Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2018 Jun 19;52(15):8521–8529. doi: 10.1021/acs.est.8b00647

Quantification of Mercury Bioavailability for Methylation Using Diffusive Gradient in Thin-Film Samplers

Udonna Ndu †, Geoff A Christensen ‡, Nelson A Rivera †, Caitlin M Gionfriddo ‡, Marc A Deshusses †, Dwayne A Elias ‡, Heileen Hsu-Kim †,*
PMCID: PMC6085726  PMID: 29920204

Abstract

graphic file with name es-2018-006472_0004.jpg

Mercury-contaminated sediment and water contain various Hg species, with a small fraction available for microbial conversion to the bioaccumulative neurotoxin monomethylmercury (MeHg). Quantification of this available Hg pool is needed to prioritize sites for risk management. This study compared the efficacy of diffusive gradient in thin-film (DGT) passive samplers to a thiol-based selective extraction method with glutathione (GSH) and conventional filtration (<0.2 μm) as indicators of Hg bioavailability. Anaerobic sediment slurry microcosms were amended with isotopically labeled inorganic Hg “endmembers” (dissolved Hg2+, Hg-humic acid, Hg-sorbed to FeS, HgS nanoparticles) with a known range of bioavailability and methylation potentials. Net MeHg production (expressed as percent of total Hg as MeHg) over 1 week correlated with mass accumulation of Hg endmembers on the DGTs and only sometimes correlated with the 0.2 μm filter passing Hg fraction and the GSH-extractable Hg fraction. These results suggest for the first time that inorganic Hg uptake in DGTs may indicate bioavailability for methylating microbes. Moreover, the methylating microbial community assessed by hgcA gene abundance was not always consistent with methylation rates between the experiments, indicating that knowledge of the methylating community should target the transcript or protein level. Altogether, these results suggest that DGTs could be used to quantify the bioavailable Hg fraction as part of a method to assess net MeHg production potential in the environment.

Introduction

The persistence of monomethylmercury (MeHg) is an environmental concern due to its propensity to biomagnify in aquatic food webs and impart health risks to humans and wildlife.1,2 MeHg production in the aquatic environment occurs primarily by anaerobic microorganisms harboring proteins encoded by the hgcA and hgcB genes.3 Hg-methylating microorganisms include certain sulfate- and iron-reducing bacteria, methanogens, and other syntrophs3−9 and are ubiquitous across anaerobic niches around the globe. Their presence and abundance appear to be independent of total Hg and MeHg concentrations.8 Thus, the amount of inorganic Hg (Hg(II)) that is bioavailable to these organisms is an important factor for understanding and possibly controlling Hg(II) methylation rates in anaerobic settings.10

While assessments of Hg-contaminated sites require an understanding of Hg bioavailability for methylation, quantification of the bioavailable fraction remains a challenge, partly due to the complexity of Hg speciation in the environment and the poor understanding of the Hg uptake process into methylating microbes. In anaerobic habitats, Hg(II) tends to be predominantly bound to organic matter and sulfides,10−12 both in dissolved and particulate forms. Because of the high affinity of Hg2+ for reduced sulfur binding site, Hg(II) uptake into microorganisms likely involves thiolate-based membrane transporter or biotic ligand.10 Thus, Hg species in soil and water have varying degrees of reactivity and bioavailability at cellular interfaces, resulting in a dependence of MeHg production rates on Hg speciation.13−16

Potential strategies for predicting or quantifying the amount of bioavailable Hg(II) in water and sediments include measurements of the “dissolved” or filter-passing phase (typically 0.2 or 0.45 μm nominal pore size) or the solid-to-aqueous Hg partition coefficient KD where the aqueous fraction is defined by this filter-passing phase. Despite this widely employed approach, the filter-passing Hg concentration or KD rarely correlate with Hg methylation rates or MeHg concentration.15,17 Another approach is to infer Hg speciation using chemical equilibrium speciation models and assume that a subset of dissolved species are bioavailable. However, these models require assumptions of equilibrium conditions that are often unjustified for Hg species undergoing dynamic transformations, or the models do not adequately account for the variety of Hg species (e.g., colloidal phases of Hg) that could fall in the filtered fraction.10 Others have also proposed the use of sequential or selective extractions of Hg as measurements of the biologically reactive fraction.17−20 For example, these methods employ chemical reagents such as a strong acid, strong base, Hg-reducing agent, ethylating agent, or a model biotic ligand such as glutathione (GSH).17−20 However, only one of these methods, the GSH-based selective extraction,17 has been tested in experiments that systematically varied Hg bioavailability in microcosms while controlling for the methylating community. Regardless, these methods are ex situ applications that are susceptible to changes in sample composition (e.g., oxidation, coagulation) during collection and storage, a potential problem for samples originating from anaerobic settings.

Passive samplers such as diffusive gradients in thin films (DGTs), are in situ sampling devices placed in water or sediment and have been widely used for a variety of trace metals, including Hg.21−31 A DGT device comprises a plastic casing that holds a membrane filter and diffusive gel (such as agarose) layered over a chelating resin.32 The high affinity of this resin for Hg results in a concentration gradient over the diffusion gel that drives the time dependent mass uptake of soluble Hg. Rates of Hg uptake and accumulation onto the chelating resin are quantified and typically used to infer the “truly dissolved” concentration of Hg based on steady state diffusive uptake equations.23−27,30,31

Although DGTs have limitations and uncertain assumptions for inferring the dissolved concentration in water and sediment,33 they have been useful for indicating the bioavailability of Hg(II) and MeHg to aquatic macroinvertebrates.28,29 These samplers might also provide information on Hg(II) bioavailability for methylating microorganisms, which are believed to take up Hg(II) and convert it to MeHg intracellularly.34 However, this hypothesis remains to be tested in controlled experiments. A thiolated chelating resin inside a DGT might simulate a thiolated biotic ligand that could drive Hg mass transfer across an extracellular layer of polymeric substances. Thus, we hypothesize that Hg mass accumulation onto DGTs is related to the reactive and bioavailable fraction of Hg for methylating microorganisms.

This hypothesis was tested in anaerobic sediment slurries amended with isotopically labeled Hg species of known spectrum of Hg methylation potential.15,35,36 These forms included dissolved Hg(II) (Hg2+, Hg(II) complexed by dissolved humic acid) and particulate species (nanocrystalline HgS, Hg(II) sorbed to FeS). For each isotopically labeled Hg spike (or “endmember”), we monitored its net conversion to MeHg. Changes to Hg reactivity and speciation were assessed by three approaches: (1) cumulative uptake of Hg into the DGT sampler; (2) quantification of the 0.2 μm filter passing fraction of Hg at each time point; and (3) quantification of the GSH-extractable fraction of Hg in the slurry at each time point.17 Microbial community abundance and diversity, including the Hg-methylating community, were assessed to understand differences in methylation rates between the slurry experiments.

Materials and Methods

Materials

Enriched stable isotopes of Hg were used to synthesize stock solutions of dissolved 204Hg2+, nanoparticulate 200HgS (39 nm average hydrodynamic diameter), 199Hg(II) sorbed to aggregates (>0.2 μm) of nanocrystalline FeS (0.95 μmol 199Hg per g FeS) (Figure S1), and 196Hg(II) bound to dissolved Suwannee humic acid (5 μmol 196Hg per mg of organic C). Preparation of these stock solutions are described in the Supporting Information (SI). Measurements of 198Hg were used to track Hg(II) and MeHg derived from the ambient or “native” mercury of the original sediment. Dissolved stock solutions of 202Hg(II) and Me202Hg were used as internal standards to correct for method extraction efficiencies for total Hg and MeHg analysis. The measured isotopic compositions of all Hg stocks are shown in Table S1.

Each DGT sampler entailed a 25 mm circular plastic housing (DGT Research Ltd., Lancaster, UK) layered with a 0.45 μm nitrocellulose filter, an agarose gel diffusion layer, and a resin binding layer of thiolated silica beads supported on a polyacrylamide gel, as described previously33 and in the SI.

Sediment and surface water samples from two different locations were used for laboratory microcosms. The first location was a mesohaline tidal saltmarsh near the Rhodes River (Edgewater, Maryland, USA). Sediment (top 15 cm, 172 ± 7 ng g–1 Hg dry weight basis) and surface water were collected on 30th September 2015, were stored at 4 °C, and were used for the experiment 1 month after collection. Slurries made from these samples were designated the “mesohaline” microcosm for this study. The second site is a small (approximately 8000 m2) retention pond of Sandy Creek (Durham, North Carolina, USA), a freshwater stream located in an urbanized watershed and known to receive moderate amounts of legacy Hg, possibly originating from historical application of turf grass pesticides in upland areas.37 The sediments (49.6 ± 6 ng g–1 Hg dry weight basis) were collected by hand from the edge of this pond and were used for the “freshwater” slurry microcosms within 2 days of collection on 3rd March 2017.

Slurry Preparation and Experimental Design

We selected batch anaerobic sediment slurries (instead of more complex laboratory or field experiments) to test our hypothesis. Reasons for this approach were to (1) control for a range of Hg bioavailabilities and methylation potentials; (2) incorporate multiple replicates across multiple time points (from 0.5 to 7 days) and microbial communities; and (3) focus on Hg methylation while minimizing confounding factors associated with MeHg degradation.

Two sediment slurry microcosm experiments (i.e., the mesohaline and freshwater microcosms) were performed independently. In both cases, the slurries were prepared by placing 80 g (wet weight) of sediment and 100 mL of surface water in 0.2 L glass jars. The mesohaline slurries (20 individual jars) were amended with 5 μM sodium pyruvate and the freshwater slurries (15 jars) were amended with 1 mM sodium pyruvate as a means to stimulate the methylating microbial community in these slurries. The slurries were sealed with gastight screw caps and incubated at room temperature in the dark for 4 days. An additional replicate slurry with a resazurin indicator (0.001%) changed from dark pink to colorless over this time, indicating the development of anoxic conditions in the jars. After this preincubation period, all further manipulations of the slurry were performed in an anaerobic chamber with a 3% H2/97% N2 atmosphere (Coy Laboratories).

Each anaerobic jar was amended with four forms of inorganic Hg (i.e., “endmembers”): dissolved 204Hg2+, 196Hg(II)-humic acid, nanoparticulate 200HgS, and 199Hg(II) sorbed to FeS. The target Hg content for each endmember was within the range of 50–200 ng g–1 d.w. (the amount of each isotope spike was verified later by analysis of total Hg content in the slurry). The jars were shaken end-over-end for a few seconds between each Hg addition. After all Hg spike additions, a DGT sampler was added to each jar, and the jars were stored under static conditions in the dark (to avoid photodegradation of MeHg). During this incubation, the jars were mixed end-over-end once per day. Replicate jars (n = 3 or 4) were sacrificed periodically from <0.5 d and up to 7 d after Hg isotope addition.

We note that results between the mesohaline and freshwater slurry experiments were not designed to be similar to each other (e.g., they utilized sediment-water from different sites and were stimulated to different extents with organic carbon). Rather, we view these slurries as distinct microbial community mixtures and geochemical compositions for us to test our hypothesis regarding the efficacy of DGTs as indicators for Hg bioavailability and methylation potential.

Microcosm Sampling

At the designated incubation time point, each jar was mixed end-over-end and held static for 10 min to allow for large particles to settle prior to subsampling in the anaerobic chamber. Porewater was collected by filtering 15 mL of the overlying water through a 0.2 μm poly(ether sulfone) filter (VWR). This filtered sample was apportioned into samples designated for total Hg analysis, major cations including Fe, sulfate, and dissolved organic carbon (DOC) analyses.

The whole slurry was then mixed end-over-end again and aliquots of the unfiltered slurry were immediately collected and frozen for later analysis of MeHg, total Hg contents, the GSH-extractable Hg fraction, dry–wet mass ratio, microbial community composition, and abundance of methylating microorganisms. The pH values of the slurries were measured at the time of sampling.

Chemical Analyses

Procedures for all chemical analyses are described in the SI. Isotope-specific MeHg and Hg contents were quantified by cold vapor inductively coupled plasma mass spectrometry (ICP-MS). Other cations were determined by conventional ICP-MS. Dry-wet mass ratio was determined by the gravimetric mass of a whole slurry sample before and after heating in an oven overnight at 105 °C.

The glutathione-extractable fraction of Hg(II) for each sample was determined as described previously.17 In brief, 1 mL of whole slurry (corresponding to 0.13 gdw sediment) was added to 10 mL of deionized water spiked with dissolved GSH to a final concentration of 1 mM and mixed end-over-end for 1 h. The supernatant was then collected and subjected to centrifugation (<1150 RCF for 5 min) and passed through a 0.22 μm polyethersulfone filter. The GSH-extractable Hg for the slurry was defined as the mass concentration of Hg in this filtered supernatant minus the filter-passing Hg concentration in replicate slurries with no GSH added. The GSH-extractable Hg contents are reported by normalizing to the sediment dry weight in the slurry.

The DGT from each microcosm jar was disassembled at the sampling time point. The three layers of the DGT unit (nitrocellulose filter, agarose gel, and thiolated silica resin layers) were each digested separately in 6 mL of concentrated HCl + HNO3 (1:1 volume ratio)33 at 55 °C for 4 h. Dilutions of these extracts were analyzed for isotope-specific total Hg content. As will be discussed later, only a small percentage of the Hg in the slurries was MeHg. Thus, MeHg content in the DGT was not measured. DGT data are generally reported as Hg mass from each isotope spike (and normalized to total Hg in the slurry for that endmember) that accumulated on the DGT resin layer for a specified incubation time.

Total Hg and MeHg contents from each isotopically labeled Hg endmember was calculated via matrix deconvolution on a Microsoft Excel spreadsheet with the measured isotopic contents of Hg and MeHg in the slurries and the measured isotopic composition of individual Hg endmembers as input values to the spreadsheet.38

Microbiological Analyses

Microbial community abundance and diversity was assessed by analysis of genomic DNA (gDNA) via 16S rRNA SSU gene sequencing, and the Hg-methylating community was quantified via PCR-based methods (qualitative broad-range, hgcAB; quantitative clade-specific, hgcA),39 as described previously. gDNA was extracted following the DNeasy PowerSoil Kit with minor modifications32 (Qiagen Inc., Germantown, MD). Final nucleic acid concentrations and purity were measured with a Qubit (Thermo Fisher Scientific) and a NanoDrop One (Thermo Fisher Scientific), respectively. High-throughput sequencing of the V4-5 region of the 16S rRNA gene was performed on an Illumina Miseq 2 × 300 bp system (Illumina, San Diego, CA) using universal primers IllCUs515F (GTGYC7AGCMGCCGCGGTAA) and Lee926R (CCGYCAATTYMTTTRAGTTT).40,41 Sequences were joined, depleted of barcodes, and quality filtered in QIIME 1.9.1.42,43 Demultiplexed sequences were then clustered into operational taxonomic units (OTUs) following a uclust-based44 open-reference OTU picking method45 in QIIME. Sequences were queried against the Greengenes 13_8 reference database at 97% sequence similarity cutoff. Chimeras were removed and QIIME outputs were further analyzed in MEGAN6.46 Hg-methylator presence (hgcAB+) and clade-specific Hg-methylator abundance (hgcA gene) were determined by PCR (5 ng template per reaction) and qPCR (100 pg template per reaction), respectively, as previously described.39 Amplicons for hgcAB were confirmed by TA cloning (Thermo Fisher Scientific) and sequencing (Eurofins Genomics, Louisville, KY) with the degenerate primers, following manufacturer’s protocol. In all cases, the best BLAST hit for a sequence in the NCBI nucleotide database was to hgcAB, indicating that the PCR hybridization was to the target gene and was not a false positive.

Results and Discussion

Hg Isotope Spikes Differ in the Extent of Net Methylation

The amount of net MeHg production depended on the initial form of Hg that was added to the slurry (Figure 1), where the %MeHg values were calculated from MeHg and total Hg values measured in the slurries (Figure S2). Overall the percentages of total Hg that was converted to MeHg were greater for the initially dissolved Hg endmembers (204Hg2+, 196Hg-humic) and lower for the initially particulate endmembers (199Hg-FeS, nano-200HgS, native Hg) in both the mesohaline and freshwater slurries (Figure 1A and B). This trend is generally consistent with previously published sediment culture studies,15,16,47 with some minor differences.

Figure 1.

Figure 1

Open in a new tab

Net production of methylmercury (MeHg) from each isotopically labeled endmember of inorganic Hg added to (A) mesohaline sediment microcosms and (B) freshwater sediment microcosms. Each time point represents the average (± std error) of 4 and 3 replicate slurries for the mesohaline and freshwater experiments, respectively. The percentages are based on measurements of total Hg and MeHg contents in the whole wet slurry of each replicate jar (sacrificially sampled at that time point).

For example, the fractions of MeHg produced from the dissolved 204Hg2+ and 196Hg-humic acid endmembers were approximately equal in both slurry experiments (Figure 1A and B) whereas in the work of Jonsson et al.,35,47 a greater proportion of dissolved Hg2+ was methylated relative to the Hg-humic endmember. A potential explanation for this difference is that we utilized a relatively high Hg:organic carbon ratio for our 196Hg-humic acid endmember, resulting in Hg predominantly bound to high abundance, weak ligand sites (e.g., carboxylates) rather than low abundance, strong ligand sites (e.g., thiolates) on the humic acid.48 Thus, the speciation of both dissolved Hg endmembers changed to similar chemical forms soon after they were added to the slurries, and this change would depend on the slurry conditions. Also, the slurries for this study and for previous studies all differ in terms of the microbial community composition and activity of the methylating community, highlighting the multiple factors that influence overall Hg methylation rates. Regardless of these complexities, the methylation results in the mesohaline and freshwater microcosms demonstrate a spectrum of methylation potentials by the added isotope tracers.

We used the 198Hg isotope to track net methylation of the native Hg in the sediment samples (assuming that the isotopic composition of the native Hg follows the natural distribution). In both mesohaline and freshwater slurries, the percentage of the native Hg as MeHg was lower than the Hg isotope spikes, suggesting that the native Hg had lower bioavailability than the spiked Hg isotopes. However, we also note that a fraction of this native Hg was already in the form of MeHg due to naturally occurring MeHg at the sampling site or production of MeHg during the 4 day preincubation step prior to Hg isotope addition. MeHg degradation processes were also occurring in these slurries, and one might expect that demethylation would have a larger impact in controlling %MeHg values of the native Hg (comprising a mixture of MeHg and inorganic Hg at the initial time point) compared to the other endmembers that were added as inorganic Hg species. Altogether, this distinction with the native Hg, as compared to the spikes, highlights the fact that %MeHg values represent a balance between Hg methylation and MeHg demethylation processes and the net methylation data of the isotopic spikes are not necessarily fully comparable to native Hg. Thus, further analyses of these data focus only on the Hg isotope spikes and exclude measurements of the native Hg fraction.

Hg Isotope Spikes as Quantified by DGTs, Filtration, and GSH-Selective Extraction

Similar to the MeHg data above, the reactivity of Hg (quantified by the DGT or by GSH selective extraction) and partitioning of Hg between solid and aqueous phases varied accordingly with the isotopically labeled endmember species. For the DGT approach, the mass uptake of Hg on the DGT for each endmember (reported as a percentage of total Hg in the slurry from that endmember) was larger for the initially dissolved Hg endmembers (204Hg2+ and 196Hg-humic acid) than the initially particulate endmembers, in both the mesohaline and freshwater slurries (Figure 2A and B). Lower percentages of Hg on the DGTs were observed for the initially solid phase endmembers (199Hg-FeS and nano-200HgS). No difference (one-way ANOVA) was observed between the dissolved Hg and Hg-humic acid endmembers for DGT contents, even though the diffusion coefficients of low molecular weight Hg(II) species (e.g., Hg(OH)x2–x, HgClx) are two times that of Hg(II)-humic complexes.10,26 These results provide additional indication that the speciation and reactivity of these dissolved Hg endmembers were similar in the slurries even though they originated from different initial forms. Therefore, we emphasize that uptake of the individual Hg endmembers into the DGTs was not necessarily as their original chemical form, especially the initially particulate Hg endmembers that contribute to the bioavailable Hg pool by releasing soluble Hg at the interface of a microbial cell (or surface of a DGT sampler).

Figure 2.

Figure 2

Open in a new tab

(A, B) Hg mass accumulated on the thiolated DGT resin layer (reported as a percent of total Hg in the slurry) for each isotopically labeled Hg endmember added to sediment slurry microcosms. (C, D) Percentage of total Hg in filtered pore water (<0.2 μm). (E, F) Percentage in the GSH-extractable fraction, as determined by selective extraction with 1 mM glutathione (GSH). Each time point represents the average of 4 and 3 replicate microcosm slurries for the mesohaline (A, C, E) and freshwater (B, D, F) microcosms, respectively.

The percentage of Hg in the 0.2 μm filtered pore water fraction (Figure 2C and D) and GSH-extractable fraction (Figure 2E and F) also varied between the different Hg isotope spikes, and the conclusions were less clear than those determined with DGT. Specifically, in the mesohaline slurry, the percentages of the sorbed 199Hg-FeS endmember in the filtered porewater fraction was the least for all the forms, a result that could be expected of solid-bound forms of Hg that would be captured by the filter. However, we also observed that the percentage of the nano-200HgS endmember in the filtered porewater was similar to the percentages of the dissolved 204Hg2+ and 196Hg-humic endmembers in the filtered porewater (Figure 2C), an unexpected result based on what had been observed with the DGTs. In contrast, in the freshwater slurries, greater percentages of the 204Hg2+ and 196Hg-humic isotope spikes (0.05 to 0.2%) than the particulate 199Hg-FeS and 200HgS isotope endmembers (0.01 to 0.04%) were observed in the filtered porewater fraction (Figure 2D), consistent with the DGT data.

The GSH-extractable Hg contents were also inconsistent with observations by the DGTs. In the mesohaline slurries, the GSH-extractable Hg content was similar for each isotope endmember at each time point. In contrast, GSH-extractable Hg in the freshwater slurries were generally greater for the dissolved Hg endmembers (2–4%) than the particulate Hg endmembers (1–2%).

We note that the accumulation of the various Hg endmembers on the top layers of the DGT (i.e., the 0.45 μm filter and agarose gel) did not correlate with observed accumulation on the thiolated resin layer (Figure S3). For example, the nano-200HgS endmember had the largest proportion on the filter layer in the mesohaline slurry even though this endmember was one of the lowest to accumulate on the resin layer. In a prior publication it was demonstrated that nano-HgS did not diffuse across agarose gel as a particle. Hence for deposition of 200Hg from nano-200HgS to occur in the resin there must have been dissolution of the nanoparticle at the gel interface. Likewise, the 196Hg-humic acid endmember showed the largest percentage in the agarose gel layer of the DGTs relative to the other endmembers, suggesting that this form was able to remain dissolved in the slurry while the others (such as the dissolved 204Hg2+ endmember) likely partitioned to solid phases immediately after addition to the slurry. Collectively, the differences between the Hg endmembers supports the suggestion that Hg accumulation on the thiolated resin layer of the DGT cannot be simply modeled by a diffusive uptake process based on a dissolved concentration gradient from bulk porewater and across the gel layer due to the predominance of Hg in the particulate form in sediments.33 Rather, uptake into the sampler depends on the reactivity of Hg in the slurry (including Hg bound to solid phases) for release of soluble Hg at the DGT surface. For these reasons, we did not use the Hg mass uptake data with the DGTs to calculate the “truly dissolved” Hg concentration.

Correlations between Hg Methylation and Hg Reactivity

For each time point, the percentage of Hg as MeHg from each endmember was compared to the percentage accumulated on the thiolated resin layer of the DGT, in the filtered pore water fraction, and the GSH-extractable Hg fraction (Figure 3). In both the mesohaline and freshwater slurries, correlations between the %MeHg and %Hg accumulated on the DGTs were observed (R2 ≥ 0.7) for each time point of the experiment (from 0.5 to 7 days).

Figure 3.

Figure 3

Open in a new tab

Percentage of total Hg as MeHg at each time point in the mesohaline microcosms (A, C, E) and the freshwater microcosms (B, D, F) plotted with (A, B) percent of total Hg in the slurry that accumulated on the DGT; (C, D) percent in the <0.2 μm filtered fraction; and (E, F) percent in the GSH-extractable fraction. For each time point category, each data point corresponds to the average across replicate microcosms for a single isotopically labeled Hg spike (204Hg2+, 196Hg-humic, 199Hg-FeS, or nano-200HgS).

Correlations of %MeHg with the percentage of Hg in the filtered pore water fraction (i.e., filtered Hg) or with the GSH-extractable fraction were mixed and generally weaker in the mesohaline slurries than the freshwater slurries. For example, R2 values were <0.1 in plots of %MeHg versus filtered Hg in the mesohaline slurries (Figure 3C), while R2 values for the freshwater slurries were greater than 0.8 (Figure 3D). We also observed that the filtered Hg values in both slurries were not increasing with incubation time (Figure 2C and D) even though the %MeHg value were increasing with time. The freshwater slurries likely contained greater amount of small Hg-bearing particles that could fall in the filtered fraction compared to mesohaline slurries where the higher ionic strength would be expected to destabilize colloidal particles. Therefore, the relevance of this filtered Hg fraction for Hg bioavailability remains uncertain, as suggested by previous reports concluding that filtered porewater Hg content is a poor indicator of Hg bioavailability.15−17,29,35

The relationships between %MeHg and the GSH-extractable fraction were also mixed, with weak correlations observed in the mesohaline slurry (Figure 3E, R2 < 0.2 for each time point) and stronger correlations observed in the freshwater slurries (Figure 3F, R2 > 0.8). This result is inconsistent with our previous study17 proposing the GSH-extractable Hg fraction as an indicator of Hg bioavailability. A key difference with the mesohaline slurry (compared to the freshwater slurry and the previous report by Ticknor et al.17) is that the %MeHg values were less than the %GSH-extractable values, suggesting that factors other than the reactivity of Hg in the bulk slurry (i.e., low microbial abundance and activity) might have limited methylation rates in the mesohaline slurries.

Differences in Biogeochemical Conditions between Slurry Microcosms

The extent of net Hg methylation in the mesohaline slurries was approximately 10 times less than net Hg methylation in the freshwater slurries (Figure 1) due to the drastically different conditions between the two experiments. During slurry preparation, far less pyruvate was added as a carbon substrate to the mesohaline slurries (5 μM) than the freshwater slurries (1 mM). The slurries were also dissimilar in geochemical characteristics relevant for Hg and microbial community composition. For example, sulfate concentrations in the mesohaline slurries were over 500 mg L–1 when the Hg isotopes were added and decreased to approximately 350 mg L–1 over the 1 week incubation (Figure S4A), while in the freshwater slurries dissolved sulfate was less than 1 mg L–1 with no measurable change during the experiment. In contrast, dissolved Fe concentration was low (<0.2 mg L–1) in the mesohaline experiment and increasing in concentration (from 20 to 40 mg L–1) in the freshwater slurries (Figure S4B). During the course of the experiment, the pH was 7.0 (±0.1) and 5.3 (±0.2) in mesohaline and freshwater slurries, respectively. Altogether, these two slurry microcosms differed in the major forms of metabolism (e.g., sulfate- versus iron-reducing microbes), which has implications for the production of metabolites such as inorganic sulfide that change inorganic Hg(II) speciation during the incubation. The relative concentration of sulfide would also influence MeHg speciation, as previous studies have suggested that abiotic MeHg decomposition can occur in sulfidic environments.49−54

Differences in Microbial Community Composition between the Microcosms

Just as geochemical conditions can markedly influence MeHg production rate and extent, so too can the presence and abundance of different methylating species, with Deltaproteobacteria isolates showing the highest and methanogen isolates showing the lowest Hg methylating potential.9 Thus, the substantial variations in Hg methylation rates between the mesohaline and freshwater slurries could also be due to the different methylating communities (Figure S5), which did not appear to significantly change over the duration of the experiment (Figure S6). The dominant microbial classes across both environments included: Bacilli (mesohaline (M) ∼20%; freshwater (F) ∼10%), Alpha- (M 5%; F ∼10%) and Gammaproteobacteria (M ∼10%; F ∼25%), unclassified (M ∼20%; F <10%), and environmental (M <5%; F ∼15%) bacteria, and Actinobacteria (M/F ∼10%). For the major groups of Hg-methylators, Clostridia spp. constituted ∼1% in both environments, while the Methanomicrobia (M ∼2%; F ∼0.1%) and Deltaproteobacteria (M ∼0.1%; F ∼1%) showed opposing trends. Combined, potential Hg-methylators constituted at most ∼2.5–4.5% (mesohaline) and 1–2% (freshwater) of the community, confirming previous findings that Hg-methylators constitute a small fraction of the total microbial community8 and yet have a great impact on MeHg production.55

However, 16S rRNA gene sequencing cannot provide reliable and robust identification of Hg-methylator diversity and relative abundance.55 When comparing percent Hg-methylators in the samples, the mesohaline had more potential methylators, but the MeHg concentrations (as a percent of total Hg) for the mesohaline samples where about 10-fold less, regardless of Hg endmember. This observation suggests that even if Hg bioavailability is taken into consideration, MeHg production potential of an anaerobic microbiome is not solely dependent on methylator abundance but is likely in combination with other factors, including transcript/protein abundance, hydrobiogeochemistry, and geochemistry as discussed above.

To better characterize the Hg-methylating community, PCR-based techniques targeting hgcAB (qualitative) and hgcA (quantitative)39 were applied. For all samples, hgcAB was confirmed via a correctly sized PCR-product (data not shown). Additionally, a subset of PCR product was cloned and sequenced and positively identified in the nucleotide NCBI database as hgcA, most often from uncultured microorganisms (data not shown). For the mesohaline and freshwater samples, hgcA genome copy number (per template comprising 5 ng DNA) was empirically determined: Deltaproteobacteria (M 1330 (±280) × 104; F 6.9 (±6.57) × 104), Firmicutes (M/F below detection, <100), and methanogenic Archaea (M below detection, <103; F 6.6 ± 1.0 × 103) (Figure S7). It is worth noting that the raw value cannot be compared among clades.39,55

The qPCR results are notably inconsistent with the potential methylator results observed by 16S rRNA gene sequencing, a discrepancy that also has been observed in other recent studies.55 However, the PCR results are consistent with expectation in that Deltaproteobacteria were greater in the mesohaline samples than in the freshwater samples, while methanogenic Archaea were greater in the freshwater samples than in the mesohaline. Firmicutes were not found with the PCR method in either experiment, likely due to limitations of the protocol that is known to target sulfate-/sulfite-reducing Firmicutes and not fermenting strains (i.e., Clostridia).39 These results highlight the differences between the two slurry conditions and the composition of the methylating microbial community and that the rates of methylation (per unit of bioavailable Hg) would be influenced in part by the community makeup.

Implications for Monitoring for Hg Bioavailability and Methylation Potential

Overall, the %MeHg values from the Hg isotope spikes correlated with the DGT data for both slurry experiments while correlations of %MeHg with the filtered porewater Hg and GSH-extractable Hg were observed only in the freshwater experiments. Furthermore, methylation rates were approximately 10 times faster for the freshwater than for the mesohaline experiment even though Hg bioavailability (as indicated by DGTs) was similar in both slurries. This difference in methylation rates was not explained by the genomic sequencing and hgcA gene abundance data. Further work in assessing the activity of the methylating community is needed.

While the DGTs produced better correlations with net Hg methylation than the other measures of bioavailability (i.e., filtered Hg and GSH-extractable Hg fractions), it is important to note that the DGT data are cumulative over the incubation time while the other two methods represent measurements at discrete time points. This distinction could be important in settings where Hg speciation is changing with time, as it likely was changing in the slurries. For example, the mesohaline slurry experiment showed a large change in sulfate concentration over 5 days (Δ[SO42–] ∼ −150 mg/L or −1.6 mM). The production of sulfide can change the speciation of Hg isotopes and their reactivity during the incubation period.12 Our previous work has shown that the bioavailability of Hg associated with freshly precipitated Hg-sulfide particles depends on the age of the particles, and conventional filtration methods do not adequately fractionate these forms.15,33

In the freshwater slurries, we did not observe measurable changes in sulfate. Thus, we hypothesize that sulfide precipitation was less important for Hg speciation. Moreover, the total Hg content in the slurry appeared to increase with time during the freshwater experiment and not in the mesohaline experiment (Figure S2). Measurements of Hg contents in the headspace are not available to complete the mass balance. Nevertheless, there may have been production and evasion of Hg0 in the headspace of the jars that slowly repartitioned to the slurry over time, replenishing the bioavailable pool that could be measured by the DGT, filtration, and GSH-extraction methods.

In summary, the results support the use of DGTs in tracer studies to track changes in bioavailability of various endmembers for methylating microbes. This study also shows that measurements of net Hg methylation potential in water and sediments require more than DGTs (or other similar measures of inorganic Hg reactivity). For example, MeHg degradation potential is an important component of net MeHg production. The slurry experiments in this study were relatively short (few days); thus, we believe that changes in MeHg concentration for the isotope spikes were dominated by Hg methylation processes rather than MeHg degradation processes for this time scale. In real settings, however, both methylation and demethylation processes would need to be considered in assessing the overall net MeHg production potential. Moreover, fluctuations of water chemistry and microbial community composition in the environment could also alter inorganic Hg speciation and bioavailability, and future testing of DGTs under field conditions and for longer time scales are needed. These field tests will need to quantify or control for gradients in activity of the Hg-methylating community and processes of MeHg degradation. In this respect, the slurry experiments described here with Hg-isotope spikes enabled an assessment of the DGTs and other measures of Hg bioavailability in a simulated environment that controlled for biological growth conditions and MeHg decomposition rates. With measurements of microbial processes as well as Hg bioavailability, researchers and site managers can improve their understanding of the factors controlling Hg methylation potential at field sites and predict how perturbations such as remediation and climate change might alter MeHg production.

Acknowledgments

We thank Grace Schwartz for collection of sediment and water samples. This study was supported by research grants from DuPont and the National Institute of Environmental Health Sciences Superfund Research Program (R01ES24344).

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.8b00647.

  • Additional descriptions of materials and methods, isotopic composition of Hg stock solutions, total Hg, MeHg, sulfate, and dissolved iron contents in slurries, and results of the microbial community sequencing and qPCR (PDF)

The authors declare no competing financial interest.

Supplementary Material

es8b00647_si_001.pdf (424.7KB, pdf)

References

  1. Hsu-Kim H.; Eckley C. S.; Achá D.; Feng X.; Gilmour C. C.; Jonsson S.; Mitchell C. P. Challenges and opportunities for managing aquatic mercury pollution in altered landscapes. Ambio 2018, 47, 141–169. 10.1007/s13280-017-1006-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Eagles-Smith C. A.; Silbergeld E. K.; Basu N.; Diaz-Barriga F.; Hopkins W. A.; Kidd K. A.; Nyland J. F.; Bustamante P. Modulators of mercury risk to wildlife and humans in the context of rapid global change. Ambio 2018, 47, 170–197. 10.1007/s13280-017-1011-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Parks J. M.; Johs A.; Podar M.; Bridou R.; Hurt R. A. Jr; Smith S. D.; Tomanicek S. J.; Qian Y.; Brown S. D.; Brandt C. C.; Palumbo A. V.; Smith J. C.; Wall J. D.; Elias D. A.; Liang L. The genetic basis for bacterial mercury methylation. Science 2013, 339 (6125), 1332–1335. 10.1126/science.1230667. [DOI] [PubMed] [Google Scholar]
  4. Yu R. Q.; Reinfelder J. R.; Hines M. E.; Barkay T. Mercury methylation by the methanogen Methanospirillum hungatei. Appl. Environ. Microbiol. 2013, 79, 6325–6330. 10.1128/AEM.01556-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Fleming E. J.; Mack E. E.; Green P. G.; Nelson D. C. Mercury methylation from unexpected sources: molybdate-inhibited freshwater sediments and an iron-reducing bacterium. Appl. Environ. Microbiol. 2006, 72 (1), 457–464. 10.1128/AEM.72.1.457-464.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Compeau G. C.; Bartha R. Sulfate-reducing bacteria: principal methylators of mercury in anoxic estuarine sediment. Appl. Environ. Microbiol. 1985, 50 (2), 498–502. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Watras C. J.; Bloom N. S. Mercury and methylmercury in individual zooplankton: Implications for bioaccumulation. Limnol. Oceanogr. 1992, 37, 1313–1318. 10.4319/lo.1992.37.6.1313. [DOI] [Google Scholar]
  8. Podar M.; Gilmour C. C.; Brandt C. C.; Soren A.; Brown S. D.; Crable B. R.; Palumbo A. V.; Somenahally A. C.; Elias D. A. Global prevalence and distribution of genes and microorganisms involved in mercury methylation. Sci. Adv. 2015, 1 (9), e1500675. 10.1126/sciadv.1500675. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Gilmour C. C.; Podar M.; Bullock A. L.; Graham A. M.; Brown S. D.; Somenahally A. C.; Johs A.; Hurt R. A. Jr; Bailey K. L.; Elias D. A. Mercury methylation by novel microorganisms from new environments. Environ. Sci. Technol. 2013, 47 (20), 11810–11820. 10.1021/es403075t. [DOI] [PubMed] [Google Scholar]
  10. Hsu-Kim H.; Kucharzyk K. H.; Zhang T.; Deshusses M. A. Mechanisms regulating mercury bioavailability for methylating microorganisms in the aquatic environment: a critical review. Environ. Sci. Technol. 2013, 47 (6), 2441–2456. 10.1021/es304370g. [DOI] [PubMed] [Google Scholar]
  11. Liem-Nguyen V.; Skyllberg U.; Bjorn E. Thermodynamic Modeling of the Solubility and Chemical Speciation of Mercury and Methylmercury Driven by Organic Thiols and Micromolar Sulfide Concentrations in Boreal Wetland Soils. Environ. Sci. Technol. 2017, 51 (7), 3678–3686. 10.1021/acs.est.6b04622. [DOI] [PubMed] [Google Scholar]
  12. Poulin B. A.; Gerbig C. A.; Kim C. S.; Stegemeier J. P.; Ryan J. N.; Aiken G. R. Effects of Sulfide Concentration and Dissolved Organic Matter Characteristics on the Structure of Nanocolloidal Metacinnabar. Environ. Sci. Technol. 2017, 51 (22), 13133–13142. 10.1021/acs.est.7b02687. [DOI] [PubMed] [Google Scholar]
  13. Graham A. M.; Aiken G. R.; Gilmour C. C. Dissolved organic matter enhances microbial mercury methylation under sulfidic conditions. Environ. Sci. Technol. 2012, 46 (5), 2715–2723. 10.1021/es203658f. [DOI] [PubMed] [Google Scholar]
  14. Pham A. L.; Morris A.; Zhang T.; Ticknor J.; Levard C.; Hsu-Kim H. Precipitation of nanoscale mercuric sulfides in the presence of natural organic matter: Structural properties, aggregation, and biotransformation. Geochim. Cosmochim. Acta 2014, 133, 204–215. 10.1016/j.gca.2014.02.027. [DOI] [Google Scholar]
  15. Zhang T.; Kucharzyk K. H.; Kim B.; Deshusses M. A.; Hsu-Kim H. Net methylation of mercury in estuarine sediment microcosms amended with dissolved, nanoparticulate, and microparticulate mercuric sulfides. Environ. Sci. Technol. 2014, 48 (16), 9133–9141. 10.1021/es500336j. [DOI] [PubMed] [Google Scholar]
  16. Jonsson S.; Skyllberg U.; Nilsson M. B.; Westlund P.; Shchukarev A.; Lundberg E.; Björn E. Mercury Methylation Rates for Geochemically Relevant HgII Species in Sediments. Environ. Sci. Technol. 2012, 46 (21), 11653–11659. 10.1021/es3015327. [DOI] [PubMed] [Google Scholar]
  17. Ticknor J. L.; Kucharzyk K. H.; Porter K. A.; Deshusses M. A.; Hsu-Kim H. Thiol-Based Selective Extraction Assay to Comparatively Assess Bioavailable Mercury in Sediments. Environ. Eng. Sci. 2015, 32 (7), 564–573. 10.1089/ees.2014.0526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Bloom N. S.; Preus E.; Katon J.; Hiltner M. Selective extractions to assess the biogeochemically relevant fractionation of inorganic mercury in sediments and soils. Anal. Chim. Acta 2003, 479 (2), 233–248. 10.1016/S0003-2670(02)01550-7. [DOI] [Google Scholar]
  19. Benoit J. M.; Gilmour C. C.; Mason R. P. The influence of sulfide on solid-phase mercury bioavailability for methylation by pure cultures of Desulfobulbus propionicus (1pr3). Environ. Sci. Technol. 2001, 35 (1), 127–132. 10.1021/es001415n. [DOI] [PubMed] [Google Scholar]
  20. Liang L.; Horvat M.; Alvarez J.; Young L.; Kotnik J.; Zhang L. The Challenge and Its Solution When Determining Biogeochemically Reactive Inorganic Mercury (RHg): Getting the Analytical Method Right. Am. J. Anal. Chem. 2013, 4 (11), 623–632. 10.4236/ajac.2013.411074. [DOI] [Google Scholar]
  21. Davison W.; Zhang H. Progress in understanding the use of diffusive gradients in thin films (DGT)–back to basics. Environmental Chemistry 2012, 9 (1), 1–13. 10.1071/EN11084. [DOI] [Google Scholar]
  22. Davlson W.; Zhang H. In situ speciation measurements of trace components in natural waters using thin-film gels. Nature 1994, 367 (6463), 546–548. 10.1038/367546a0. [DOI] [Google Scholar]
  23. Divis P.; Szkandera R.; Brulik L.; Docekalova H.; Matus P.; Bujdos M. Application of new resin gels for measuring mercury by diffusive gradients in a thin-films technique. Anal. Sci. 2009, 25 (4), 575–578. 10.2116/analsci.25.575. [DOI] [PubMed] [Google Scholar]
  24. Divis P.; Leermakers M.; Docekalova H.; Gao Y. Mercury depth profiles in river and marine sediments measured by the diffusive gradients in thin films technique with two different specific resins. Anal. Bioanal. Chem. 2005, 382 (7), 1715–1719. 10.1007/s00216-005-3360-8. [DOI] [PubMed] [Google Scholar]
  25. Merritt K. A.; Amirbahman A. Mercury mobilization in estuarine sediment porewaters: a diffusive gel time-series study. Environ. Sci. Technol. 2007, 41 (3), 717–722. 10.1021/es061659t. [DOI] [PubMed] [Google Scholar]
  26. Hong Y. S.; Rifkin E.; Bouwer E. J. Combination of diffusive gradient in a thin film probe and IC-ICP-MS for the simultaneous determination of CH3Hg+ and Hg2+ in oxic water. Environ. Sci. Technol. 2011, 45 (15), 6429–6436. 10.1021/es200398d. [DOI] [PubMed] [Google Scholar]
  27. Fernandez-Gomez C.; Dimock B.; Hintelmann H.; Diez S. Development of the DGT technique for Hg measurement in water: comparison of three different types of samplers in laboratory assays. Chemosphere 2011, 85 (9), 1452–1457. 10.1016/j.chemosphere.2011.07.080. [DOI] [PubMed] [Google Scholar]
  28. Clarisse O.; Lotufo G. R.; Hintelmann H.; Best E. P. H. Biomonitoring and assessment of monomethylmercury exposure in aqueous systems using the DGT technique. Sci. Total Environ. 2012, 416, 449–454. 10.1016/j.scitotenv.2011.11.077. [DOI] [PubMed] [Google Scholar]
  29. Amirbahman A.; Massey D. I.; Lotufo G.; Steenhaut N.; Brown L. E.; Biedenbach J. M.; Magar V. S. Assessment of mercury bioavailability to benthic macroinvertebrates using diffusive gradients in thin films (DGT). Environ. Sci. Process. Impacts 2013, 15 (11), 2104–2114. 10.1039/c3em00355h. [DOI] [PubMed] [Google Scholar]
  30. Hong Y.; Dan N. P.; Kim E.; Choi H. J.; Han S. Application of diffusive gel-type probes for assessing redox zonation and mercury methylation in the Mekong Delta sediment. Environ. Sci. Process. Impacts 2014, 16 (7), 1799–1808. 10.1039/C3EM00728F. [DOI] [PubMed] [Google Scholar]
  31. Noh S.; Hong Y. S.; Han S. Application of diffusive gradients in thin films and core centrifugation methods to determine inorganic mercury and monomethylmercury profiles in sediment porewater. Environ. Toxicol. Chem. 2016, 35 (2), 348–356. 10.1002/etc.3193. [DOI] [PubMed] [Google Scholar]
  32. Zhang H.; Davison W. Use of diffusive gradients in thin-films for studies of chemical speciation and bioavailability. Environmental Chemistry 2015, 12 (2), 85–101. 10.1071/EN14105. [DOI] [Google Scholar]
  33. Pham A. L.; Johnson C.; Manley D.; Hsu-Kim H. Influence of Sulfide Nanoparticles on Dissolved Mercury and Zinc Quantification by Diffusive Gradient in Thin-Film Passive Samplers. Environ. Sci. Technol. 2015, 49 (21), 12897–12903. 10.1021/acs.est.5b02774. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. 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. Microbiol. 2011, 77 (12), 3938–3951. 10.1128/AEM.02993-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Jonsson S.; Skyllberg U.; Nilsson M. B.; Westlund P. O.; Shchukarev A.; Lundberg E.; Bjorn E. Mercury methylation rates for geochemically relevant Hg(II) species in sediments. Environ. Sci. Technol. 2012, 46 (21), 11653–11659. 10.1021/es3015327. [DOI] [PubMed] [Google Scholar]
  36. Zhang T.; Kim B.; Levard C.; Reinsch B. C.; Lowry G. V.; Deshusses M. A.; Hsu-Kim H. Methylation of mercury by bacteria exposed to dissolved, nanoparticulate, and microparticulate mercuric sulfides. Environ. Sci. Technol. 2012, 46 (13), 6950–6958. 10.1021/es203181m. [DOI] [PubMed] [Google Scholar]
  37. Deonarine A.; Hsu-Kim H.; Zhang T.; Cai Y.; Richardson C. J. Legacy source of mercury in an urban stream–wetland ecosystem in central North Carolina, USA. Chemosphere 2015, 138 (Supplement C), 960–965. 10.1016/j.chemosphere.2014.12.038. [DOI] [PubMed] [Google Scholar]
  38. Hintelmann H.; Ogrinc N.. Determination of Stable Mercury Isotopes by ICP/MS and Their Application in Environmental Studies, In Biogeochemistry of Environmentally Important Trace Elements; American Chemical Society, 2002; Vol. 835; pp 321–338. [Google Scholar]
  39. Christensen G. A.; Wymore A. M.; King A. J.; Podar M.; Hurt R. A. Jr; Santillan E. U.; Soren A.; Brandt C. C.; Brown S. D.; Palumbo A. V.; Wall J. D.; Gilmour C. C.; Elias D. A. Development and Validation of Broad-Range Qualitative and Clade-Specific Quantitative Molecular Probes for Assessing Mercury Methylation in the Environment. Appl. Environ. Microbiol. 2016, 82 (19), 6068–6078. 10.1128/AEM.01271-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Parada A. E.; Needham D. M.; Fuhrman J. A. Every base matters: assessing small subunit rRNA primers for marine microbiomes with mock communities, time series and global field samples. Environ. Microbiol. 2016, 18 (5), 1403–1414. 10.1111/1462-2920.13023. [DOI] [PubMed] [Google Scholar]
  41. Quince C.; Lanzen A.; Davenport R. J.; Turnbaugh P. J. Removing noise from pyrosequenced amplicons. BMC Bioinf. 2011, 12, 38–12–38. 10.1186/1471-2105-12-38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Caporaso J. G.; Kuczynski J.; Stombaugh J.; Bittinger K.; Bushman F. D.; Costello E. K.; Fierer N.; Pena A. G.; Goodrich J. K.; Gordon J. I.; Huttley G. A.; Kelley S. T.; Knights D.; Koenig J. E.; Ley R. E.; Lozupone C. A.; McDonald D.; Muegge B. D.; Pirrung M.; Reeder J.; Sevinsky J. R.; Turnbaugh P. J.; Walters W. A.; Widmann J.; Yatsunenko T.; Zaneveld J.; Knight R. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 2010, 7 (5), 335–336. 10.1038/nmeth.f.303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Bokulich N. A.; Subramanian S.; Faith J. J.; Gevers D.; Gordon J. I.; Knight R.; Mills D. A.; Caporaso J. G. Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing. Nat. Methods 2013, 10 (1), 57–59. 10.1038/nmeth.2276. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Edgar R. C. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 2010, 26 (19), 2460–2461. 10.1093/bioinformatics/btq461. [DOI] [PubMed] [Google Scholar]
  45. Rideout J. R.; He Y.; Navas-Molina J. A.; Walters W. A.; Ursell L. K.; Gibbons S. M.; Chase J.; McDonald D.; Gonzalez A.; Robbins-Pianka A.; Clemente J. C.; Gilbert J. A.; Huse S. M.; Zhou H. W.; Knight R.; Caporaso J. G. Subsampled open-reference clustering creates consistent, comprehensive OTU definitions and scales to billions of sequences. PeerJ 2014, 2, e545. 10.7717/peerj.545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Huson D. H.; Beier S.; Flade I.; Gorska A.; El-Hadidi M.; Mitra S.; Ruscheweyh H. J.; Tappu R. MEGAN Community Edition - Interactive Exploration and Analysis of Large-Scale Microbiome Sequencing Data. PLoS Comput. Biol. 2016, 12 (6), e1004957. 10.1371/journal.pcbi.1004957. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Jonsson S.; Andersson A.; Nilsson M. B.; Skyllberg U.; Lundberg E.; Schaefer J. K.; Åkerblom S.; Björn E. Terrestrial discharges mediate trophic shifts and enhance methylmercury accumulation in estuarine biota. Sci. Adv. 2017, 3 (1), e1601239. 10.1126/sciadv.1601239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Haitzer M.; Aiken G. R.; Ryan J. N. Binding of mercury(II) to dissolved organic matter: the role of the mercury-to-DOM concentration ratio. Environ. Sci. Technol. 2002, 36 (16), 3564–3570. 10.1021/es025699i. [DOI] [PubMed] [Google Scholar]
  49. Deacon J. B. Volatilization of methyl-mercuric chloride by hydrogen sulphide. Nature 1978, 275, 344. 10.1038/275344a0. [DOI] [Google Scholar]
  50. Baldi F.; Parati F.; Filippelli M. Dimethylmercury and dimethylmercury-sulfide of microbial origin in the biogeochemical cycle of Hg. Water, Air, Soil Pollut. 1995, 80, 805–805. 10.1007/BF01189732. [DOI] [Google Scholar]
  51. Craig P. J.; Moreton P. A. The role of sulphide in the formation of dimethyl mercury in river and estuary sediments. Mar. Pollut. Bull. 1984, 15 (11), 406–408. 10.1016/0025-326X(84)90257-1. [DOI] [Google Scholar]
  52. Jonsson S.; Mazrui N. M.; Mason R. P. Dimethylmercury Formation Mediated by Inorganic and Organic Reduced Sulfur Surfaces. Sci. Rep. 2016, 6, 27958. 10.1038/srep27958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wallschläger D.; Hintelmann H.; Evans R. D.; Wilken R. D. Volatilization of dimethylmercury and elemental mercury from River Elbe floodplain soils. Water, Air, Soil Pollut. 1995, 80, 1325. 10.1007/BF01189798. [DOI] [Google Scholar]
  54. Kanzler C. R.; Lian P.; Trainer E. L.; Yang X.; Govind N.; Parks J. M.; Graham A. Emerging investigator series: Methylmercury Speciation and Dimethylmercury Production in Sulfidic Solutions. Environ. Sci.: Processes Impacts 2018, 20, 584–594. 10.1039/C7EM00533D. [DOI] [PubMed] [Google Scholar]
  55. Christensen G. A.; Somenahally A. C.; Moberly J. G.; Miller C. M.; King A. J.; Gilmour C. C.; Brown S. D.; Podar M.; Brandt C. C.; Brooks S. C.; Palumbo A. V.; Wall J. D.; Elias D. A. Carbon Amendments Alter Microbial Community Structure and Net Mercury Methylation Potential in Sediments. Appl. Environ. Microbiol. 2018, 84, e01049–17. 10.1128/AEM.01049-17. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

es8b00647_si_001.pdf (424.7KB, pdf)

Articles from Environmental Science & Technology are provided here courtesy of American Chemical Society

RESOURCES