of Lloyd et al. (13), who showed that mutants of E. coli defective in the synthesis of transcription factor FNR, of molybdenum cofactor, or of formate dehydrogenase H were unable to reduce Tc(VII), indicating a role for the formate-hydrogenlyase complex in the reduction.

Desulfovibrio fructosovorans (24) is a sulfate-reducing bacterium amenable to molecular biological study. Three different hydrogenases have been identified in this bacterium: [NiFe] and [Fe] hydrogenases (1, 10, 27), localized in the periplasm, and a heterotetrameric NADP-reducing [Fe] hydrogenase, localized in the cytoplasm (4, 21, 22). In the present work, we have combined physiological, genetic, and biochemical approaches to determine, at the molecular level, the exact roles of these different hydrogenases in Tc(VII) reduction and precipitation.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Growth of organisms and preparation of extracts. The wild-type strain D. fructosovorans DSM 3604 (24), D. fructosovorans strain MR400 carrying a deletion in the nickel-iron hydrogenase operon (26), or strain MR400 complemented with the nickel-iron hydrogenase genes (28) was grown for 72 h at 37°C in stoppered 100-ml bottles supplemented, when required, with kanamycin and gentamicin (50 µg/ml) in a minimal medium defined by Widdel and Pfennig (32). For Tc(VII) reduction assay, the strains were subcultured three times in a medium containing 20 mM fructose as an electron donor and 20 mM fumarate as an electron acceptor.

Tc(VII) reduction by resting cell suspensions or purified proteins. Bacteria grown for 72 h at 37°C were transferred anaerobically in a centrifuge tube stoppered with a rubber septum (Suba seal no. 37; Aldrich) and washed four times in 50 mM Tris-HCl buffer (pH 8.0). The bacterial pellet was resuspended anaerobically in either Tris-HCl (20 mM, pH 8.0 or 8.5), MES (morpholineethanesulfonic acid) (20 mM, pH 5.5), MOPS (morpholinepropanesulfonic acid) (20 mM, pH 6.5 or 7.5), or citrate-sodium phosphate buffer (20 mM, pH 4.5) to a concentration of about 0.5 mg of cells (dry weight) per ml. Aliquots (1.9 ml) of the washed cell suspension were transferred under nitrogen to 10-ml serum bottles sealed with butyl rubber stoppers. Electron donors (fructose, fumarate, lactate, pyruvate, or formate) were added from concentrated stock solutions to a final concentration of 10 mM. For these experiments, all of the bottles were depleted of oxygen by three cycles of vacuum-nitrogen and then flushed under nitrogen for 10 min. When hydrogen was supplied as an electron donor for metal reduction, the gas was flushed into the headspace of the bottles for 20 min with resting cell suspensions or for 180 min with soluble extracts or purified proteins in order to activate the nickel-iron hydrogenase, as described by Fernandez et al. (6) and Hatchikian et al. (10). A solution (100 µl) of ammonium pertechnetate (NH₄TcO₄) (Amersham Life Science Products, Orsay, France, and NEN Life Science Products, Paris, France), deaerated by flushing with argon 10 min before use, was added to a final concentration of 1 mM for cell suspensions. Concentrations of 250 µM to 6 mM were used for K_m and V_max determination, and a concentration of 0.5 mM was used for reduction by soluble extracts or by purified proteins.

Measurements of Tc. Total Tc in solution was assayed by autoradiography with a STORM 840 PhosphorImager (Molecular Dynamics) as described by Lloyd and Macaskie (14). Tc uptake was expressed as the percentage or the concentration of Tc remaining in solution after centrifugation in an Eppendorf 5415C centrifuge (14,000 rpm, 20 min) in comparison with total Tc. Tc(VII) (R_f = 0.7) was also separated from reduced, nonmobile [Tc(V); R_f = 0.0] and mobile [mainly Tc(IV); R_f = 0.9] soluble Tc species using paper chromatography (29) prior to autoradiography and quantification using a PhosphorImager. In most experiments, concentrations of Tc were also quantified using a Packard 1900 TR analyzer. Each sample (10 µl) was added to a glass scintillation vial with 10 ml of Ultima Gold or Insta-Gel-Plus scintillation fluid (Packard Instrument S. A., Rungis, France). Disintegration counts per minute were recorded at between 20 and 250 keV for 5 min.

Purification of cytochrome c₃ and [NiFe] hydrogenase. Pure cytochrome c₃ and [NiFe] hydrogenase were obtained in four steps, and all operations were performed at pH 7.6. In the first step, the soluble fraction was loaded on a DEAE-Sepharose (Pharmacia) column. This column retains all of the hydrogenase but not the cytochrome c₃. The fraction containing cytochrome c₃ was subsequently loaded on an SP-Sepharose (Pharmacia) column equilibrated with Tris-HCl (10 mM, pH 7.6) and eluted at 200 mM NaCl. The eluted fraction was then concentrated in a 15-ml Centriprep YM-10 (Centrifugal Filter Device; Amicon) and filtered through a Sephacryl S-200 high-resolution column (Pharmacia). Finally, cytochrome was loaded on a hydroxylapatite (Bio-Rad) column and eluted at 200 mM potassium phosphate (pH 7.6). Purified cytochrome exhibited a broad single band of 16.5 kDa in sodium dodecyl sulfate-15% polyacrylamide gel electrophoresis and a purity index (A_553reduced  A_570reduced/A_280oxidized) of 3.02.

Analytical procedures. Protein concentrations were measured with a bicinchoninic acid assay kit (Pierce) by the method of Smith et al. (30), using bovine serum albumin as a standard. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was carried out by the method of Laemmli (11). The hydrogen uptake activity was visualized in the gel after native 7.5% polyacrylamide gel electrophoresis as previously described (4). UV-visible spectra of the pure nickel-iron hydrogenase and the cytochrome c₃ were recorded on a Varian spectrophotometer. The hydrogen uptake activity was measured spectrophotometrically by monitoring the reduction of methyl viologen (6) at 30°C in a tightly closed quartz cuvette filled with 1 ml of reaction buffer (50 mM Tris-HCl [pH 8], 1 mM methyl viologen) bubbled for 20 min with H₂. The proton-deuterium (H-D) exchange reaction was measured in whole-cell suspensions by a mass spectrometric method as described previously (31). Production of H₂ and HD was used to calculate the exchange activity.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Microbial reduction of technetium: physiology and kinetic parameters. In order to study exclusively the enzymatic contribution to Tc(VII) reduction by sulfate-reducing bacteria, chemical Tc(VII) reduction was prevented by growing D. fructosovorans in a mineral medium in the absence of sulfate. Cells grown under these conditions had a greyish color with no black precipitate of iron sulfide. These conditions were preferred to those used by Lloyd et al. (15) for D. desulfuricans, which did not completely abolish the formation of iron sulfide in the case of D. fructosovorans (data not shown).

View larger version (17K): [in this window] [in a new window]   FIG. 1.   (A) Tc(VII) reduction and precipitation by resting cells of D. fructosovorans supplied with hydrogen (closed circles) as an electron donor. Control cultures with nitrogen (open circles) contained no added electron donor. The cells were incubated for 24 h at 23°C and pH 8.0. (B) Effect of pH on Tc(VII) reduction by D. fructosovorans. Hydrogen was supplied as the electron donor, and the cells were incubated at room temperature (23°C) and pH 4.5 (open circles), 5.5 (closed triangles), 6.5 (open squares), 7.5 (open triangles), 8.0 (closed circles), or 8.5 (inverted closed triangles).

Inhibitors and genetic determinants. The Tc(VII) reduction activity of D. fructosovorans was not inhibited by exposure of the cells to air, which indicated oxygen tolerance of the enzyme (Fig. 2). Moreover, this activity was irreversibly inhibited by a 10-min preincubation of the cells with 0.5 mM CuCl₂, a specific inhibitor of the periplasmic hydrogen uptake activity in vivo (8) (Fig. 2). These results excluded a possible role of NADP-reducing hydrogenase in the reduction mechanism, as this enzyme is cytoplasmic and oxygen sensitive (4, 22). To investigate the role of the [NiFe] hydrogenase, which is periplasmic and oxygen tolerant and is the major hydrogenase produced by D. fructosovorans (10), Tc(VII) reduction by strain MR400, which carries a specific deletion of the structural genes of this hydrogenase (26), was studied. This deletion strain showed a dramatic decrease in the rate of reduction (Fig. 2): only 20% of the Tc(VII) was reduced in 3 h, although iron-only hydrogenase activity still represented about 16% of the wild-type level in both the hydrogen uptake and deuterium-hydrogen exchange activities (data not shown). Moreover, the complementation of strain MR400 by the nickel-iron hydrogenase genes carried on a multicopy plasmid (28) restored hydrogenase activity (hydrogen uptake and deuterium exchange) (data not shown) and Tc(VII) reductase activity (Fig. 2) to wild-type levels. These results demonstrated the essential role of the nickel-iron hydrogenase in the in vivo reduction of Tc(VII) by the sulfate-reducing bacterium D. fructosovorans.

View larger version (17K): [in this window] [in a new window]   FIG. 2.   Effect of hydrogenase inhibitors or hydrogenase contents on Tc(VII) reduction in D. fructosovorans. Cells were incubated at 23°C and pH 5.5 with hydrogen supplied as an electron donor. Open circles, wild-type strain (control); closed circles, wild-type strain after preincubation with air; closed squares, wild-type strain after preincubation with 0.5 mM CuCl2; closed triangles, mutant MR400 with the nickel-iron hydrogenase genes deleted; open triangles, complemented mutant MR400.

Reduction and precipitation of Tc(VII) by soluble extracts and purified proteins. In order to provide biochemical evidence of the involvement of the [NiFe] hydrogenase and possibly other electron carriers in the biological reduction of Tc(VII), this activity was tested in different fractions: (i) crude extract (soluble and membranes proteins) and (ii) soluble fraction (supernatant from centrifugation at 120,000 × g) suspended in MOPS (20 mM, pH 6.5). In both cases, 85% of the Tc(VII) was reduced and precipitated as shown by the appearance of brownish particles after 18 h of incubation in the presence of hydrogen. No reduction or precipitation occurred when hydrogen was omitted. This experiment demonstrates that reduction of Tc(VII) is performed in vitro by soluble proteins. This further shows that the precipitation of reduced forms of Tc does not necessitate the presence of membranes as nucleation sites. On the other hand, the precipitation process is highly dependent on the buffer used. Indeed, when Tris-HCl (50 mM, pH 8.0) was used, 80% of the Tc(VII) (R_f = 0.7) was reduced to Tc(V) (R_f = 0) in the first 24 h as shown by paper chromatography (Fig. 3) (29), but no precipitation occurred. This result indicates that Tc(VII) reduction mediated by the soluble fraction is not obligately followed by precipitation. The precipitation process seems to be dependent mainly on the chemical nature of the buffer used and the pH. This behavior is different from that observed in vivo, where the precipitation of Tc is independent of the buffer used (Fig. 1B).

View larger version (94K): [in this window] [in a new window]   FIG. 3.   Chromatograms of Tc(VII) reduction by soluble fractions of D. fructosovorans at pH 8.0. Soluble proteins were incubated at 23°C and pH 8.0 (20 mM Tris-HCl) with hydrogen as an electron donor and 0.5 mM Tc(VII). Tc(VII), Rf = 0.7; Tc(V), Rf = 0.0.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In the present work, we report that D. fructosovorans reduces Tc(VII) and removes it efficiently from solution. The reduction process occurs at wide ranges of temperature (10 to 40°C) and pH (5.5 to 8.0). The optimum pH (around pH 5.5) probably reflects the best affinity of hydrogenase for TcO₄. Reduction of Tc(VII) with increasing Tc concentrations gave an apparent K_m of 2 mM, which is slightly higher than but consistent with the K_m of 0.5 mM determined by Lloyd et al. (18) with E. coli and D. desulfuricans supplied with formate as an electron donor. At a high concentration of TcO₄ (6 mM), D. fructosovorans exhibits the highest rate of reduction described so far, i.e., 7 mmol of Tc reduced per g (dry weight) of bacteria per h. The corresponding values for E. coli and D. desulfuricans were 12.5 and 800 µmol of Tc reduced per g (dry weight) of bacteria per h, respectively (18). The high efficiency of the Tc(VII)-reducing activity of D. fructosovorans within wide ranges of pH and temperature makes this bacterium a good candidate for the removal of this radionuclide from solution in a bioremediation process.

The Tc(VII)-reducing activity of D. fructosovorans requires the presence of hydrogen as an electron donor, whereas organic electron donors such as lactate, pyruvate, fumarate, fructose, and formate are inefficient (Table 1). The essential role of the nickel-iron hydrogenase in the process of reduction of Tc(VII) was further proved by genetic and biochemical studies. This role is supported by (i) in vivo and in vitro inhibition of the activity by Cu(II), (ii) oxygen tolerance of the activity, (iii) the dramatic decrease of Tc(VII) reduction in a mutant lacking [NiFe] hydrogenase structural genes, and (iv) demonstration of direct reduction by purified [NiFe] hydrogenase. This is the first report which demonstrates the reduction and removal of Tc(VII) from solution with purified nickel-iron hydrogenase. Reduction of Tc(VII) at the expense of molecular hydrogen is the most efficient and most widespread mechanism in the bacteria tested so far. Even though alternative electron donors can be used by several species, such as E. coli (13), S. putrefaciens (14, 33), and D. desulfuricans (16), the oxidation of these organic substrates often leads to the production of hydrogen or formate. Therefore, hydrogenase (alone or in a formate-hydrogenlyase complex) appears to be the major component involved in enzymatic Tc(VII) reduction.

In addition to the ability to reduce Tc(VII), selenite- and chromate-reducing activities have been reported for the [Fe] hydrogenase of Clostridium pasteurianum (35) and the [Fe] hydrogenase of sulfate-reducing bacteria (23), respectively. Metal reductase activity of hydrogenases therefore appears to be a widespread property.