Reduction of Net Mercury Methylation by Iron in Desulfobulbus

May 31, 2003 - In sediment-free cultures, net mercury methylation decreased with ... A pure culture of the sulfate-reducing Desulfobulbus propionicus ...
0 downloads 0 Views 110KB Size
Environ. Sci. Technol. 2003, 37, 3018-3023

Reduction of Net Mercury Methylation by Iron in Desulfobulbus propionicus (1pr3) Cultures: Implications for Engineered Wetlands ANNA S. MEHROTRA, ALEX J. HORNE, AND DAVID L. SEDLAK* Department of Civil and Environmental Engineering, University of California, Berkeley, Berkeley, California 94720-1710

Although one potential drawback of wetland construction and restoration is the formation of monomethylmercury, it may be possible to decrease net mercury methylation with the use of an appropriate sediment amendment. Using pure cultures of the sulfate-reducing bacterium Desulfobulbus propionicus (1pr3), we tested the hypothesis that adding ferrous iron to sulfidic wetland sediments decreases mercury solubility and bioavailability and, therefore, net methylation. In sediment-free cultures, net mercury methylation decreased with increasing [Fe(II)]. After 72 h of incubation, more than four times as much net methylmercury formed in the lowest ([Fe(II)] ) 10-6 M) treatment (180 ( 33 pM) as compared with the highest ([Fe(II)] ) 10-2 M) treatment (42 ( 14 pM). In cultures containing a model wetland sediment, more than three times as much methylmercury was observed in 10-6 M Fe(II) treatments (1,010 ( 95 pM) as compared with treatments amended with 10-2 M Fe(II) (300 ( 46 pM). Initial filterable mercury measurements and chemical equilibrium speciation predictions suggest that the lower net methylmercury production in the high-iron treatments was due to a decrease in sulfide activity and a concomitant decrease in the concentration of dissolved mercury. Although iron amendments could potentially minimize net mercury methylation in engineered wetland sediments, further research under field conditions is required to assess the efficacy of this approach.

Introduction Wetland restoration and construction have received considerable attention in the past few decades. Between 1986 and 1997, federal programs alone led to a 108 000 ha (43,740 acre) increase in U.S. wetland acreage (approximately 0.1% of total acreage) (1). Among the benefits of wetland restoration and construction are habitat for threatened and endangered species, improved water quality, flood mitigation, aquifer recharge, and aesthetic and heritage values (2). One potential drawback is the formation of monomethylmercury (MHg) (3-6). Of all the chemical forms of mercury, methylmercury poses the greatest toxicity concern to humans and wildlife (7, 8) and readily accumulates in aquatic food webs, being * Corresponding author phone: (510)643-0256; fax: (510)642-7483; e-mail: [email protected]. 3018

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 13, 2003

implicated in 74% of all fish consumption advisories in the United States (9). Mercury methylation in wetlands has been shown to occur in anoxic sediments through a process mediated by sulfatereducing bacteria (10, 11). Methylation rates depend on bacterial growth rates (12) as well as on the concentration of dissolved mercury species. In anoxic waters, the activity of Hg2+ and dissolved Hg(II) complexes is controlled by sulfide in the presence of cinnabar (HgS(s)) (13). Complexation with dissolved natural organic matter (14-16) or interactions with minerals and organic-coated surfaces (17) may also affect the concentration and speciation of dissolved Hg(II). It has been hypothesized that only small, uncharged mercury complexes (e.g., HgS0(aq) and Hg(HS)20) capable of passive diffusion across bacterial cell membranes are available for methylation (18-20). As a result, the total activity of uncharged Hg(II) complexes should govern the extent of mercury methylation in anoxic sediments. To maximize the benefits of wetland restoration and construction, it is desirable to minimize the potential for MHg formation. Although it is difficult, and usually undesirable, to decrease biological activity or to avoid reducing conditions in wetland sediments, changes in Hg(II) solubility and speciation may be feasible with an appropriate sediment amendment. For example, the presence of Fe2+ should decrease the activity of S2- through the formation of FeS(s). Since S2- is an important ligand for Hg(II) in anoxic environments (13), the decrease in sulfide activity could affect sediment mercury chemistry. We hypothesize that adding reduced iron to wetland sediments will decrease the extent of net mercury methylation by decreasing S2- activity and hence Hg(II) solubility and bioavailability. To test this hypothesis, we added Hg(II) and varying concentrations of Fe(II) to pure cultures of the sulfate-reducing bacterium Desulfobulbus propionicus (1pr3) grown in both sedimentfree media and in model wetland sediment.

Methods Bacteria and Culture Media. A pure culture of the sulfatereducing Desulfobulbus propionicus (1pr3), known to methylate mercury, was obtained from the American Type Culture Collection (ATCC 33891). The bacteria were maintained using strict anaerobic techniques on Baar’s medium for sulfate reducers (ATCC medium 1249), modified to contain 0.1 mM, instead of 2 mM, Fe(NH4)2(SO4)2. All chemicals were obtained from Fisher Scientific unless otherwise noted and were of at least 99% purity and used without further purification. Media for the experiments consisted of 1 mM KH2PO4, 3 mM (NH4)3PO4, 15 mM NaCl, 5 mM KCl, 25 mM NaHCO3, 20 mM sodium lactate, 4 mM Na2SO4, 10 mL/L 50× vitamin solution (21), and 1 mL/L 200× trace metals stock (21) with 1 mg/L resazurin added to deaerated (sparged with ultrahigh-purity N2 for approximately 3 h) Nanopure (Barnstead II) water. After the pH was adjusted to 7.0 (with HCl), 50 mL of media was dispensed into serum bottles that were then sealed with butyl rubber stoppers and autoclaved. Filter-sterilized (0.22-µm syringe filters; Gelman Acrodisc, Fisher Scientific) stocks of MgSO4, calcium lactate, and TiNTA (as a reductant; prepared according to ref 22) were added to the media for final concentrations of 1.5, 0.5, and 0.1 mM, respectively. All additions were performed in an anaerobic chamber. The media turned clear upon addition of Ti-NTA (indicating low redox potential), and only clear media was used in the experiments. During the experiments, the quantity of SO42- reduced, rather than direct cell counts or optical density, was used to 10.1021/es0262838 CCC: $25.00

 2003 American Chemical Society Published on Web 05/31/2003

FIGURE 1. SO42- and measurable S(-II) in sediment-free cultures with 10-6 M Fe(II) and 10-2 M Fe(II). Error bars represent one standard error for quadruplicate cultures and fall within the symbols when not visible. monitor cell growth. The bacteria were in exponential growth to early stationary phases during the course of the experiment, as evidenced by increasing rates of SO42- reduction over time (Figure 1). Inocula transferred into oxic media were used as purity checks, and no cell growth (measured by absorbance at 660 nm) was detected in the oxic controls. Experimental Procedure. Experiments were carried out both without sediment and with model wetland sediment. The model wetland sediment consisted of 25 g of very coarse (1-2 mm) acid-washed sand, 2.5 g of dried and sieved cattail (Typha latifolia) stems (