Startseite Enzymatic reduction of U60 nanoclusters by Shewanella oneidensis MR-1
Artikel Öffentlich zugänglich

Enzymatic reduction of U60 nanoclusters by Shewanella oneidensis MR-1

  • Qiang Yu EMAIL logo und Jeremy B. Fein
Veröffentlicht/Copyright: 29. Juni 2017

Abstract

In this study, a series of reduction experiments were conducted using a representative uranyl peroxide nanocluster, U60 (K16Li44[UO2(O2)OH]60) and a bacterial species, Shewanella oneidensis MR-1, that is capable of enzymatic U(VI) reduction. U60 was reduced by S. oneidensis in the absence of O2, but the reduction kinetics for U60 were significantly slower than was observed in this study for aqueous uranyl acetate, and were faster than was reported in previous studies for solid phase U(VI). Our results indicate that U60 aggregates bigger than 0.2 μm formed immediately upon mixing with the bacterial growth medium, and that these aggregates were gradually broken down during the process of reduction. Neither reduction nor dissolution of U60 was observed during 72 h of control experiments open to the atmosphere, indicating that the breakdown and dissolution of U60 aggregates is caused by the reduction of U60, and that S. oneidensis is capable of direct reduction of the U(VI) within the U60 nanoclusters, likely due to the adsorption of U60 aggregates onto bacterial cells. This study is first to show the reduction capacity of bacteria for uranyl peroxide nanoclusters, and the results yield a better understanding of the long term fate of uranium in environmental systems in which uranyl peroxide nanoclusters are present.

1 Introduction

The anthropogenic use of uranium (U), such as in mining of U ore and production of nuclear weapons and fuel, has resulted in numerous sites of U contamination of groundwater and sediments [1], [2]. Microbial reduction of soluble U(VI) to insoluble U(IV) under anoxic conditions can affect the mobility of U in the environment [3], [4], [5], and a bacterially-mediated reductive precipitation approach has been viewed as a promising strategy for the remediation of U-contaminated sites [6], [7], [8]. In order to predict the long-term fate of U in these contaminated systems, and to develop a practical approach for accelerating U immobilization, the controls and mechanisms of microbial reduction of the important forms of U(VI) that may be present in these systems must be determined.

Bacterial reduction of various aqueous U(VI) species has been studied extensively in both laboratory and field-scale experiments 4], [9], [10], [11], [12], [13], [14], [15], and a large number of bacterial species are now known to have the ability to reduce U(VI) [16]. Most of previous studies focused on the reduction of aqueous monomeric U(VI) species [4], [9], [13]. In general, these studies demonstrate that reduction can be rapid [4], [9], and is largely controlled by the extent of adsorption of bioavailable U(VI) complexes onto bacterial cell envelopes [13], [17], unless electron shuttles are present [18]. A range of factors, such as solution pH, the presence of aqueous ligands and the specific electron donor used, can strongly affect the reduction kinetics of dissolved U(VI) by bacteria. For example, the presence of Ca2+ can significantly inhibit U(VI) reduction by bacteria [13], [19], likely due to the formation of Ca2+-U(VI)-carbonate complexes that lower the redox potential of the U(VI)/U(IV) couple [19]. On the other hand, increasing aqueous EDTA concentrations enhance U(VI) reduction by S. oneidensis possibly due to U(IV)-EDTA complexation which removes U from the cell surface and keeps binding and reduction sites free on the cell [13]. Only a few studies have examined the reduction of U(VI) in a solid phase [11], [14], [15], [20]. These studies document much slower reduction kinetics compared to reduction of aqueous monomeric U(VI), likely due to the difficulty of transporting electrons from bacterial cells to the solid phase U(VI), and solid phase U(VI) associated with sediments can be inaccessible to microbial reduction [11]. In other cases, bacteria may promote dissolution of the U(VI) solid phase before the U(VI) can be reduced [14], [15].

Although enzymatic reduction of aqueous and solid phase U(VI) has been studied, the ability of bacteria to reduce U(VI) present in nanoclusters in solution is unknown. A wide range of uranyl peroxide nanoclusters have been described over the past 10 years [21], [22]. In general, most uranyl peroxide nanocluster crystals are synthesized by combining uranyl acetate, hydrogen peroxide and other reagents in ultrapure water under slightly to highly basic pH conditions [21]. After placing nanocluster crystals in an unsaturated solution, complete dissolution of the crystals occurs rapidly, liberating isolated nanoclusters that typically have outer diameters of 2–3 nm [21]. Uranyl peroxide nanoclusters can persist in ultrapure water for extended periods. For example, Flynn et al. [23] demonstrate that 97.4% of U is still present in nanoclusters after dissolving U60 nanocluster crystals in ultrapure water for 15 days, and there is qualitative evidence that isolated U60 nanoclusters remain intact in solution for at least several months [24]. While the presence of uranyl peroxide nanoclusters has not yet been documented in natural systems, there is a high potential for them to form at some U-contaminated sites, such as at Fukushima, Japan and Hanford, USA, where all the requirements for nanocluster formation have been met, including the presence of peroxide and high concentrations of dissolved U(VI) in basic solutions [25]. For example, the Fukushima accident led to seawater with a pH of ~8 interacting with exposed solid phase UO2, likely producing significant concentrations of dissolved U(VI) as well as peroxide by radiolysis of seawater [24], and likely resulting in the formation of uranyl peroxide nanoclusters [25]. The structures of approximately 40 nanocluster types have been described, but their behavior in natural systems, for example their bioavailability for microbial reduction, remains largely unexplored. As a hybrid type of U species between solid phase and aqueous monomeric U species, uranyl peroxide nanoclusters are smaller than even colloidal-sized solid phase U species, but their structures are much more complex than any dissolved monomeric U species. They exhibit some properties similar to solid phases, but in other ways are similar to large aqueous complexes [23]. If uranyl peroxide nanoclusters are present in U-contaminated sites, their behavior in the environment will be unpredictable based on our current knowledge.

The primary objective of this study is to test whether uranyl peroxide nanoclusters can be reduced by bacteria that are capable of U(VI) reduction, and to examine the mechanisms responsible for the microbial reduction of these nanoclusters if reduction does occur. Toward this end, we conducted a series of U(VI) reduction experiments using Shewanella oneidensis MR-1 and a representative uranyl peroxide nanocluster, U60 (K16Li44[UO2(O2)OH]60) [26]. U60 was chosen because it is persistent in circumneutral ultrapure water for at least 294 days [24], and its solubility behavior in water has been studied previously [23]. Shewanella oneidensis MR-1 is an extensively-studied Fe(III) reducing strain of bacteria that can reduce a variety of U(VI) species including both monomeric U(VI) species and solid phase crystalline U(VI) [13], [15]. Our results indicate that some of the U(VI) present in U60 can be reduced enzymatically, and we conduct and report on a range of complementary experiments that constrain the aggregation state and stability of the U60 during the reduction process.

2 Materials and methods

2.1 Synthesis of U60 nanoclusters

The synthesis of U60 nanoclusters followed a modified procedure of Sigmon et al. [26]. Briefly, 1 mL of 0.5 M uranyl nitrate hexahydrate, 1 mL of 30% hydrogen peroxide, and 0.1 mL of 0.4 M potassium chloride were added to a 20 mL scintillation vial. The solution pH was adjusted to approximately 9 by adding 0.4 mL of 4 M lithium hydroxide. After 7 days of reaction at room temperature, large, yellow, cubic crystals precipitated from solution and were harvested by vacuum filtration using a Whatman cellulose filter (11 μm pore) and were then rinsed with 18 MΩ ultrapure water. The washed crystals were then dissolved in ultrapure water to form a parent suspension of U60 nanoclusters, as confirmed by electrospray ionization mass spectrometry (ESI-MS) [27].

2.2 Bacterial cell preparation

The procedures for growth and washing of Shewanella oneidensis MR-1 (ATCC#: 700550) cells were similar to those described previously [13]. Initially, a stock of S. oneidensis at −80 °C was transferred to trypticase soy agar (TSA) plates and was cultured at 32 °C for 24 h. A single colony of S. oneidensis was then transferred from TSA plates to 3 mL of trypticase soy broth with 0.5% yeast extract (TSB) and was cultured aerobically at 32 °C for 24 h. The harvested bacterial suspensions were transferred to 1 L of TSB medium and allowed to grow aerobically at 32 °C for another 24 h until early stationary phase was reached, at which point the cells were harvested by centrifugation at 10,970 g for 5 min. The collected biomass pellet was then rinsed twice with ~15 mL of sterile anoxic 0.1 M NaCl solution that was stored in a sealed serum bottle, with each rinse followed by centrifugation at 8100 g for 5 min. Finally, the biomass pellet was transferred into a sterile centrifuge tube with cap and was mixed with ~15 mL of sterile anoxic 0.1 M NaCl solution to make a parent bacterial suspension, the cell density of which was determined by optical density measurement using UV-vis spectrophotometry at 600 nm (OD600). After the UV-vis measurement, the parent bacterial suspension was sealed and was immediately transferred into an anaerobic glovebox for the reduction experiments. The washing of S. oneidensis cells was conducted in the air, but sterile anoxic 0.1 M NaCl solution was used as the washing solution in order to minimize the oxygen that was introduced to the parent bacterial suspensions during the washing. Each reduction experiment was conducted with a freshly prepared parent bacterial suspension.

2.3 U60 aggregation measurements

The bacterial growth medium used in the reduction experiments consisted of 1 mM KCl, 5 mM NH4Cl, 50 mM Na-lactate, 30 mM NaHCO3, with a vitamins and trace elements solution similar to that used by Lovley and Phillips [28]. The pH of the experimental medium was adjusted to 7.0 using small aliquots of 1 M NaOH and/or 1 M HCl. In order to determine the aggregation state of the U60 nanoclusters before the reduction experiments, U60 nanoclusters were added to either ultrapure water or the bacterial growth medium to create U60 suspensions with concentrations ranging from 0 to 700 μM, as determined by fluorometry (see below). The OD600 values of these prepared U60 suspensions were immediately measured using UV-vis, and the concentrations of U(VI) in the filtrates after filtration through a 0.2 μm PTFE membrane were also measured using fluorometry. U60 suspensions in ultrapure water yield an OD600 reading of approximately 0 at any of our tested U60 concentrations, but the OD600 value of the U60 nanoclusters in the bacterial growth medium increases with increasing U60 concentration (Figure S1A). Similarly, although all U60 nanoclusters that are suspended in ultrapure water pass through a 0.2 μm PTFE membrane, yielding no significant change in U(VI) concentration between the unfiltered and filtered samples, when U60 nanoclusters that were suspended in bacteria-free growth medium were filtered through the same type of filter, no U(VI) was detected in the filtrate, indicating that U60 nanoclusters in the growth medium aggregate rapidly to form aggregates larger than 0.2 μm.

The above measurements of the extent of U60 aggregation in the bacterial growth medium with ~300 μM of U60 were extended to 72 h, with each experiment conducted in an anaerobic glovebox with modest shaking. At selected sampling times, approximately 2 mL of suspension was removed from the glovebox for measurement of an OD600 value, and measurement of U(VI) in the filtrate by fluorometry after filtration through a 0.2 μm membrane. During the course of 3 days, the OD600 of a U60 suspension in the growth medium did not change significantly, and no U(VI) was detected after filtration (Figure S1B), indicating that the size of the U60 aggregates remained larger than 0.2 μm in these bacteria-free control experiments after 72 h.

2.4 U(VI) reduction experiments

All of the reduction experiments were conducted in an anaerobic glovebox filled with 5% H2/95% N2 and equipped with a palladium catalyst to maintain an oxygen concentration below 1 ppm. After 30 min of heating and degassing by bubbling a 5% H2/95% N2 mixture through the solution outside the glovebox, the growth medium was transferred into the anaerobic glovebox, and distributed into sterile serum bottles with 50 mL of medium in each bottle. All the serum bottles were then sealed with rubber septa and aluminum crimp caps, and were autoclaved outside the glovebox at 120 °C for 30 min. Prior to the reduction experiments, U(VI) stock solutions were prepared using either uranyl acetate or U60 as U(VI) sources, with 18 MΩ ultrapure water, and the U concentrations of these solutions were determined by inductively coupled plasma optical emission spectroscopy (ICP-OES). The U stock solutions were degassed for 30 min using 5% H2/95% N2, filter-sterilized through a 0.2 μm PTFE membrane filter, and finally injected into the sterile and sealed serum bottles in the glovebox.

In order to compare the reduction kinetics of aqueous monomeric U(VI) and U60, a small amount of uranyl acetate or U60 stock solution was injected into serum bottles containing 50 mL growth medium to achieve an initial U(VI) concentration in each case of ~300 μM. The reaction was initiated by injecting an aliquot of parent bacterial suspension to each serum bottle to achieve the desired cell density (OD600: ~0.6). Two control experiments were conducted in order to test if reduction of U60 by S. oneidensis is an enzymatic reduction. The first control started with ~300 μM of U60, and no biomass was added to the bottle. The second control started with ~300 μM of U60 and ~0.6 OD600 of dead cells of S. oneidensis with the cells killed by autoclaving them at 121 °C for 30 min. At selected sampling times, approximately 2 mL of suspension was removed from each bottle for determination of the U(VI) concentration. In order to measure the potential extent of re-oxidation of reduced U, in selected reduction experiments approximately 1 mL of the removed suspension at each sampling time was exposed to the atmosphere with modest shaking for 12 h, and the final U(VI) concentration was determined using fluorometry. The effect of varying the initial cell density on the U(VI) reduction rate was studied by varying the cell density in the experiments using cell concentrations with OD600 values of 0.6, 1.0 or 1.5, and 300 μM of U60 nanoclusters in each. The effect of varying the U60 concentration was studied using a cell density of 0.6 OD600 and U60 concentrations of 300, 500 or 1000 μM.

In order to probe the aggregation state of U60 nanoclusters during reduction and its effect on U60 reduction by S. oneidensis, a range of different filtrations were conducted on otherwise identical experimental systems. In each of these tests, a cell density of 0.6 OD600 and a U60 concentration of 500 μM was used, and at selected sampling times, approximately 3 mL of suspension was removed from each bottle, and then either was analyzed for U(VI) concentration without filtration, was filtered using a 0.2 μm PTFE membrane to remove U60 aggregates, or was filtered using a 10 kDa molecular weight sieve to remove isolated U60 nanoclusters and U60 nanocluster aggregates as only monomeric U(VI) species can pass through these filters [23]. Before reduction was initiated in each experiment, filtration through a 0.2 μm membrane removed virtually all U in the solution, indicating extensive formation of U60 aggregates greater than 0.2 μm. In order to test whether breakdown or dissolution of the U60 aggregates occurs when reduction does not proceed, a control experiment was conducted open to the atmosphere by shaking a suspension with ~300 μM of U60 and S. oneidensis cells at an OD600 value of 1.0 in the bacterial growth medium for 48 h. At selected sampling times, two 2 mL samples of suspension were removed from the bottle. One of these samples was analyzed for U(VI) concentration using fluorometry without any filtration; the other was filtered using a 0.2 μm PTFE membrane prior to the U(VI) concentration analysis. In all the above reduction experiments, duplicate tests were conducted, and the reported results represent the average values of the duplicate experiments.

2.5 U60 adsorption experiments

Adsorption experiments were conducted by adding U60 nanoclusters and S. oneidensis cells to 50 mL of 0.1 M NaCl to reach an initial U60 concentration of ~500 μM with varying initial cell densities ranging from OD600 values of 0–1.0. All of the experimental systems were allowed to react for 24 h with modest shaking open to the atmosphere. At 0 h and 24 h, approximately 1 mL of the suspension was filtered through 0.2 μm PTFE membranes, and the concentration of U(VI) in each filtrate was determined by fluorometry.

2.6 Analysis of U(VI) concentrations

A PTI Quantamaster QM-4 fluorometer was used to determine the concentration of U(VI) in suspensions by monitoring the phosphorescence decay of U(VI) at 515 nm as a function of time after excitation at 420 nm [29], using the following procedure. The first step in the analysis was to acidify 200 μL of each sample with 100 μL of 12.1 M HCl. Each acidified sample was then diluted 150 times with 18 MΩ ultrapure water, followed by mixing with a U(VI) complexing agent (Uraplex®, Chemcheck Instruments, Inc.) in a ratio of 2:3 to prevent the uranyl ion from quenching. The samples were analyzed immediately after pretreatment. Before analysis of the samples, matrix-matched blanks and standards covering the desired U(VI) concentration range were measured in order to construct a calibration curve, and the determined detection limit of this approach was approximately ~2 ppm. This method is specific for the detection of U(VI) because other valences of U are essentially nonluminescent [29].

In order to test whether all U(VI) that is present in the U60 nanoclusters becomes available to the Uraplex complexing agent and therefore is detected in the subsequent U(VI) analysis, we compared calibration curves that were constructed from standards made from either U60 nanoclusters or from dissolved uranyl acetate. Total U(VI) concentrations in each type of standard were measured both by ICP-OES and using the fluorometry approach described above. The resulting calibration curves for dissolved uranyl acetate and for the U60 nanoclusters were essentially the same (Figure S2A), indicating that all of the U(VI) that is present in the U60 nanoclusters is available to the Uraplex complexing agent and is detected during the U(VI) analysis by fluorometry. The availability of the U(VI) from the U60 nanoclusters is likely due to complete breakdown of the nanoclusters during the acidification step of the analysis. Because both U60 and uranyl acetate standards yield identical calibration curves, in all subsequent U(VI) analyses by fluorometry, we used calibration curves constructed using dissolved uranyl acetate standards.

Because it is difficult to separate U60 aggregates from bacterial cells, we tested whether the presence of bacteria affects the fluorometry analysis of U(VI) concentrations in mixures of U60 and bacterial cells. In this test, different amounts of a parent bacterial suspension were added to 400 μM U60 suspensions in ultrapure water to achieve OD600 bacterial cell densities of 0.29, 0.53 and 0.68, and the U(VI) concentration in each sample was measured using fluorometry both before and after the addition of the bacterial cells. The presence of bacteria does not significantly or consistently affect the concentration of U(VI) as measured by fluorometry (Figure S2B), and thus for all subsequent U(VI) analyses for the experimental samples, we did not filter the samples to remove bacteria before acidification. Filtration would remove not only the bacteria, but also any U60 adsorbed onto the bacteria as well as aggregates of U60 larger than 0.2 μm, thereby yielding inaccurate U(VI) analyses.

3 Results

3.1 Comparison of reduction of U60 and uranyl acetate

The reduction kinetics of U60 nanoclusters and an aqueous monomeric uranium species, uranyl acetate, by S. oneidensis, are presented in Figure 1. Over the course of 72 h, no significant change in the U(VI) concentration of the system was observed in the abiotic controls that contained U60 nanoclusters in growth medium without cells, suggesting that abiotic reduction of U(VI) under the experimental conditions is negligible. Similarly, a control experiment that contained U60 nanoclusters and dead S. oneidensis cells in growth medium for 72 h also did not result in reduction of U(VI). In contrast, the concentration of U(VI) in systems containing U60 nanoclusters in the presence of live S. oneidensis cells continuously decreased over the course of the experiment, with ~60% of U(VI) left in the system after 72 h. After exposure of these suspensions to the atmosphere, all of the reduced U(VI) was oxidized and recovered (Figure 1). These observations demonstrate that the U(VI) in U60 nanoclusters is susceptible to enzymatic reduction by S. oneidensis under the studied conditions. However, the reduction kinetics of U(VI) in U60 nanoclusters is significantly slower than that of U(VI) in the uranyl acetate experiment. For example, after 24 h ~80% of the U(VI) was reduced in the uranyl acetate experiment, but only ~20% of the U(VI) in the U60 experiment was reduced by S. oneidensis over the same time. Extending the reaction to 72 h slightly increased the extent of U60 reduction to ~40%, suggesting that at least some of the U(VI) in the U60 nanoclusters was not available for bioreduction by S. oneidensis over this time period.

Figure 1: Reduction kinetics of U60 nanoclusters and aqueous uranyl acetate by S. oneidensis. The unlabeled data (open symbols) represent the results of: 1) the abiotic control experiment with only U60 (○
), 2) the control experiment with U60 and killed cells (∆), and 3) the re-oxidation experiment using samples from the reduction of U60 by S. oneidensis (□).
Figure 1:

Reduction kinetics of U60 nanoclusters and aqueous uranyl acetate by S. oneidensis. The unlabeled data (open symbols) represent the results of: 1) the abiotic control experiment with only U60 ( ), 2) the control experiment with U60 and killed cells (∆), and 3) the re-oxidation experiment using samples from the reduction of U60 by S. oneidensis ().

3.2 Effect of cell density on U60 reduction

Increasing the initial cell density significantly enhances the rate and extent of U60 reduction by S. oneidensis (Figure 2). In the experiment starting with a cell density of 0.6 OD600 and an initial concentration of U(VI) of 300 μM in suspension as U60, we observed that only ~29% of the U(VI) was reduced after 120 h. In addition, most of the U(VI) reduction occurred within 48 h, and extending the reaction time from 48 h to 120 h only slightly enhanced the extent of reduction. In contrast, in experiments starting with OD600S. oneidensis values of 1.0 and 1.5, rapid and continuous reduction was observed. After 120 h, almost all of the U(VI) in the U60 was reduced in the 1.5 OD600 experiment.

Figure 2: Effect of initial cell density (OD600 values of 0.6, 1.0 or 1.5) on the reduction of U60 nanoclusters by S. oneidensis as a function of time.
Figure 2:

Effect of initial cell density (OD600 values of 0.6, 1.0 or 1.5) on the reduction of U60 nanoclusters by S. oneidensis as a function of time.

3.3 Effect of U60 concentration on U60 reduction

The concentration of U60 in the experimental system affects the reduction kinetics of U60 by S. oneidensis (Figure 3). In experiments with low initial U60 concentrations (300 μM and 500 μM), the reduction behavior displays a similar two-stage feature: a fast reduction within the first 48 h, followed by a plateau stage where very little further U(VI) was reduced. Although two-stage reduction kinetics were also observed in experiments starting with 1000 μM of U60, a plateau in reduction extent was not reached, and the extent of reduction kept increasing from 48 to 120 h. After 120 h, the reduced U(VI) was 82.3, 179.5 and 431.3 μM in experiments with initial U60 concentrations of 300, 500 and 1000 μM, respectively.

3.4 Size analysis of U60 suspensions

Our filtration tests document a dramatic decrease in average U(VI) particle size that accompanies the reduction of U60 in the bioreduction experiments (Figure 4). At the start of the experiments, virtually all of the U(VI) was present in the >0.2 μm size fraction, as filtration through a 0.2 μm filter removed nearly 100% of the U(VI) from the samples due to U60 aggregation. However, in subsequent samples, both monomeric U(VI) (determined from the 10 kDa filtration test) and the U(VI)<0.2 μm increased rapidly as U(VI) reduction proceeded. After 5 h of reaction, all of the remaining U(VI) passed through a 0.2 μm filter, implying that complete breakdown of the original U60 aggregates occurred. However, a significant difference between the concentration of total U(VI) and the concentration of monomeric U(VI) was observed, with ~25% of the remaining U(VI) in the system present as either isolated U60 nanoclusters or as U60 aggregates. From 5 h to 96 h, while aqueous monomeric U(VI) concentrations continued to decrease, the concentration of U(VI) present as nanoclusters did not change markedly with time. These results suggest that most of the U(VI) that was reduced during this time period was from the aqueous monomeric U(VI) pool, and that the U(VI) that remained in the U60 nanoclusters was likely not bioavailable.

Figure 4: Size analysis of samples during a U(VI) reduction experiment conducted with 500 μM U60 and an initial cell density of 0.6 OD600. The total U(VI) values refer to the concentration of U(VI) in the unfiltered samples. The U(VI)<0.2 μm values and the monomeric U(VI) values refer to the concentrations of U(VI) in the sample fractions that pass through a 0.2 μm PTFE filter and a 10 kDa molecular weight filter, respectively. The nanocluster U(VI) values represent the total U(VI) present within U60 nanoclusters in the unfiltered samples, and were calculated by subtracting the monomeric U(VI) values from the total U(VI) values.
Figure 4:

Size analysis of samples during a U(VI) reduction experiment conducted with 500 μM U60 and an initial cell density of 0.6 OD600. The total U(VI) values refer to the concentration of U(VI) in the unfiltered samples. The U(VI)<0.2 μm values and the monomeric U(VI) values refer to the concentrations of U(VI) in the sample fractions that pass through a 0.2 μm PTFE filter and a 10 kDa molecular weight filter, respectively. The nanocluster U(VI) values represent the total U(VI) present within U60 nanoclusters in the unfiltered samples, and were calculated by subtracting the monomeric U(VI) values from the total U(VI) values.

3.5 Adsorption of U60 onto Shewanella oneidensis

Although both the bacterial cells and the U60 nanoclusters are negatively charged, our results provide evidence that U60 nanoclusters can adsorb onto S. oneidensis cells in 0.1 M NaCl. As shown in Figure 5, without bacterial cells present (data points at OD600=0), ~84% of the U60 in 0.1 M NaCl passed through a 0.2 μm filter, and this percentage decreased slightly to ~71% after 24 h. These results indicate that some aggregation of the U60 nanoclusters occurs in the NaCl solution, but most nanoclusters remain isolated or present only as small (<0.2 μm) aggregates. In contrast, although the presence of S. oneidensis did not change the measured U(VI) concentrations in the filtrate when we filter the suspensions immediately after adding bacterial cells (the 0 h symbols in Figure 5), the concentration of U60 that passed through a 0.2 μm filter significantly decreased after 24 h of reaction. In addition, higher biomass concentrations in the system resulted in lower concentrations of U60 that passed through a 0.2 μm filter. As the experiments were conducted open to the atmosphere and an electron donor was not present in the system, microbial reduction could not occur, and thus the decrease in U60 concentration in the filtrate can be attributed to U60 adsorption onto bacterial cells. In the presence of a suspension of S. oneidensis with an OD600 value of 1.0, only ~20% of the total U60 was detected in the filtrate after 24 h of reaction, suggesting that approximately 50% of U60 adsorbed onto the bacterial cells (assuming that approximately 30% were present as aggregates larger than 0.2 μm, as shown in the 24 h cell-free controls).

Figure 5: Evidence for the adsorption of U60 nanoclusters onto S. oneidensis: the effect of cell density on the measured U(VI) concentration in the filtrate (0.2 μm) of U60 and S. oneidensis suspensions in 0.1 M NaCl that are open to the atmosphere. 0 h samples are taken immediately after the addition of cells to the U60 suspension, and the 24 h samples depict U(VI) concentrations 24 h later. Data points at OD600=0 represent control experiments in the absence of S. oneidensis cells.
Figure 5:

Evidence for the adsorption of U60 nanoclusters onto S. oneidensis: the effect of cell density on the measured U(VI) concentration in the filtrate (0.2 μm) of U60 and S. oneidensis suspensions in 0.1 M NaCl that are open to the atmosphere. 0 h samples are taken immediately after the addition of cells to the U60 suspension, and the 24 h samples depict U(VI) concentrations 24 h later. Data points at OD600=0 represent control experiments in the absence of S. oneidensis cells.

4 Discussion

In ultrapure water, U60 nanoclusters are highly negatively charged, and are largely present as isolated nanoclusters with a diameter of approximately 2.4 nm [26], [30]. Isolated U60 nanoclusters are meta-stable in ultrapure water for at least several months without structural breakdown [24]. Although self-assembly of isolated U60 nanoclusters into aggregates of U60 nanoclusters in ultrapure water may occur, this process is usually slow. For example, Soltis et al. [30] using cryo-TEM observed that U60 nanoclusters that were aged in ultrapure water for 7 months aggregated only slightly, and formed aggregates with an average size of 17.7±10.4 nm. However, the cations in our bacterial growth medium, such as K+ and Na+, can accumulate on nanocluster surfaces, thereby reducing the electrostatic repulsion between nanoclusters. As a result, aggregates of the U60 with a size of at least 0.2 μm were observed immediately after adding the nanocluster parent suspension to the growth medium, as depicted in Figure 6a. Similarly, Soltis et al. [30] reported that U60 aggregates larger than 100 nm in diameter formed within 30 min after adding ~0.1 μM of Na+ or K+ to U60 solutions in ultrapure water. These results suggest that uranyl peroxide nanoclusters likely would be present as aggregates rather than as isolated nanoclusters if they form in natural systems since most surface or groundwaters contain abundant dissolved cations including Na+ and K+. U60 aggregates were stable in the growth medium in our abiotic control experiments over the course of 72 h without significant dissolution or disaggregation (Figure S1B). Therefore, instead of isolated U60 nanoclusters, large U60 aggregates represent the major U(VI)-bearing species in suspension before reduction was initiated under our experimental conditions.

Figure 6: Proposed reaction pathways of enzymatic reduction of U60 nanoclusters by S. oneidensis: (a) Aggregation of U60 nanoclusters in bacterial growth medium; (b) Adsorption of U60 aggregates onto bacterial cells; (c) Reduction of some of the U(VI) present in the U60 aggregates, followed by the breakdown of U60 aggregates and the dissolution of some U60 nanoclusters; (d) Reduction of aqueous monomeric U(VI) to U(IV).
Figure 6:

Proposed reaction pathways of enzymatic reduction of U60 nanoclusters by S. oneidensis: (a) Aggregation of U60 nanoclusters in bacterial growth medium; (b) Adsorption of U60 aggregates onto bacterial cells; (c) Reduction of some of the U(VI) present in the U60 aggregates, followed by the breakdown of U60 aggregates and the dissolution of some U60 nanoclusters; (d) Reduction of aqueous monomeric U(VI) to U(IV).

Two possible mechanisms could be involved in the reduction by S. oneidensis of the U(VI) atoms that are introduced to our experimental systems within U60 nanoclusters: 1) a process similar to that responsible for bacterial reduction of solid-phase U(VI) wherein a first step of dissolution of U60 nanoclusters into aqueous monomeric U(VI) species occurs, followed by enzymatic bacterial reduction of the aqueous monomeric U(VI); or 2) a process similar to that responsible for bacterial reduction of aqueous-phase U(VI) wherein the U60 nanoclusters remain intact and direct reduction of U(VI) on the U60 aggregates occurs due to contact between the aggregates and the electron transport chain located within the bacterial cell walls.

The moderate reduction kinetics of U60 observed in this study does not support the dominance of the first mechanism. In general, the microbial reduction of dissolved monomeric U(VI) by pure cultures is rapid and is complete within several hours [4], [31]. Conversely, the reduction of solid phase U(VI) may occur over several weeks, and the reduction kinetics are strongly dependent on the dissolution and release rates of U(VI) from the solid phase [20]. While the reduction rate of U60 nanoclusters is significantly slower than that of uranyl acetate (Figure 1), a representative form of dissolved monomeric U(VI) widely used in previous studies, the reduction rate is still much faster than the reported reduction rate of solid phase U(VI). For example, Liu et al. [20] reported that the half-life of a solid phase U(VI), sodium boltwoodite, during the reduction by S. oneidensis MR-1 ranged from 5 to 10 days. In our study, using the same bacterial species, it took 2–3 days for the system to reduce half of the initial U60 (Figure 2). These results indicate that the reduction mechanism for U60 probably differs from both that of monomeric U(VI) and solid phase U(VI).

Furthermore, if the reduction mechanism for solid phase U(VI) dominated the reduction of U60 in this study, we would expect that dissolution of U60 nanoclusters by S. oneidensis would occur no matter whether reduction is initiated or not. Because reduction does not occur in the presence of oxygen, we conducted an oxic control experiment using otherwise identical conditions to those of the U60 reduction experiments in order to test whether U60 dissolution into aqueous monomeric U(VI) species occurs to a significant extent under the experimental conditions. This control experiment showed that filtration through a 0.2 μm membrane removed virtually all of the U(VI) present in the samples, indicating that no dissolution or disaggregation of the U60 nanoclusters occurred in the presence of S. oneidensis even after 48 h when reduction was not initiated (Figure S3). These results suggest that neither S. oneidensis nor other abiotic components in our reduction experiments caused the dissolution of U60 aggregates before reduction occurred, and direct reduction of some U(VI) within the U60 aggregates by S. oneidensis must have occurred at least in the early stages of the reduction experiment. This conclusion is supported by our adsorption results (Figure 5), which suggest that U60 nanocluster adsorption onto S. oneidensis cells is possible, and thus direct contact between the nanoclusters and the electron transport chain located within the bacterial cell walls can occur, as depicted in Figure 6b. As shown in Figures 2 and 3, increasing the concentration of either the bacterial cells or the U60 nanoclusters in the experiments enhanced the initial 24 h reduction kinetics, likely due to the enhanced concentration of U60 that was adsorbed onto, and hence bioavailable to, S. oneidensis under these conditions.

Figure 3: Effect of initial U60 concentration (300, 500 and 1000 μM) on the reduction of U60 nanoclusters by S. oneidensis as a function of time. Note that this figure depicts the concentration of U(VI) that was reduced by the bacteria rather than the normalized concentration of U(VI) remaining in solution that is shown in Figures 1 and 2.
Figure 3:

Effect of initial U60 concentration (300, 500 and 1000 μM) on the reduction of U60 nanoclusters by S. oneidensis as a function of time. Note that this figure depicts the concentration of U(VI) that was reduced by the bacteria rather than the normalized concentration of U(VI) remaining in solution that is shown in Figures 1 and 2.

After some U(VI) in the U60 aggregates was reduced in our reduction experiments, breakdown and dissolution of the U60 aggregates occurred, as depicted in Figure 6c. The concentration of aqueous monomeric U(VI) dramatically increased within the first few hours of the experiments, and these species soon became the major reservoir of U(VI) in the system. The continuous decrease in aqueous monomeric U(VI) concentration from 5 h to 96 h in Figure 4 suggests that the reduction during this period mainly involved monomeric U(VI), as depicted in Figure 6d. Over this time period, the concentration of U60 nanoclusters, represented by the difference between the ‘total U(VI)’ values and the ‘monomeric U(VI)’ values in Figure 4, remained nearly constant, suggesting that the remaining U60 nanoclusters in suspension were not bioavailable for reduction and did not undergo further degradation, dissolution, or reduction. As shown in Figure 4, a considerable concentration of dissolved monomeric U(VI) was still detected in solution and was not reduced after 96 h, implying that the bioavailability of the dissolved monomeric U(VI) that originated from the U60 nanoclusters was less than that of aqueous uranyl acetate. However, more than 90% of the U(VI) in the U60 nanoclusters was reduced after 120 h of reaction in the presence of a cell suspension with an initial OD600 value of 1.5 (Figure 2), suggesting that a higher initial cell density can strongly enhance the bioavailability of the various U(VI) species that were produced during U60 reduction.

Our results show that although direct reduction of some of the U(VI) within U60 nanoclusters is possible, direct contact between the nanoclusters and the bacteria does not lead to complete reduction of the U(VI) within the nanoclusters under most of the experimental conditions. Furthermore, our results indicate that enzymatic reduction of a portion of the U(VI) within the U60 nanoclusters causes a breakdown of the U60 aggregates and a partial dissolution of the U60 nanoclusters to form aqueous monomeric U(VI) species that are small enough to pass through the 10 kDa molecular weight filter. Our findings demonstrate that the reduction of U60 nanoclusters by S. oneidensis is initiated by direct electron transfer to the U60 nanoclusters made possible due to adsorption of the nanoclusters onto the bacterial cells, similar to the reduction mechanism for aqueous uranyl species. However, the dominant reduction mechanism quickly changes to one that involves an indirect process of electron transfer from the bacterial cells to the breakdown or dissolution products from the U60 nanoclusters. It is likely that this dual nature of U60 reduction causes the relatively slow overall U(VI) reduction kinetics observed here.

5 Conclusions

In this study, we demonstrate that a representative uranyl peroxide nanocluster, U60, can be reduced by Shewanella oneidensis MR-1 under anoxic conditions, and that the produced U(IV) is fully re-oxidized after exposure to air. The reduction kinetics of U60 is significantly slower than that of aqueous uranyl acetate, and higher initial cell densities or U60 concentrations result in faster reduction kinetics. Upon contact with the growth medium, U60 nanoclusters form aggregates with a size of at least 0.2 μm, and these aggregates are the primary reservoir of U(VI) at the beginning of the experiments. The initial reduction of U60 by S. oneidensis in the experiments involves direct reduction of a portion of the U(VI) that is present within U60 nanocluster aggregates via adsorption, and this initial reduction is followed by, and likely causes, the breakdown of the U60 aggregates and the conversion of some of the U60 nanoclusters to aqueous monomeric U(VI) species that are small enough to pass through the 10 kDa molecular weight filter used here. After the initial period of direct U60 reduction, most of the reduction involves the newly produced aqueous monomeric U(VI) species. Under most of the experimental conditions, measureable concentrations of U60 and dissolved U(VI) were still present in solution after several days of reaction. Our results suggest that the reduction of U60 nanoclusters is different from either aqueous monomeric U(VI) or solid-phase U(VI), and that complex processes such as nanocluster adsorption onto the reducing bacterial cells, nanocluster aggregation behavior, and the breakdown of the nanoclusters to monomeric aqueous U(VI) species all are involved in the overall U60 reduction reactions. This study is first to show the reduction capacity of bacteria for uranium peroxide nanoclusters, and the results improve our understanding and ability to predict the long term fate of uranium in the environment.

Acknowledgements

This work was supported by the Office of Basic Energy Sciences of the U.S. Department of Energy as part of the Materials Science of Actinides Energy Frontier Research Center (DE-SC0001089). We thank Jennifer Szymanowski for synthesis of the U60 nanoclusters and for technical support and assistance. The ICP-OES analyses were conducted at the Center for Environmental Science and Technology (CEST) at University of Notre Dame.

References

1. da Cunha, K. M. D., Henderson, H., Thomson, B. M., Hecht, A. A.: Ground water contamination with U-238, U-234, U-235, Ra-226 and Pb-210 from past uranium mining: cove wash, Arizona. Environ. Geochem. Hlth. 36(3), 477 (2014).10.1007/s10653-013-9575-2Suche in Google Scholar PubMed

2. Campbell, K. M., Kukkadapu, R. K., Qafoku, N. P., Peacock, A. D., Lesher, E., Williams, K. H., Bargar, J. R., Wilkins, M. J., Figueroa, L., Ranville, J., Davis, J. A., Long, P. E.: Geochemical, mineralogical and microbiological characteristics of sediment from a naturally reduced zone in a uranium-contaminated aquifer. Appl. Geochem. 27(8), 1499 (2012).10.1016/j.apgeochem.2012.04.013Suche in Google Scholar

3. Lovley, D. R., Phillips, E. J. P., Gorby, Y. A., Landa, E. R.: Microbial reduction of uranium. Nature 350(6317), 413 (1991).10.1038/350413a0Suche in Google Scholar

4. Gorby, Y. A., Lovley, D. R.: Enzymatic uranium precipitation. Environ. Sci. Technol. 26(1), 205 (1992).10.1021/es00025a026Suche in Google Scholar

5. Senko, J. M., Istok, J. D., Suflita, J. M., Krumholz, L. R.: In-situ evidence for uranium immobilization and remobilization. Environ. Sci. Technol. 36(7), 1491 (2002).10.1021/es011240xSuche in Google Scholar PubMed

6. Finneran, K. T., Anderson, R. T., Nevin, K. P., Lovley, D. R.: Potential for bioremediation of uranium-contaminated aquifers with microbial U(VI) reduction. Soil Sediment. Contam. 11(3), 339 (2002).10.1080/20025891106781Suche in Google Scholar

7. Anderson, R. T., Vrionis, H. A., Ortiz-Bernad, I., Resch, C. T., Long, P. E., Dayvault, R., Karp, K., Marutzky, S., Metzler, D. R., Peacock, A., White, D. C., Lowe, M., Lovley, D. R.: Stimulating the in situ activity of Geobacter species to remove uranium from the groundwater of a uranium-contaminated aquifer. Appl. Environ. Microb. 69(10), 5884 (2003).10.1128/AEM.69.10.5884-5891.2003Suche in Google Scholar PubMed PubMed Central

8. Barton, L. L., Choudhury, K., Thomson, B. M., Steenhoudt, K., Groffman, A. R.: Bacterial reduction of soluble uranium: The first step of in situ immobilization of uranium. Radioact. Waste Manag. 20(2–3), 141 (1996).Suche in Google Scholar

9. Liu, C. X., Gorby, Y. A., Zachara, J. M., Fredrickson, J. K., Brown, C. F.: Reduction kinetics of Fe(III), Co(III), U(VI) Cr(VI) and Tc(VII) in cultures of dissimilatory metal-reducing bacteria. Biotechnol. Bioeng. 80(6), 637 (2002).10.1002/bit.10430Suche in Google Scholar PubMed

10. Elias, D. A., Krumholz, L. R., Wong, D., Long, P. E., Suflita, J. M.: Characterization of microbial activities and U reduction in a shallow aquifer contaminated by uranium mill tailings. Microbial. Ecol. 46(1), 83 (2003).10.1007/s00248-002-1060-xSuche in Google Scholar PubMed

11. Ortiz-Bernad, I., Anderson, R. T., Vrionis, H. A., Lovley, D. R.: Resistance of solid-phase U(VI) to microbial reduction during in situ bioremediation of uranium-contaminated groundwater. Appl. Environ. Microb. 70(12), 7558 (2004).10.1128/AEM.70.12.7558-7560.2004Suche in Google Scholar PubMed PubMed Central

12. Gu, B. H., Wu, W. M., Ginder-Vogel, M. A., Yan, H., Fields, M. W., Zhou, J., Fendorf, S., Criddle, C. S., Jardine, P. M.: Bioreduction of uranium in a contaminated soil column. Environ. Sci. Technol. 39(13), 4841 (2005).10.1021/es050011ySuche in Google Scholar PubMed

13. Sheng, L., Szymanowski, J., Fein, J. B.: The effects of uranium speciation on the rate of U(VI) reduction by Shewanella oneidensis MR-1. Geochim. Cosmochim. Acta 75(12), 3558 (2011).10.1016/j.gca.2011.03.039Suche in Google Scholar

14. Rui, X., Kwon, M. J., O’Loughlin, E. J., Dunham-Cheatham, S., Fein, J. B., Bunker, B., Kemner, K. M., Boyanov, M. I.: Bioreduction of hydrogen uranyl phosphate: mechanisms and U(IV) products. Environ. Sci. Technol. 47(11), 5668 (2013).10.1021/es305258pSuche in Google Scholar PubMed

15. Yang, Y., Wang, S., Albrecht-Schmitt, T. E.: Microbial dissolution and reduction of uranyl crystals by Shewanella oneidensis MR-1. Chem. Geol. 387, 59 (2014).10.1016/j.chemgeo.2014.08.020Suche in Google Scholar

16. Wall, J. D., Krumholz, L. R.: Uranium reduction. Annu. Rev. Microbiol. 60, 149 (2006).10.1146/annurev.micro.59.030804.121357Suche in Google Scholar PubMed

17. Sheng, L., Fein, J. B.: Uranium reduction by Shewanella oneidensis MR-1 as a function of NaHCO3 concentration: surface complexation control of reduction kinetics. Environ. Sci. Technol. 48(7), 3768 (2014).10.1021/es5003692Suche in Google Scholar PubMed

18. Cologgi, D. L., Lampa-Pastirk, S., Speers, A. M., Kelly, S. D., Reguera, G.: Extracellular reduction of uranium via Geobacter conductive pili as a protective cellular mechanism. P. Natl. Acad. Sci. USA 108(37), 15248 (2011).10.1073/pnas.1108616108Suche in Google Scholar PubMed PubMed Central

19. Brooks, S. C., Fredrickson, J. K., Carroll, S. L., Kennedy, D. W., Zachara, J. M., Plymale, A. E., Kelly, S. D., Kemner, K. M., Fendorf, S.: Inhihition of bacterial U(VI) reduction by calcium. Environ. Sci. Technol. 37(9), 1850 (2003).10.1021/es0210042Suche in Google Scholar PubMed

20. Liu, C. X., Jeon, B. H., Zachara, J. M., Wang, Z. M., Dohnalkova, A., Fredrickson, J. K.: Kinetics of microbial reduction of solid phase U(VI). Environ. Sci. Technol. 40(20), 6290 (2006).10.1021/es0608601Suche in Google Scholar PubMed

21. Qiu, J., Burns, P. C.: Clusters of actinides with oxide, peroxide, or hydroxide bridges. Chem. Rev. 113(2), 1097 (2013).10.1021/cr300159xSuche in Google Scholar PubMed

22. Burns, P. C., Kubatko, K. A., Sigmon, G., Fryer, B. J., Gagnon, J. E., Antonio, M. R., Soderholm, L.: Actinyl peroxide nanospheres. Angew. Chem. Int. Edit. 44(14), 2135 (2005).10.1002/anie.200462445Suche in Google Scholar PubMed

23. Flynn, S. L., Szymanowski, J. E. S., Gao, Y. Y., Liu, T. B., Burns, P. C., Fein, J. B.: Experimental measurements of U60 nanocluster stability in aqueous solution. Geochim. Cosmochim. Acta 156, 94 (2015).10.1016/j.gca.2015.02.021Suche in Google Scholar

24. Armstrong, C. R., Nyman, M., Shvareva, T., Sigmon, G. E., Burns, P. C., Navrotsky, A.: Uranyl peroxide enhanced nuclear fuel corrosion in seawater. P. Natl. Acad. Sci. USA. 109(6), 1874 (2012).10.1073/pnas.1119758109Suche in Google Scholar PubMed PubMed Central

25. Burns, P. C., Ewing, R. C., Navrotsky, A.: Nuclear fuel in a reactor accident. Science 335(6073), 1184 (2012).10.1126/science.1211285Suche in Google Scholar PubMed

26. Sigmon, G. E., Unruh, D. K., Ling, J., Weaver, B., Ward, M., Pressprich, L., Simonetti, A., Burns, P. C.: Symmetry versus minimal pentagonal adjacencies in uranium-based polyoxometalate fullerene topologies. Angew. Chem. Int. Edit. 48(15), 2737 (2009).10.1002/anie.200805870Suche in Google Scholar PubMed

27. McGrail, B. T., Sigmon, G. E., Jouffret, L. J., Andrews, C. R., Burns, P. C.: Raman spectroscopic and ESI-MS characterization of uranyl peroxide cage clusters. Inorg. Chem. 53(3), 1562 (2014).10.1021/ic402570bSuche in Google Scholar PubMed

28. Lovley, D. R., Phillips, E. J. P.: Novel mode of microbial energy-metabolism – organic-carbon oxidation coupled to dissimilatory reduction of iron or manganese. Appl. Environ. Microb. 54(6), 1472 (1988).10.1128/aem.54.6.1472-1480.1988Suche in Google Scholar PubMed PubMed Central

29. Brina, R., Miller, A. G.: Direct detection of trace levels of uranium by laser-induced kinetic phosphorimetry. Anal. Chem. 64(13), 1413 (1992).10.1021/ac00037a020Suche in Google Scholar

30. Soltis, J. A., Wallace, C. M., Penn, R. L., Burns, P. C.: Cation-dependent hierarchical assembly of U60 nanoclusters into macro-ion assemblies imaged via cryogenic transmission electron microscopy. J. Am. Chem. Soc. 138(1), 191 (2016).10.1021/jacs.5b09802Suche in Google Scholar PubMed

31. Liu, C., Zachara, J. M. Fredrickson, J. K. Kennedy, D. W., Dohnalkova, A.: Modeling the inhibition of the bacteral reduction of U(VI) by beta-MnO2(s). Environ. Sci. Technol. 36(7), 1452 (2002).10.1021/es011159uSuche in Google Scholar PubMed


Supplemental Material:

The online version of this article (DOI: https://doi.org/10.1515/ract-2017-2759) offers supplementary material, available to authorized users.


Received: 2017-1-10
Accepted: 2017-5-23
Published Online: 2017-6-29
Published in Print: 2018-1-26

©2018 Walter de Gruyter GmbH, Berlin/Boston

Heruntergeladen am 8.9.2025 von https://www.degruyterbrill.com/document/doi/10.1515/ract-2017-2759/html
Button zum nach oben scrollen