Previous Article | Next Article ![]()
Applied and Environmental Microbiology, January 2004, p. 413-420, Vol. 70, No. 1
0099-2240/04/$08.00+0 DOI: 10.1128/AEM.70.1.413-420.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Joseph M. Suflita, Michael J. McInerney, and Lee R. Krumholz*
Institute for Energy and the Environment and the Department of Botany and Microbiology, University of Oklahoma, Norman, Oklahoma 73019
Received 26 August 2003/ Accepted 21 October 2003
|
|
|---|
|
|
|---|
The most intensely studied of these carriers in Desulfovibrio vulgaris are cytochromes (53, 59), with the foremost being the periplasmic cytochrome c3 (31, 44). The latter is used as a biomarker for Desulfovibrio spp. (31). Similarities that exist between cytochrome c3 from different strains include heme 4 interacting with the hydrogenase for electron transfer (1, 5, 15, 40), as well as an N-terminal amino acid sequence extension for periplasm export (1, 15, 19, 61, 62).
Recently, a periplasmic-facing but membrane-associated cytochrome c3 was found in D. vulgaris Hildenborough (58) with similar E0 values and reduced minus oxidized spectrum to the soluble protein (7). Rather than being regarded as a general electron carrier, this protein is speculated to function with the high-molecular-weight complex (hmc) as a membrane-bound oxidoreductase complex, since they are part of the same operon. The periplasmic cytochrome c3, on the other hand, functions in the reduction of a variety of electron acceptors, such as O2 (11, 12, 18, 28), Cr(VI) (24), Fe(III) (22), and U(VI) (27).
We are specifically interested in how sulfate-reducing bacteria reduce U(VI) to U(IV). While these microorganisms can tolerate up to 24 mM uranium (22), most species cannot grow using it as the sole electron acceptor (56). Exceptions are D. vulgaris UFZ B 490 (41) and Desulfotomaculum reducens (54). Reports of the kinetic parameters for U(VI) reduction show enhanced reduction rates in the presence of sulfate by a sulfate-reducing enrichment culture (specific activity of 23 µM min-1 mg of cells-1; apparent Km of 250 µM) (49, 50), while Desulfovibrio desulfuricans showed no such stimulation (specific activity of 30 µM min-1 mg of cells-1; Km of 500 µM) (25, 50). In both cases, sulfate and uranium were reduced concomitantly. Whereas D. desulfuricans showed zero-order kinetics for sulfide production with or without U(VI), loss of the latter was first-order (49), indicating that U reduction is not a typical enzymatic reaction.
Given the ability to reduce numerous electron acceptors, along with first-order U-reduction kinetics, this cytochrome may be more than a simple electron carrier. Here, we report for the first time that cytochrome c3 is likely involved in sulfate reduction from lactate and pyruvate, but not from H2, in D. vulgaris. Further, Fe(III) and U(VI) completely inhibit lactate- and pyruvate-mediated sulfate reduction, but only slow this process when H2 is the electron donor. By using washed cells in reduction assays and cytochrome spectra, we propose that electrons derived from organic donor oxidation may flow through two or more pathways, one of which involves cytochrome c3, while the latter is not implicated in H2-dependent sulfate reduction. Finally, we present a modified electron transport pathway, for which both the present work and previous reports agree.
|
|
|---|
Cultures of wild-type D. vulgaris Hildenborough (ATCC 29579) and its hyd (42) and hmc (14, 17, 43) mutants (a gift from G. Voordouw) were grown with the medium previously described (35) at 37°C. The hyd (lacking the Fe-only hydrogenase) and hmc mutants were grown in the presence of 10 µg of chloramphenicol/ml. Cultures (1 liter) were grown with either 20 mM Na-lactate or H2 (80% H2-20% CO2 headspace) as the electron donor and Na-sulfate (20 mM) as the electron acceptor, or fermentatively with 20 mM Na-pyruvate. All cultures were grown in 2-liter Pyrex bottles sealed with a rubber stopper that was fitted with a serum tube and a serum stopper. Once grown (A660 of
0.80; late log phase), cultures were dispensed into centrifuge bottles inside of an anaerobic chamber (Coy Industries, Ann Arbor, Mich.), sealed, and then centrifuged (27,000 x g, 20 min, 10°C). Cells used for experiments with lactate or pyruvate as the electron donor were washed three times by centrifugation as above, resuspending the pellet in 30 mM NaHCO3 buffer (pH 6.8, under an 80% N2-20% CO2 gas phase) (27). After washing, the pellet was resuspended in 15 ml of the buffer per liter of culture to yield a cell concentration of 8 to 8.75 mg of protein/ml of buffer. When H2 was the electron donor, cysteine-HCl (0.25 g liter-1; 1.6 mM) was added to the buffer, as it is required to facilitate H2-mediated sulfate reduction (25). Cysteine-HCl was not added to the buffer when either lactate or pyruvate was tested.
Modeling of U(VI) and Fe(III) concentrations.
The major species of U(VI) and Fe(III) in our experiments were determined using Phreeqci (version 2.7) (http://typhoon.mines.edu/zipfiles/phreeqc.htm) with the Lawrence Livermore National Laboratory database. All chemical components added to the medium were incorporated into the model to calculate the chemical form of electron acceptor present. The model was also used to argue that the electron acceptors remained in solution. Concentrations of 2 mM each of U(VI) and Fe(III) were used to model the reduction experiments.
Electron acceptor reduction assays.
All manipulations were conducted in an anaerobic chamber except for electron acceptor measurement. Experiments were conducted in duplicate.
The reduction of sulfate, Fe(III), and U(VI) by washed cell suspensions was followed over time, and progress curves were constructed to determine the kinetic parameters and to characterize the ability of the organism to use individual, or combinations of, electron acceptors. Each assay was performed in 160-ml serum bottles sealed with rubber serum stoppers. Each serum bottle contained 100 ml (final volume) of 30 mM NaHCO3 buffer (pH 6.8) with an 80% N2-20% CO2 gas phase, except when hydrogen was the electron donor, where the gas phase was 80% H2-20% CO2. When hydrogen was the electron donor, the buffer contained 1.6 mM cysteine · HCl. U(VI) (as uranyl carbonate), FeCl3, or Na2SO4 was added from filter-sterilized, anaerobic stock solutions prepared in the 30 mM NaHCO3 buffer. Uranyl-carbonate was prepared (26) and quantified by using a kinetic phosphorescence analyzer (model KPA-11; Chemchek Instruments, Inc.) as described previously (4, 30). The reaction was started by the addition of 8 ml of the washed cell suspension to give a final protein concentration of 0.65 to 0.71 mg of protein per ml. For all assays, cells were grown in medium with the same electron donor as was used in the assay. For example, if the H2-mediated loss of sulfate, Fe(III), or U(VI) was studied, the cells were grown with H2-CO2 and sulfate. In order to preclude H2 phase-transfer limitations and to ensure that electron acceptors remained in solution, the cell density was kept low (as indicated above) and the reaction mixture was continuously stirred. Stir bars were placed in serum bottles which were then flushed with N2-CO2, sealed, and autoclaved (121°C, 20 min).
After addition of the washed cell suspension, the loss of the electron acceptor was followed over time by removing periodically 0.2 ml of the cell suspension. The sample was injected immediately into 0.2 ml of anaerobic 1.2 M HCl and centrifuged (13,000 x g, 10 min, 20°C). The supernatant was removed for metal or sulfate quantitation. The loss of U(VI) was monitored as described above. Sulfate concentrations were measured by using a Dionex DX-500 ion chromatograph with an AS-4A anion column as previously described (57). Levels of Fe(II) and Fe(III) were determined by a colorimetric method (23). Lactate was measured using a Dionex DX-500 chromatography system with an AS-11A anion column (48). Protein was determined by adding 0.2 ml of cell suspension to 0.2 ml of 1 N NaOH and incubating at 37°C for 4 h. The protein content of these cell preparations was quantitated using the bicinchoninic protein assay (Pierce Chemical Co.).
In order to ensure that cysteine did not affect metal reduction rates, control assays without cells were also performed as above for each metal, with cysteine present in the buffer. Also, assays with cells suspended in the cysteine-containing buffer and without the addition of an electron acceptor were conducted to confirm that cysteine was not metabolized under the conditions of our assays.
Kinetics of Fe(III), sulfate, and U(VI) reduction.
The kinetics of sulfate reduction were determined by incubating cell suspensions prepared as described above at concentrations of 0.2, 0.5, 1.0, and 2.0 mM Na2SO4 with 20 mM Na-lactate and measuring sulfate loss over time. The apparent Km and Vmax associated with sulfate reduction were determined by calculating the initial reduction rate and fitting this rate to the Michaelis-Menten equation. Sulfate reduction kinetics were also determined by monitoring sulfate loss over the entire time course and using the integrated Michaelis-Menten equation (47). The Michaelis-Menten equation was then fitted to the progress curve data using Matlab software, where the apparent Km and Vmax were again calculated. In each case, the two methods were in good agreement with the latter results presented here. Each experiment was performed twice with duplicate bottles. Fe(III) reduction kinetics were determined in a similar manner with a starting concentration of 2 mM Fe(III) and using the integrated Michaelis-Menten equation (47).
In order to determine the kinetic parameters of U(VI) reduction, assays were carried out as described above at initial U(VI)-carbonate concentrations ranging from 0.05 to 5 mM, with 20 mM lactate as electron donor. Samples were removed every 20 min for the first 6 h and periodically thereafter until U(VI) reduction ceased. Kinetic parameters were then calculated by using both initial rate measurements as well as the integrated Michaelis-Menten equation (47).
Competition for electron acceptors.
Competition assays were performed to see if the wild-type strain of D. vulgaris could reduce U(VI) or Fe(III) in the presence of sulfate. If sulfate loss was somehow inhibited while metal reduction occurred, this would argue that a common electron transfer protein was involved in both sulfate and metal reduction. Each assay was carried out (as described above) with 20 mM lactate (80% N2-20% CO2) or H2 (80% H2-20% CO2) as the electron donor, with 2 mM each of Fe(III) and U(VI), Fe(III) and sulfate, or U(VI) and sulfate. Positive controls used 2 mM Fe(III), U(VI), or sulfate alone, and assays with heat-killed cells served as negative controls. The assays were sampled over time as described above.
Inhibition of sulfate reduction versus U(VI) or Fe(III) reduction.
Molybdate has often been shown to inhibit sulfate reduction, with the proposed site of inhibition being the adenosine triphosphate sulfurylase (36). In order to determine if this enzyme may be involved in U(VI) and/or Fe(III) reduction, 20 mM Na-molybdate was added to reduction assay mixtures with 2 or 20 mM Na2SO4, 2 mM uranyl-carbonate, or 2 mM FeCl3. Assays were performed as described above using the wild-type strain of D. vulgaris with 20 mM Na-lactate as the electron donor.
Effect of periplasm removal on electron acceptor reduction.
The proposed U(VI) reductase, and most likely the Fe(III) reductase (27), is the periplasmic cytochrome c3. The periplasm of lactate and sulfate-grown, wild-type D. vulgaris cells was removed (2) to determine if periplasmic components were required for metal and sulfate reduction. The cells were resuspended in 15 ml of original culture liter-1 in 30 mM NaHCO3, pH 6.8. Cells were inspected microscopically to ensure that cell lysis had not taken place.
Lactate dehydrogenase activity (a cytoplasmic enzyme) of treated cells demonstrated that the cytoplasm was intact after this treatment. Treated cells were assayed as described above with 20 mM Na-lactate as the electron donor and 2 mM methyl viologen as electron acceptor. Lactate was measured as described above. Methyl viologen reduction was measured spectrophotometrically at 560 nm (46, 63). The effectiveness of this treatment in removal of the periplasm was determined by the loss of the cytochrome c3 spectrum from the treated cell suspension (see below).
Cytochrome spectra.
The oxidation of the periplasmic c-type cytochrome via sulfate, U(VI), or Fe(III) was determined for all three strains of D. vulgaris by using a modified reduced minus oxidized spectral method (29). Sample cuvettes were constructed with 12- by 75-mm glass test tubes, sealed with 00 butyl rubber stoppers, and flushed with N2-CO2 (80:20) or H2-CO2 (80:20) (10). A spectrophotometer (Beckman DU64; Beckman Coulter, Inc.) scanned the assay mixture from 400 to 650 nm at 500 nm/min. To generate the cytochrome spectra, we first added 0.2 ml of the washed cell suspension (about 1.7 mg of protein) to 1.7 ml of sterile 30 mM NaHCO3 buffer. Cysteine was present in the buffer if H2 was used as the electron donor (see above). A reductant, Na-dithionite, 40 µM lactate, pyruvate under N2-CO2 (80:20), or a headspace of H2-CO2 (80:20) was then added. The cells were allowed to incubate in the dark (1 h), and the cuvette was scanned to obtain the reduced spectrum. Prior to electron acceptor addition, the H2-CO2 (80:20) gas phase (if used) was aseptically exchanged by venting with sterile needles to N2-CO2 (80:20) to limit electron generation and facilitate cytochrome oxidation. Then, 0.1 ml of 25 mM Na2SO4, uranyl-carbonate, or FeCl3 (2 mM final concentration) was added, and the spectrum was recorded immediately and then periodically thereafter for 2 h. The reduced spectrum was used as the baseline scan by the spectrophotometer software, yielding an inverted spectrum which was reversed for presentation in Fig. 4. A positive control used H2-reduced cell suspensions where the stopper was removed after incubation (1 h) and replaced with sterile cotton so that the cytochrome was oxidized by exposure to air. Negative controls used either no electron donor in the presence of all three electron acceptors or H2-reduced cell suspensions amended with 0.1 ml of assay buffer instead of electron acceptor.
![]() View larger version (19K): [in a new window] |
FIG. 4. Production of sulfide monitored over time with concomitant cytochrome spectra. Sulfidogenesis was found when lactate ( ) (A), pyruvate () (B), and H2 with cysteine ( ) (C), but not H2 without cysteine ( ), were used to reduce the cytochrome. Inserts show the spectrum obtained after 30 min from lactate and pyruvate and after 2 h from H2 with and without the addition of cysteine.
|
|
|
|---|
Modeling of U(VI) and Fe(III) concentrations.
Computer (Phreeqci) modeling of solutions of 2 mM uranyl (UO22+) or FeCl3 in 30 mM NaHCO3, pH 6.8, indicated two major species of each electron acceptor. In the case of uranyl, UO2(CO3)34- (1.079 mM) and UO2(CO3)22- (0.787 mM) accounted for 93.3% of the U(VI) added. For Fe(III), Fe(OH)3 (1.448 mM, likely as a mineral based on the saturation index) and Fe(OH)2+ (0.550 mM) made up greater than 99% of the Fe(III) present. These analyses indicated that rather than several species being involved, two uranyl species and Fe(OH)2+ were the predominant soluble forms. Since Fe(III) was completely reduced to Fe(II) in the reduction assays (see Fig. 2), it is likely that the Fe(OH)3 dissolves slowly as the equilibrium of the reaction is altered by microbial action.
![]() View larger version (24K): [in a new window] |
FIG. 2. Experimental results in which washed cell suspensions of D. vulgaris were incubated in the presence of competing electron acceptors. (A and B) Lactate was used as the electron donor with U(VI) ( ) and sulfate ( ) (both at 2 mM) as electron acceptors (A) or with Fe(III) ( ) and U(VI) ( ) as electron acceptors (B). (C and D) H2 was the electron donor with U(VI) ( ) and sulfate ( ) or sulfate alone ( ) (C), or with Fe(III) ( ) and U(VI) ( ) (D).
|
![]() View larger version (24K): [in a new window] |
FIG. 1. Comparison of the kinetics of sulfate (A), Fe(III) (B), and U(VI) (C) reduction by washed cell suspensions of wild-type D. vulgaris. The inserts are semilogarithmic plots of the respective reduction processes for sulfate (2 mM), U(VI) (0.05 to 5.0 mM), and Fe(III) (0.2 to 1.0 mM). The solid line in panel A is a simulation of the sulfate depletion curve based on the Michaelis-Menten equation, using the derived Km and Vmax values and an initial starting concentration of 2 mM. First-order decay curves (broken lines) modeled in panels B and C are based on the measured decay coefficient and the given starting concentration. The data shown are from a single progress curve and are not an average of duplicate data.
|
|
View this table: [in a new window] |
TABLE 1. Comparison of initial electron acceptor reduction rates when grown and assayed with H2, lactate, or pyruvate as electron donor
|
Inhibition of sulfate reduction versus U(VI) and Fe(III) reduction.
The addition of 20 mM Na-molybdate inhibited lactate-mediated sulfate reduction with both 2 mM (Fig. 3) and 20 mM (data not shown) Na2SO4 as expected. However, the presence of molybdate did not influence Fe(III) (data not shown) or U(VI) reduction (Fig. 3). This indicated that electron transfer to Fe(III) and U(VI) does not involve components that are inhibited by molybdate.
![]() View larger version (17K): [in a new window] |
FIG. 3. Graph showing the reduction of sulfate (triangles) and U(VI) (diamonds) by washed cell suspensions of wild-type D. vulgaris. Sulfate reduction was completely inhibited by 20 mM molybdate ( ) with transient inhibition of sulfate reduction by the presence of 2 mM U(VI) ( ). U(VI) reduction in the presence of sulfate was not inhibited by molybdate ( ) compared to U(VI) reduction in the absence of molybdate ( ).
|
Cytochrome c3 spectra.
The difference spectra for all three strains of D. vulgaris showed a Soret peak at 419 nm and
and ß peaks at 553 and 525 nm, respectively, which is consistent with cytochrome c3 as the major c-type cytochrome in the cell (31, 55). Reduction of the cytochrome using dithionite, H2-CO2, lactate, and pyruvate was successful after a 1-h incubation, as reduced minus oxidized spectra were detected when the cuvettes were subsequently exposed to air. The addition of 2 mM Fe(III), U(VI), or sulfate caused immediate oxidation of the cytochrome when reduced with lactate, pyruvate, or dithionite, confirming the involvement of this protein in electron transfer from organic donors. The coupling of H2 with the electron acceptors was not as straightforward. The addition of 2 mM Fe(III) or U(VI) to H2-reduced cells caused an immediate oxidation of the cytochrome, but oxidation was not observed with sulfate addition even after a 2-h incubation (Fig. 4). However, the subsequent addition of 2 mM Fe(III) or U(VI) gave an immediate difference spectrum, suggesting that the cytochrome remained in the reduced state while sulfate reduction proceeded (data not shown). We then carried out sulfide production assays to be certain that sulfate was reduced when it was present as the oxidant. In all cases, sulfate reduction occurred as determined by the production of sulfide (Fig. 4). These data suggest that the predominant c-type cytochrome involved in pyruvate- or lactate-coupled sulfate reduction (presumably cytochrome c3) is not involved in H2-mediated sulfate reduction.
|
|
|---|
A comparison of the Soret,
, and ß peaks from our cytochrome spectra with the spectral properties of various c-type cytochromes implicates cytochrome c3 involvement (44) rather than hmc (6), cytochrome c553 (59), or cytochrome c7 (53). Additional evidence that the spectrum is not due to the hmc is provided by the observation that the hmc-deficient mutant showed a cytochrome spectrum similar to that of the wild type. Our metal-dependent spectra provided further evidence that cytochrome c3 is intimately involved in U(VI) (27) and Fe(III) (22) reduction. This cytochrome was also implicated in O2 reduction in sulfate-reducing bacteria (11, 12) where sulfidogenesis was interrupted during the time that O2 was being reduced (28). Our results are similar in that Fe(III) and U(VI) must be reduced before sulfate reduction will occur. Thus, it could be suggested that the periplasmic cytochrome c3 functions in a protective role (9) whereby oxidized compounds with higher redox potentials than that of sulfate (-220 mV) are immediately reduced to prevent cell oxidation. This could imply that electron transfer to the cytochrome is a dead-end pathway in the absence of alternate electron acceptors but, since it plays an essential role in pyruvate and sulfate reduction (45), its function as a central electron reservoir (13, 33) is more likely. Consistent with a central role for cytochrome c3, we found that the pyruvate- or lactate-reduced c-type cytochrome is oxidized by Fe(III), U(VI), and sulfate, suggesting a common pathway for metal and sulfate reduction from organic electron donors. However, the use of H2 as an electron donor showed that Fe(III) still inhibited U(VI) reduction, but sulfate loss was slowed instead of being halted by U(VI). These results suggest that cytochrome c3 may not be directly involved in H2-mediated sulfate reduction. The presence of a difference spectrum during H2-mediated Fe(III) or U(VI) reduction suggests that some electrons from H2 are used to reduce a c-type cytochrome, presumably cytochrome c3. The fact that no spectrum was observed during H2-mediated sulfate reduction but was present after the addition of a metal provides additional evidence that sulfate reduction from H2 does not involve this cytochrome. It could be argued that the difference spectrum was produced by oxidation of cytochrome c3 to a steady-state redox potential and that the H2 and sulfate redox state was more reduced than the H2 and U(VI) or Fe(III) state. We believe that this explanation is unlikely, as a spectrum was observed after oxidizing lactate-reduced cells with sulfate.
Hence, given the reports cited along with the present results, we propose the pathway shown in Fig. 5. D. vulgaris contains several periplasmic hydrogenases (Fe-only, NiFe, and NiFeSe) that could be involved in H2 oxidation (60). To account for the lack of c-type cytochrome spectrum during H2-mediated sulfate reduction, electrons from the Fe-only hydrogenase are likely donated to the hmc complex without cytochrome c3 participation. Further evidence for this pathway is the suppression of H2-mediated, but not lactate- or pyruvate-mediated, sulfate reduction with either the hyd (Fe-only hydrogenase mutant) or hmc mutants. As metal reduction rates from H2 are increased in the hyd mutant, it is likely that electrons needed for H2-mediated Fe(III) or U(VI) reduction could be supplied through an alternative enzyme, such as one of the NiFe hydrogenases (13, 33). This interpretation is consistent with the reported decreased ability of the hyd and hmc mutants to carry out H2-mediated sulfate reduction (14, 42, 60).
![]() View larger version (28K): [in a new window] |
FIG. 5. Proposed model for the electron transfer pathways for sulfate reduction from H2 versus that from lactate and pyruvate. Solid black lines represent currently proposed electron flow. Grey broken lines indicate the formation of H2 from lactate or pyruvate via cytochrome c3 and NiFe hydrogenase. H2ase = hydrogenase; hmc = high-molecular-weight complex.
|
Electron flow from lactate oxidation also results in H2 production (42), but it does not always involve cytochrome c3, since its deletion had no effect on growth or sulfate reduction in the presence of lactate (45). However, Fe(III) and U(VI) inhibited lactate-mediated sulfate reduction in our assays, and cytochrome c3 reduced with lactate was oxidized with Fe(III), U(VI), and sulfate. Hence, we propose that electrons derived from lactate oxidation may flow through two or more pathways, one of which involves cytochrome c3 (42).
We thank Gerrit Voordow and Judy Wall for providing mutant strains and for their critiques of the manuscript and Matthew Ramsey for helpful discussion.
Present address: Biological Sciences Division, Pacific Northwest National Laboratory, Richland, WA 99352. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»