Dissolved Organic Matter Enhances Microbial Mercury Methylation

Article pubs.acs.org/est

Dissolved Organic Matter Enhances Microbial Mercury Methylation Under Sulfidic Conditions Andrew M. Graham,*,† George R. Aiken,‡ and Cynthia C. Gilmour† †

Smithsonian Environmental Research Center, 647 Contees Wharf Road, Edgewater, Maryland 21037, United States U.S. Geological Survey, 3215 Marine Street, Suite E127, Boulder, Colorado 80303, United States

‡

Downloaded via UNIV OF GOTHENBURG on January 29, 2019 at 06:59:40 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Dissolved organic matter (DOM) is generally thought to lower metal bioavailability in aquatic systems due to the formation of metal−DOM complexes that reduce free metal ion concentrations. However, this model may not be pertinent for metal nanoparticles, which are now understood to be ubiquitous, sometimes dominant, metal species in the environment. The influence of DOM on Hg bioavailability to microorganisms was examined under conditions (0.5−5.0 nM Hg and 2−10 μM sulfide) that favor the formation of β-HgS(s) (metacinnabar) nanoparticles. We used the methylation of stable-isotope enriched 201HgCl2 by Desulfovibrio desulf uricans ND132 in short-term washed cell assays as a sensitive, environmentally significant proxy for Hg uptake. Suwannee River humic acid (SRHA) and Williams Lake hydrophobic acid (WLHPoA) substantially enhanced (2to 38-fold) the bioavailability of Hg to ND132 over a wide range of Hg/DOM ratios (9.4 pmol/mg DOM to 9.4 nmol/mg DOM), including environmentally relevant ratios. Methylmercury (MeHg) production by ND132 increased linearly with either SRHA or WLHPoA concentration, but SRHA, a terrestrially derived DOM, was far more effective at enhancing Hg-methylation than WLHPoA, an aquatic DOM dominated by autochthonous sources. No DOM-dependent enhancement in Hg methylation was observed in Hg−DOM− sulfide solutions amended with sufficient L-cysteine to prevent β-HgS(s) formation. We hypothesize that small HgS particles, stabilized against aggregation by DOM, are bioavailable to Hg-methylating bacteria. Our laboratory experiments provide a mechanism for the positive correlations between DOC and MeHg production observed in many aquatic sediments and wetland soils.

■

INTRODUCTION The bioavailability of Hg for microbial uptake and methylation provides a useful model system for examining the role of DOM in metal nanoparticle bioavailability. MeHg is produced mainly in anoxic sediments and soils, primarily by dissimilatory sulfate- and iron-reducing bacteria (DSRB and DFeRB)1,2 belonging to the δ-Proteobacteria. Field and laboratory experiments show that speciation of inorganic mercury (Hg(II)i) is a major control on its bioavailability, uptake, and methylation. Microbial Hg methylation occurs most rapidly in mildly sulfidic environments.3 Two models for Hg uptake by Hg-methylating organisms have been developed: (1) passive diffusion of small neutrally charged Hg−ligand complexes, such as HgCl or HgS complexes,3,4 and (2) active transport of specific Hg−ligand complexes such as Hg−amino acid complexes.5−7 There are, however, numerous uncertainties surrounding these models. Importantly, the extant models consider only dissolved Hg(II)i species while evidence is mounting for nanoparticulate metacinnabar (β-HgS(s)) formation in the sulfidic environments where Hg methylation occurs.8−10 In oxic waters, metal−ligand complexes of Hg(II)i with thiol moieties of DOM are the dominant forms of filterable Hg.11,12 DOM generally reduces Hg bioavailability in surface waters.4,13 © 2012 American Chemical Society

In the presence of sulfide, however, HgS complexes dominate aqueous Hg speciation3,14 (Supporting Information Table S1). Neutral HgS complexes are highly bioavailable to Hgmethylating bacteria.3 Deonarine et al.8 recently demonstrated nanoparticulate β-HgS(s) formation and suggested that the bioavailable neutral Hg−S species may be nanoparticulate βHgS(s) or polynuclear Hg−S clusters rather than the aqueous HgS0 (or HOHgSH 0) monomers previously proposed. Importantly, DOM interacts strongly with β-HgS(s) 15 effectively limiting particle growth8−10,16 and potentially influencing Hg bioavailability. Filter-passing Hg−S−DOM polynuclear clusters or HgS(s) nanoparticles stabilized by DOM are probably the dominant forms of Hg(II)i in anaerobic soil and sediment pore waters where MeHg is produced.8−10,17 Further, the kinetics of particle formation and growth, rather than equilibrium speciation, may control Hg(II)i bioavailability to Hg-methylating bacteria inhabiting anoxic, sulfidic waters. Here, we investigate Hg(II)i bioavailability under sulfidic conditions using Desulfovibrio desulf uricans ND132, a HgReceived: Revised: Accepted: Published: 2715

October 17, 2011 February 2, 2012 February 6, 2012 February 6, 2012 dx.doi.org/10.1021/es203658f | Environ. Sci. Technol. 2012, 46, 2715−2723

Environmental Science & Technology

Article

Table 1. Characteristics of Dissolved Organic Matter Isolates Used in This Studya average MW (Da)

aromatic C (% total C)

reduced S (mol % total S)

SUVA280 (L (mg C)‑1 m‑1)

isolate

site description

C (%)

H (%)

O (%)

N (%)

S (%)

ash (%)

Suwannee River humic acid (SRHA)

black water river draining Okeefenokee Swamp (GA) seepage lake in northcentral MN

52.6b

4.3b

42.0b

1.2b

0.5b

1.04b

1399c

35.1c

18.3c

4.26d

55.2d

5.7d

36.5d

1.8d

0.75d

2.98c

772c

13.8c

n.d.

1.20d

Williams Lake hydrophobic acid (WLHPoA)

a Elemental data are corrected for ash content. n.d. = not determined. bData from International Humic Substance Society (IHSS). cData from Waples et al.19 dThis study.

purification. Williams Lake (White Oak, MN) hydrophobic acid (WLHPoA) was isolated according to the method of Aiken et al.24 In short, Williams Lake water sampled in September 2010 was filtered (0.45 μm), acidified to pH 2 with trace metal grade HCl, and passed through a column of Amberlite XAD-8 resin. The hydrophobic acid fraction (HPoA; comprised of humic and fulvic acids) was retained on the XAD-8 resin and eluted with 0.1 N NaOH. The eluate was hydrogen-saturated, desalted, and freeze-dried. For each isolate, specific ultraviolet absorbance (SUVA), a measure strongly correlated to DOM aromaticity, was determined at λ =280 nm.25 Characteristics of the two isolates are summarized in Table 1. Washed Cell Hg Methylation Assays. We evaluated the effect of Hg/DOM ratio (ratios ranged from ∼9 pmol Hg/mg DOM to ∼9 nmol/mg DOM) and DOM source on MeHg production by strain ND132 in short-term washed cell assays in minimal medium. These assays permit greater control of Hg(II)i speciation than is achievable under batch growth in complex culture media.6 To prepare cultures for methylation assays, cells were grown to midlog phase on EPF growth medium and harvested by centrifugation (15 min at 3000g). In an anaerobic glovebag (95% N2, 5% H2 atmosphere), cells were washed by resuspending in sterile, prewarmed (31 °C), pH 7.4 ± 0.1 EPF wash buffer consisting of 1 mM Na-pyruvate, 1 mM fumaric acid, 10.1 mM KCl, 171 mM NaCl, 4.4 mM KH2PO4, 7.5 mM NH4Cl, 5 mM MOPS, and 1 mg/L resazurin. Wash buffer was reduced with 40 μM of filter-sterilized Ti-NTA. Cells were then reharvested by centrifugation (15 min at 3000g). To conduct methylation assays, washed cells were resuspended in EPF wash buffer amended with DOM isolates, or 0.5 mM cysteine as a control, plus enriched 201HgCl2. Methylation was assessed from production of Me201Hg after 3 h incubation at 31 °C. Assays were conducted in triplicate in filter-sterilized, prewarmed assay buffer at pH 7.2 ± 0.1. Assay buffer was reduced with 40 μM filter-sterilized Ti-NTA just prior to cell resuspension. DOM amendments were 0−48.4 mg/L of SRHA or 0−50.9 mg/L of WLHPoA. Hg was added at 0.5 or 5 nM as stable isotope-enriched 201HgCl2 (Oak Ridge Laboratory, 98.11% 201Hg). The stable isotope approach was used to separate the Hg spike from any Hg found in the DOM isolate or other medium components. We report DOM concentrations based on the known mass of isolate added; nominal DOC concentrations were estimated based upon DOM concentration and elemental composition. 201HgCl2 was pre-equilibrated with the DOM isolate for 24 h in the dark prior to challenging ND132, as formation of Hg−DOM complexes was expected to occur on the time scale of hours.26,27 To achieve low sulfide levels in sulfate-free washed-cell assays, we utilized ND132’s ability to cleave cysteine to sulfide. Although cysteine was not a component of the wash buffer for

methylating DSRB exhibiting high rates of MeHg production that can be grown under chemically controlled conditions.18 Because Hg methylation occurs intracellularly, 18 MeHg production serves as a relevant proxy for Hg uptake. Suwannee River humic acid (SRHA) and Williams Lake hydrophobic acid (WLHPoA) were selected as nonmarine end-members of the aromaticity spectrum of DOMs found in natural waters.19 Both DOM isolates, and SRHA in particular, have been utilized in previous studies of Hg−DOM and Hg−S−DOM interactions, including Hg−DOM binding12 and HgS(s) precipitation,9,16 aggregation,8,16 and dissolution.9 Inorganic Hg(II) bioavailability in the presence of both sulfide and SRHA or WLHPoA was examined in short-term washed-cell assays6 with careful attention paid to Hg(II)i speciation and mass balance. We evaluated MeHg production at low μM sulfide concentrations over a wide range of Hg/DOM ratios, including ratios approximating natural environments (e.g., ∼9 pmol Hg/mg DOM). By using a microorganism to query Hg(II)i speciation, we were able to work at Hg/DOM ratios well below those currently possible for study by spectroscopic techniques.

■

EXPERIMENTAL SECTION Cell Cultivation and Maintenance. Desulfovibrio desulf uricans ND132 was selected as a model organism for studying MeHg production in the presence of DOM because of its high rates of MeHg production, our experience growing it under controlled conditions, and prior study18,20 including a full genome sequence. 21 Strain ND132 was isolated from Chesapeake Bay bottom sediments22 where high rates of MeHg production occur in the presence of micromolar sulfide and mg/L concentrations of DOM.23 When originally named, the species was considered D. desulf uricans. However, a recent phylogeny of the genus suggested that strain ND132 might be better classified with the halophilic and dechlorinating Desulfuricans strains;18 it awaits formal renaming of the species. Cells were maintained on pH 7.20 ± 0.1 deoxygenated estuarine pyruvate-fumarate (EPF) growth medium consisting of the following: 40 mM Na-pyruvate, 40 mM fumaric acid, 1.7 mM NaH2PO4·H2O, 18.8 mM NH4Cl, 6.7 mM KCl, 171 mM NaCl, 10 mM MOPS, 1.5 mM MgCl2, 1.5 mM CaCl2, 0.5 mM L-cysteine, 4.4 μM FeCl2, 25 nM Na2SeO4, 25 nM Na2WO4, 0.5 g/L of yeast extract, 0.5 mL/L of trace metals stock,18 5 mL/L of vitamins stock,18 and 1 mg/L of resazurin as a redox indicator. Cells were grown anaerobically in serum bottles or Hungate tubes at 31 °C. DOM Isolates. The effect of DOM on Hg bioavailability and methylation was tested using isolated fractions of organic matter from two locations. Suwannee River humic acid (SRHA) Standard II was purchased from the International Humic Substances Society (IHSS) and used without further 2716

dx.doi.org/10.1021/es203658f | Environ. Sci. Technol. 2012, 46, 2715−2723


Article

Figure 1. Effect of Suwannee River humic acid (SRHA) concentration or L-cysteine (CYS) addition on MeHg production by Desulfovibrio desulfuricans ND132. Bars are measurements made at the end of 3 h washed cell assays at 31 °C. Error bars indicate standard deviation of triplicate reaction tubes. Top panels show the distribution of MeHg between cells and other particulates (gray bars) and the aqueous phase (white bars); total MeHg concentrations in the medium are shown with black bars. Bottom panels show the distribution of total Hg (THg) between phases. The concentration of total added Hg is shown with dashed lines; the difference between the total Hg measured in the medium (black bars in panels c and d) and the amount added is the loss to reaction tube walls.

samples were stored frozen until analysis by the Bradford protein assay.29 MeHg was determined by isotope-dilution (ID) gaschromatography (GC) inductively coupled plasma mass spectrometry (ICP-MS), following distillation and ethylation, as previously described.30,31 ID spikes were Me 199 Hg (synthesized in-house by reaction of methylcobalamin with 91.95% enriched 199HgCl2, Oak Ridge National Laboratories). Samples were analyzed by a Brooks Rand MERX automated methylmercury system interfaced to a Perkin-Elmer Elan DRCII ICP-MS.30 The average relative percent difference of duplicate analyses was 3.5% (n = 7 pairs) and the instrument detection limit for Me201Hg (calculated as 3 times the standard deviation for distillation blanks) was 0.03 pM, corresponding to a method detection limit of 0.7 pM for 1-mL sample distillations. Total mercury (THg) was determined after digestion by SnCl2 reduction and ICP-MS, either with or without the use of dual gold-trap amalgamation preconcentration.30 Both filters and unfiltered aqueous culture medium were digested in hot 7:4 v/v HNO3/H2SO4 (1:2 v/v ratio for sample to digest acid). These digestates and filtered THg samples were amended with 1% v/v BrCl. Excess 201THg concentrations were calculated based on the isotope abundances of the isotope-enriched tracer and ambient Hg after correcting for impurities in the 201HgCl2 tracer.31

methylation assays, washed cells continued to produce low levels of sulfide during the assays from the cysteine provided in growth medium prior to washing. This approach yielded sulfide concentrations in assays of 2−6 μM except in controls where cysteine was added to the assay buffer (see Figures S1−S3 in Supporting Information). Uninoculated assay buffer with either 19.2 mg/L SRHA and 5 nM 201HgCl2 or 20.7 mg/L WLHPoA and 0.5 nM 201HgCl2 served as controls for possible abiotic Hg methylation. For positive methylation controls, we made use of the known cysteine stimulation of Hg-methylation by ND132,7 and amended assay buffer with 0.5 mM L-cysteine and Hg, but no DOM. Prior to the start of the assays, aliquots were removed for analysis of optical density (OD660), pH, sulfide, and total cell protein. Following the 3 h incubation period, aliquots were removed for final OD660, pH, sulfide, total cell protein, unfiltered Me201Hg, and total 201Hg (201THg). Samples for filterable Me201Hg and 201THg were obtained by filtration through a 0.2-μm polycarbonate track-etched filter (Whatman). Filters were retained for analysis of particulate Me201Hg and 201 THg. MeHg and THg samples were acidified to 0.5% v/v HCl with trace metal grade HCl (Fisher), and stored frozen until analysis. All incubations and sample preparation were carried out inside the glovebag. Analytical Methods. Sulfide was measured using an ionspecific electrode on samples preserved in sulfide antioxidant buffer, and calibrated with Pb-titrated standards.28 Protein 2717



■

Article

RESULTS AND DISCUSSION MeHg Production and Hg Partitioning in Washed Cell Assays. Both of the DOM isolates tested enhanced the uptake and methylation of 0.5 to 5 nM Hg(II)i by D. desulf uricans ND132 in the presence of low μM sulfide levels (Figures 1 and 2). SRHA increased MeHg production up to ∼40-fold for 5 nM

The observed increase in MeHg production across the DOM concentration gradients could not be attributed to DOM-driven differences in basic medium chemistry, cell density, or metabolic activity. Medium pH, optical density, total cell protein, and sulfide concentrations were similar across treatments at a given 201 HgCl 2 addition (Supporting Information Figures S1−S3). ND132 appeared unable to metabolize SRHA. In separate experiments, we evaluated sulfate reduction rates by ND132 with SRHA (0−33.6 mg/L) supplied as the sole carbon source/electron donor, and found no significant (p > 0.05; Student's t-test) sulfide production over 3 h in the presence of SRHA (1.1 ± 0.2 μM H2ST/h for all SRHA treatments compared to 95.3 ± 2.4 μM H2ST/h with lactate as the carbon source/electron donor). Further, enhanced methylation could not be attributed to abiotic methylation by DOM. Me201Hg production in uninoculated control treatments with DOM (Figures 1 and 2) was ≤1 pM (method detection limit = 0.7 pM), significantly lower than MeHg production in any active culture assays (p < 0.01). Finally, DOC-driven changes in MeHg production could not be explained by changes in Hg sorption to surfaces, including bottle walls and cell walls. Ligand-driven changes in Hg partitioning can dramatically alter the available Hg concentration and can confound interpretation of bioavailability experiments. To determine how the DOM isolates affected Hg distribution in our experiments, we followed the partitioning of 201Hg between reaction tube walls and aqueous and particulate (0.2 μm) phases (Figures 1 and 2). In these figures, the difference between the measured total Hg concentration in the culture medium (black bars) and the total added Hg (dashed line) represents the loss of Hg from the medium to reaction tube walls. Direct measurements of Hg on bottle walls (by adding digest acid to empty tubes) and lack of elemental Hg production by ND132 in this medium18 confirmed this loss mechanism. Even in these studies with a strong Hg ligand, a substantial fraction of added Hg is lost to bottle walls. The amount of 201THg remaining in the medium was weakly positively correlated with [SRHA] (r2 = 0.63, p < 0.001 for 0.5 nM 201HgCl2 experiments and r2 = 0.39, p = 0.013 for 5.0 nM 201HgCl2 experiments, respectively); but weakly negatively correlated with WLHPoA (r2 = 0.28, p = 0.042). Filterable Hg might be presumed to be more bioavailable than particulate Hg. Overall there was only a very weak relationship between total MeHg and filterable 201THg for any experiments with DOM (Supporting Information Figure S4). These data show that DOM is not enhancing MeHg production simply by keeping Hg in solution and off bottle walls. However, because “filterable” Hg passing a 0.2-μm pore size filter may include nanoparticulate β-HgS(s) and HgS-DOM colloids, these data do not provide information on the role of DOM in Hg solubility per se. As we argue later in the paper, DOM-driven differences in β-HgS(s) precipitation within the filterable pool can dramatically affect availability of Hg for uptake. The addition of ∼500 μM L-cysteine, a thiol-bearing amino acid, to washed cell assays significantly enhanced MeHg production relative to the highest DOM concentrations evaluated (Figures 1 and 2). In assays with added cysteine, virtually all of the filterable 201Hg was converted to Me201Hg within 3 h (Figure S4). These results are consistent with recent findings that cysteine enhances Hg(II)i bioavailability to Hgmethylating organisms, including ND132.6,7 As discussed below, speciation calculations indicate that Hg−cys complexes

Figure 2. Effect of Williams Lake hydrophobic acid (WLHPoA) concentration or L-cysteine (cys) addition on MeHg production by strain ND132. Bars are measurements made at the end of 3 h washed cell assays at 31 °C. Error bars indicate standard deviation of triplicate reaction tubes. Top panel shows the distribution of MeHg between cells and other particulates (gray bars) and the aqueous phase (white bars); total MeHg in the medium is shown with black bars. Bottom panel shows the distribution of total Hg (THg) between phases. The concentration of total added Hg is shown with dashed lines; the difference between the total Hg measured in the medium (black bars in panel b) and the amount added is the loss to reaction tube walls.

added Hg (Figure 1a) and up to ∼10-fold for 0.5 nM Hg (Figure 1c), relative to controls without added DOM or cysteine. The effect of WLHPoA addition on MeHg production was more modest, yielding up to a ∼2-fold increase for 0.5 nM added Hg (Figure 2). Importantly, the majority of MeHg produced by ND132 was rapidly exported (as noted by Gilmour et al.18), and the amount of export was independent of DOM source or concentration (Figures 1 and 2). Across all experiments, 65 ± 23% of the total MeHg produced was filterable. Rapid export of MeHg may represent a detoxification strategy for Hg-methylating bacteria that might otherwise accumulate MeHg. However, there is no evidence suggesting that MeHg production confers resistance to Hg(II)i in Desulfovibrio.18 2718



Article

Formation of β-HgS at 5 nM THg is expected over a wide range of DOM-thiol and sulfide concentrations at KS0 = 36 and pH 7.30 (Figure 4a). The stability field for β-HgS(s) expands with increasing log KS0 (Figure 4b). A slightly smaller stability field for β-HgS(s) is observed at 0.5 nM THg (Figure 4 c,d). Overlaying the “experimental window” of DOM−thiol and sulfide concentrations (assuming that thiols are ∼0.15−0.3% of SRHA12,32 and similar for WLHPoA given its similar total S content) onto these diagrams indicates that DOM concentrations were generally one to many orders of magnitude below that required to prevent β-HgS(s) formation. We intentionally selected a relatively high value for the equilibrium constant for the 1:2 Hg-DOM thiol complex (log K = 42.032), further emphasizing that, even allowing for very strong Hg−DOM binding, DOM is unable to outcompete sulfide for Hg. Our results are consistent with those reported by Wang and Tessier33 who utilized WHAM34 in conjunction with MINEQL to model Hg(II)i speciation in sediment porewaters containing μM concentrations of sulfide and polysulfides. They estimated that sulfide concentrations less than 5 × 10−10 M were required for Hg complexes with humic substances to predominate inorganic Hg(II)i speciation at pH 7. Further, the slope of the plot of log Hg/DOM ratio vs log fraction Hg methylated (Figure 3) is inconsistent with Hg− DOM as the bioavailable complex. Assuming Hg forms twocoordinate complexes with thiol moieties in DOM, the fraction of total Hg as Hg(SR)2 should increase by 2 orders of magnitude for every order of magnitude increase in total thiol (RSHT)/HgT ratio. The SRHA experimental data yield a slope of about −0.5 which is significantly lower than that predicted (−2) based on the stoichiometry of Hg binding to DOM thiols. This supports the idea that Hg complexed directly with thiol moieties in DOM was unlikely to be the form available for uptake and methylation in our assays containing both sulfide and DOM. Role of DOM Character in HgS Bioavailability Comparison of DOM Isolates. Both the SRHA and WLHPoA isolates enhanced MeHg production by ND132, but SRHA was about 7 times more effective at doing so (Figure 5; note that C content is quite similar for these isolates). To make a valid comparison across assays in which medium chemistry, cell density, or activity can vary, we normalized Me 201 Hg production to the positive control for each experimental set (no DOM, 500 μM L-cysteine). SRHA and WLHPoA represent nonmarine end-members of a spectrum of DOM characteristics. SRHA has a higher average molecular weight and aromaticity, reflecting its terrestrial origin. WLHPoA is an aquatic DOM derived primarily from autochthonous sources. It has a lower average molecular weight and is more aliphatic in character (Table 1). The two isolates are roughly comparable in terms of total sulfur content. In the absence of sulfide, Hg complexation with thiol moieties determines the binding strength of an organic material. Haitzer et al.12 demonstrated that DOM aromaticity and molecular weight have limited impact on this interaction. Conversely, DOM size and aromaticity are both strongly associated with the strength of HgS−DOM interactions (e.g., cinnabar dissolution19) and also with MeHg production in natural environments.30,35 Our finding that MeHg production is preferentially enhanced by the more aromatic DOM isolate (SRHA) under conditions of supersaturation with respect to β-HgS(s) suggests these observations may be linked.

dominated Hg(II)i speciation in these assays, preventing βHgS(s) formation. DOM and Hg Concentration Dependence. MeHg production linearly depended upon DOM concentration for both SRHA and WLHPoA (Supporting Information Figure S5). Using SRHA, we evaluated MeHg production by ND132 at two different initial 201HgCl2 concentrations (0.5 and 5.0 nM), but with similar sulfide (5.4 ± 1.0 vs 6.1 ± 0.4 μM) and DOM concentrations. The slope of the regression lines for SRHA concentration (mg/L) vs protein-normalized MeHg (pmol MeHg/mg protein) was about 2-fold greater in the experiments at the higher (5.0 nM) Hg concentrations suggesting that DOM-enhancement of Hg bioavailability becomes more pronounced as Hg concentrations rise. In the absence of DOM, however, MeHg production was greater in the lower Hg concentration experiment. The Hg/DOM ratio was a significant predictor of the fraction of Hg(II)i methylated (Figure 3) across both of the Hg

Figure 3. Fraction of Hg methylated by ND132 decreases with increasing Hg/DOM ratio. Experiments were performed with 0.5 or 5.0 nM 201HgCl2, 0−48.4 mg/L Suwannee River humic acid, and 5.8 ± 0.9 μM sulfide at pH 7.2 ± 0.1. Error bars indicating standard deviations of triplicate methylation assays are smaller than data markers.

concentrations assayed. Experiments with different initial Hg concentrations but identical Hg/DOM ratio yielded similar fractional MeHg. This is consistent with the hypothesis that overall Hg(II)i bioavailability depends upon the distribution of Hg(II)i among various Hg(II)i−ligand complexes each with different degrees of bioavailability. Equilibrium Speciation Modeling. Equilibrium speciation modeling indicates that β-HgS(s) formation occurred during the methylation assays containing Hg, DOM, and sulfide (Figure 4). Using MINEQL+ (v. 4.5, Environmental Research Software), we constructed predominance diagrams for Hg(II)i as a function of the two dominant ligands in the washed cell assays, which are sulfide and the thiol moieties in DOM. Except where noted (Supporting Information Table S1), values of pertinent equilibrium constants were taken directly from the MINEQL+ database. We assumed that Hg forms twocoordinate complexes with DOM−thiols.32 To capture the uncertainty associated with the value of the solubility product (KS0) for β-HgS(s), we performed calculations using log KS0 values spanning the reported uncertainty (log KS0 = 38 ± 2). 2719



Article

Figure 4. Calculated inorganic Hg(II) speciation as a function of DOM thiol concentration ([RSH]T) and total sulfide concentration ([H2S]T) for total Hg ([Hg]T) equal to either 5.0 nM (a and b) or 0.5 nM (c and d) and pH equal to 7.30. Metacinnabar (β-HgS(s)) precipitates in the area indicated by the diagonal lines. Predominant dissolved Hg species (either the sum of all dissolved Hg−S species (ΣHgxSy(2−2x+y) (aq)) or Hg-DOM thiol species (Hg(SR)2 (aq) are listed for regions under-saturated with respect to β-HgS(s). Calculations were performed using two values of the solubility product (KS0) for the reaction Hg2+ + HS− = β-HgS(s) + H+; in (a, c) log KS0 = 36.0, and in (b, d) log KS0 = 40.0.

mechanism for the positive correlations between DOC and MeHg production observed in many aquatic sediments and wetland soils. A growing body of literature shows that DOM slows HgS formation/aggregation8,9,16 and reduces HgS crystallinity.9,10 Dynamic light scattering (DLS) experiments have shown that HgS forms precipitates 10−200 nm in diameter in the presence of DOM (including SRHA), and that the rate of HgS particle growth is inversely related to DOM concentration.8,9 Gerbig et al.10 recently determined Hg(II)i speciation in Hg−DOM− sulfide solutions under similar experimental conditions and noted the formation of Hg−S species with bond lengths consistent with β-HgS(s). Most importantly, EXAFS data collected toward the high end of the Hg/DOC range evaluated here (∼4 × 10−5 mole Hg/mol C) unequivocally show that βHgS(s) particle size and/or crystallinity decreases with decreasing Hg/DOM ratio.10 Due to the low concentrations of Hg(II)i employed in this study, we could not directly observe predicted changes in HgS nanoparticle morphology, size, structure, or aggregation rate as a function of DOM concentration (using techniques such as electron microscopy, X-ray based spectroscopy, or DLS). Several lines of evidence support our hypothesis, however. Previous EXAFS data show that β-HgS(s) becomes smaller or more disordered with increasing DOM concentration.10 Also significant, we found that Hg(II)i was relatively more bioavailable at lower Hg concentrations. Because HgS(s) formation kinetics are anticipated to be slower at lower Hg concentrations,9 this observation is consistent with the idea that smaller, less-organized HgS particles are more bioavailable.

Figure 5. Comparison of the effect of Suwannee River humic acid (SRHA) and Williams Lake hydrophobic acid (WLHPoA) on Hg methylation by ND132. MeHg concentrations were normalized by MeHg production in control experiments with 500 μM L-cysteine. Methylation assays contained 0.5 nM 201HgCl2, 1.8 ± 0.3 μM sulfide (WLHPoA), or 6.1 ± 0.4 μM sulfide (SRHA), and various amounts of DOM. Error bars indicating standard deviations of triplicate methylation assays are smaller than data markers.

Bioavailability of HgS Nanoparticles for Microbial Uptake and Methylation. In our experiments, DOM enhanced Hg biomethylation in solutions supersaturated with respect to β-HgS(s). Based on these data and modeling, we hypothesize that β-HgS(s) nanoparticles are bioavailable to Hgmethylating bacteria, and that DOM enhances their bioavailability by slowing aggregation into larger, more crystalline, less bioavailable forms of HgS. This hypothesis provides a 2720



Article

suggest that inhibitory effects of DOM on nanoparticle−cell surface interactions may be counterbalanced by reduced particle growth in the presence of DOM. Our work builds on the large body of work on the intimate relationship between the Hg and sulfur cycles.8−10,15,16 We hypothesize that observed sulfide-inhibition of MeHg production in natural environments3 may be diminished by DOC. Sulfide concentration has provided a reasonable predictive tool for MeHg production within many environments.3 DOC concentration, or better, the Hg/DOM ratio, may provide additional predictive power, perhaps across ecosystems. Our investigation provides additional insight into observations30,35 that organic-rich environments such as wetlands are hotspots for Hg methylation. Overall, our data demonstrate that Hg uptake and methylation are considerably more complex than previously thought, and existing equilibrium-speciation based models for predicting MeHg production likely underestimate MeHg production in organic-rich sulfidic waters. More broadly, this demonstration of DOM-dependent enhancement of metal uptake implies that existing paradigms for metal acquisition by bacteria (free ion activity model,43 passive diffusion of neutral, lipophilic species model3) may require significant amendment.

Importantly, no DOM-dependent enhancement in Hgmethylation was observed under conditions where β-HgS(s) formation was prevented (Figure 6). To prevent β-HgS

Figure 6. No enhancement in Hg methylation by Suwannee River humic acid (SRHA) addition in solutions under-saturated with respect to metacinnabar. Metacinnabar formation was prevented by addition of 500 μM L-cysteine. Methylation assays contained 0.5 nM 201HgCl2, 9.9 μM sulfide, and 0−38.4 mg/L SRHA. Error bars indicate standard deviations of triplicate methylation assays and where not visible are smaller than data markers.

■

ASSOCIATED CONTENT

* Supporting Information S

Further ancillary and Hg partitioning data for methylation assays and thermodynamic data for equilibrium speciation modeling. This material is available free of charge via the Internet at http://pubs.acs.org.

formation, we added 500 μM of L-cysteine to washed cell assay solutions containing 0.5 nM 201HgCl2, ∼10 μM sulfide, and variable amounts of SRHA. In these experiments, the predicted saturation index for β-HgS(s) was −0.97, and the predicted dominant aqueous complex was Hg−cys. With Hg− cys as the dominant complex, SHRA did not enhance MeHg production, or impact the partitioning of Hg between cells, bottle walls, and the aqueous phase. This implies that DOM is not acting as a cell membrane permeabilizing agent36 enhancing the uptake of all forms of Hg(II)i. Finally, metal−sulfide nanoparticle growth rates in DOM solutions have been shown to negatively correlate with DOM size and aromaticity.37 Our observation that higher molecular weight, more aromatic DOM was more effective than low molecular weight, more aliphatic DOM at enhancing MeHg production under sulfidic conditions further supports the idea that DOM impacts methylation primarily by altering the dynamics of HgS(s) growth. Implications for Hg Uptake and MeHg Production Models. The idea that HgS nanoparticles are bioavailable to Hg-methylating bacteria is a significant departure from the conventional wisdom that only dissolved species are bioavailable to Hg-methylating bacteria.38 We suggest that the neutral HgS uptake hypothesis3 be extended to include HgS nanoparticles. Recent evidence suggests that neutral HgS species, originally identified based on octanol−water partitioning behavior,14 may actually be HgS(s) nanoparticles.8 The nature of Hg uptake from HgS nanoparticles is unknown, but previous studies suggest penetration of sufficiently small particles39 or ligand-exchange/dissolution at the cell surface. Regardless of uptake mechanism, smaller HgS nanoparticles or clusters are likely more bioavailable to Hg-methylating bacteria. Particle size exerts strong control on nanoparticle solubility,40 reactivity,41 and bioavailability39 beyond simple surface area effects. DOM has been proposed to limit adhesion of nanoscale particles to bacterial cell surfaces,42 presumably via electrostatic or steric barriers to nanoparticle−cell surface interaction. Our results

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]; phone: (443) 482-2472. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank A. Bullock, A. Maizel, G. Riedel, T. Bell, K. Butler, and B. Poulin for laboratory assistance and C. Gerbig for his comments on the manuscript. This work was supported by the U.S. Department of Energy under the Subsurface Biogeochemical Research Program, Office of Biological and Environmental Research, Office of Science, through the Mercury Science Focus Area Program at Oak Ridge National Laboratory, by U.S. National Science Foundation grant DEB0351050 to A. Heyes and C. Gilmour, and by the U.S. Geological Survey Priority Ecosystems and Toxics Substances Hydrology Programs. A.M.G. acknowledges support from the Smithsonian Institution Postdoctoral Fellowship Program. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. government.

■

REFERENCES

(1) Gilmour, C. C.; Henry, E. A.; Mitchell, R. Sulfate stimulation of mercury methylation in freshwater sediments. Environ. Sci. Technol. 1992, 26, 2281−2287. (2) Kerin, E. J.; Gilmour, C. C.; Roden, E.; Suzuki, M. T.; Coates, J. D.; Mason, R. P. Mercury methylation by dissimilatory iron-reducing bacteria. Appl. Environ. Microb. 2006, 72, 7919−7921. (3) Benoit, J.; Gilmour, C.; Mason, R.; Heyes, A. Sulfide controls on mercury speciation and bioavailability to methylating bacteria in sediment pore waters. Environ. Sci. Technol. 1999, 33, 951−957.

2721



Article

(4) Barkay, T.; Gillman, M.; Turner, R. Effects of dissolved organic carbon and salinity on bioavailability of mercury. Appl. Environ. Microb. 1997, 63, 4267−4271. (5) Golding, G.; Kelly, C.; Sparling, R.; Loewen, P.; Rudd, J.; Barkay, T. Evidence for facilitated uptake of Hg(II) by Vibrio anguillarum and Escherichia coli under anaerobic and aerobic conditions. Limnol. Oceanogr. 2002, 47, 967−975. (6) Schaefer, J. K.; Morel, F. M. M. High methylation rates of mercury bound to cysteine by Geobacter sulfurreducens. Nat. Geosci. 2009, 2, 123−126. (7) Schaefer, J. K.; Rocks, S. S.; Zheng, W.; Liang, L.; Gu, B.; Morel, F. M. M. Active transport, substrate specificity, and methylation of Hg(II) in anaerobic bacteria. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 8714−8719. (8) Deonarine, A.; Hsu-Kim, H. Precipitation of mercuric sulfide nanoparticles in NOM-containing water: Implications for the natural environment. Environ. Sci. Technol. 2009, 43, 2368−2373. (9) Slowey, A. J. Rate of formation and dissolution of mercury sulfide nanoparticles: The dual role of natural organic matter. Geochim. Cosmochim. Acta 2010, 74, 4693−4708. (10) Gerbig, C.; Kim, C.; Stegemeier, J.; Ryan, J. N.; Aiken, G. R. Formation of nanocolloidal metacinnabar in mercury-DOM-sulfide systems. Environ. Sci. Technol. 2011, 45, 9180−9187. (11) Han, S.; Gill, G.; Lehman, R.; Choe, K. Complexation of mercury by dissolved organic matter in surface waters of Galveston Bay, Texas. Mar. Chem. 2006, 98, 156−166. (12) Haitzer, M.; Aiken, G.; Ryan, J. Binding of mercury(II) to aquatic humic substances: Influence of pH and source of humic substances. Environ. Sci. Technol. 2003, 37, 2436−2441. (13) Gorski, P. R.; Armstrong, D. E.; Hurley, J. P.; Krabbenhoft, D. P. Influence of natural dissolved organic carbon on the bioavailability of mercury to a freshwater alga. Environ. Pollut. 2008, 154, 116−123. (14) Benoit, J.; Mason, R.; Gilmour, C. Estimation of mercury-sulfide speciation in sediment pore waters using octanol-water partitioning and implications for availability to methylating bacteria. Environ. Toxicol. Chem. 1999, 18, 2138−2141. (15) Miller, C. L.; Mason, R. P.; Gilmour, C. C.; Heyes, A. Influence of dissolved organic matter on the complexation of mercury under sulfidic conditions. Environ. Toxicol. Chem. 2007, 26, 624−633. (16) Ravichandran, M.; Aiken, G.; Ryan, J.; Reddy, M. Inhibition of precipitation and aggregation of metacinnabar (mercuric sulfide) by dissolved organic matter isolated from the Florida Everglades. Environ. Sci. Technol. 1999, 33, 1418−1423. (17) Aiken, G. R.; Hsu-Kim, H.; Ryan, J. N. Influence of dissolved organic matter on the environmental fate of metals, nanoparticles, and colloids. Environ. Sci. Technol. 2011, 45, 3196−3201. (18) Gilmour, C. C.; Elias, D. A.; Kucken, A. M.; Brown, S. D.; Palumbo, A. V.; Schadt, C. W.; Wall, J. D. Sulfate-reducing bacterium Desulfovibrio desulfuricans ND132 as a model for understanding bacterial mercury methylation. Appl. Environ. Microb. 2011, 77, 3938− 3951. (19) Waples, J.; Nagy, K.; Aiken, G.; Ryan, J. Dissolution of cinnabar (HgS) in the presence of natural organic matter. Geochim. Cosmochim. Acta 2005, 69, 1575−1588. (20) Jay, J. A.; Murray, K. J.; Gilmour, C. C.; Mason, R. P.; Morel, F. M. M.; Roberts, A. L.; Hemond, H. F. Mercury methylation by Desulfovibrio desulfuricans ND132 in the presence of polysulfides. Appl. Environ. Microb. 2002, 68, 5741−5745. (21) Brown, S. D.; Gilmour, C. C.; Kucken, A. M.; Wall, J. D.; Elias, D. A.; Brandt, C. C.; Podar, M.; Chertkov, O.; Held, B.; Bruce, D. C.; Detter, J. C.; Tapia, R.; Han, C. S.; Goodwin, L. A.; Cheng, J.; Pitluck, S.; Woyke, T.; Mikhailova, N.; Ivanova, N. N.; Han, J.; Lucas, S.; Lapidus, A. L.; Land, M. L.; Hauser, L. J.; Palumbo, A. V. Genome sequence of the mercury-methylating strain Desulfovibrio desulf uricans ND132. J. Bacteriol. 2011, 193, 2078−2079. (22) Gilmour, C. C.; Tuttle, J. H.; Means, J. C. Anaerobic microbial methylation of inorganic tin in estuarine sediment slurries. Microb. Ecol. 1987, 14, 233−242.

(23) Hollweg, T. A.; Gilmour, C. C.; Mason, R. P. Methylmercury production in sediments of Chesapeake Bay and the mid-Atlantic continental margin. Mar. Chem. 2009, 114, 86−101. (24) Aiken, G.; McKnight, D.; Thorn, K.; Thurman, E. Isolation of hydrophilic organic-acids from water using nonionic macroporous resins. Org. Geochem. 1992, 18, 567−573. (25) Weishaar, J. L.; Aiken, G. R.; Bergamaschi, B. A.; Fram, M. S.; Fujii, R.; Mopper, K. Evaluation of specific ultraviolet absorbance as an indicator of the chemical composition and reactivity of dissolved organic carbon. Environ. Sci. Technol. 2003, 37, 4702−4708. (26) Gasper, J.; Aiken, G.; Ryan, J. A critical review of three methods used for the measurement of mercury (Hg2+)-dissolved organic matter stability constants. Appl. Geochem. 2007, 22, 1583−1597. (27) Miller, C. L.; Southworth, G.; Brooks, S.; Liang, L.; Gu, B. Kinetic controls on the complexation between mercury and dissolved organic matter in a contaminated environment. Environ. Sci. Technol. 2009, 43, 8548−8553. (28) Clesceri, L. S., Greenberg, A. E., Eaton, A. D., Eds. Standard Methods for the Examination of Water and Wastewater; American Public Health Association: Washington, DC, 2000. (29) Bradford, M. M. Rapid and sensitive method for quantitation of microgram quantities of protein utilizing principle of protein-dye binding. Anal. Biochem. 1976, 72, 248−254. (30) Mitchell, C. P. J; Gilmour, C. C. Methylmercury production in a Chesapeake Bay salt marsh. J. Geophys. Res., [Biogeosci.] 2008, 113, G00C04. (31) Hintelmann, H.; Ogrinc, N. Determination of stable mercury isotopes by ICP/MS and their application in environmental studies. In Biogeochemistry of Environmentally Important Trace Metals; Cai, Y., Braids, C. O., Eds.; American Chemical Society: Washington, DC, 2003; pp 321−338. (32) Skyllberg, U. Competition among thiols and inorganic sulfides and polysulfides for Hg and MeHg in wetland soils and sediments under suboxic conditions: Illumination of controversies and implications for MeHg net production. J. Geophys. Res., [Biogeosci.] 2008, 113, G00C03. (33) Wang, F.; Tessier, A. Zero-Valent Sulfur and Metal Speciation in Sediment Porewaters of Freshwater Lakes. Environ. Sci. Technol. 2009, 43, 7252−7257. (34) Tipping, E. Modeling the interactions of Hg(II) and methylmercury with humic substances using WHAM/Model VI. Appl. Geochem. 2007, 22, 1624−1635. (35) Hall, B. D.; Aiken, G. R.; Krabbenhoft, D. P.; MarvinDipasquale, M.; Swarzenski, C. M. Wetlands as principal zones of methylmercury production in southern Louisiana and the Gulf of Mexico region. Environ. Pollut. 2008, 154, 124−134. (36) Parent, L.; Twiss, M. R.; Campbell, P. G. C. Influences of natural dissolved organic matter on the interaction of aluminum with the microalga Chlorella: A test of the free-ion model of trace metal toxicity. Environ. Sci. Technol. 1996, 30, 1713−1720. (37) Deonarine, A.; Lau, B. L. T.; Aiken, G.; Ryan, J.; Hsu-Kim, H. Effects of humic substances on precipitation and aggregation of zinc sulfide nanoparticles. Environ. Sci. Technol. 2011, 45, 3217−3223. (38) Benoit, J.; Gilmour, C.; Mason, R. The influence of sulfide on solid phase mercury bioavailability for methylation by pure cultures of Desulfobulbus propionicus (1pr3). Environ. Sci. Technol. 2001, 35, 127− 132. (39) Dehner, C. A.; Barton, L.; Maurice, P. A.; DuBois, J. L. Sizedependent bioavailability of hematite (α-Fe2O3) nanoparticles to a common aerobic bacterium. Environ. Sci. Technol. 2011, 45, 977−983. (40) Gilbert, B.; Banfield, J. F. Molecular-scale processes involving nanoparticulate minerals in biogeochemical systems. Rev. Mineral. Geochem. 2005, 59, 109−155. (41) Liu, J.; Aruguete, D. M.; Murayama, M.; Hochella, M. F. Jr. Influence of size and aggregation on the reactivity of an environmentally and industrially relevant nanomaterial (PbS). Environ. Sci. Technol. 2009, 43, 8178−8183. (42) Li, Z.; Greden, K.; Alvarez, P. J. J.; Gregory, K. B.; Lowry, G. V. Adsorbed polymer and NOM limits adhesion and toxicity of nano 2722



Article

scale zerovalent iron to E. coli. Environ. Sci. Technol. 2010, 44, 3462− 3467. (43) Morel, F. M. M.; Hering, J. G. Principles and Applications of Aquatic Chemistry; John Wiley and Sons, Inc.: New York, 1993.

2723


Dissolved Organic Matter Enhances Microbial Mercury Methylation

Recommend Documents