2f-3: Mercury methylation: microbial and geochemical constraints

Freshwaters throughout the Globe are contaminated by mercury (Hg) due to anthropogenic activities. Considerable part of the global atmospheric Hg deposition takes place in boreal regions. Biomining that exploits bacterial leaching on-site is suggested as a clean technology to recover metals from low-grain ores. However, biomining effluents have been shown to cause salting and multi-metal contamination of the effluent-receiving waterways. Risk assessors are challenged by the multitude of adverse effects biomining effluents are suggested to have towards freshwater ecosystems. Of those the potential enhancement of Hg biomethylation to methylmercury (MeHg) is surely not the least in importance. Mercury biomethylation in surficial sediments can become enhanced due to sulfate-containing effluents that enhance hypoxic and anoxic conditions in the lake bottoms that again favor activity of sulfate reducing bacteria besides of the sulfate addition itself. These bacteria can methylate Hg to MeHg that biomagnifies in the food web. We sampled surficial sediments from biomining-affected Lake Kalliojärvi and Lake Jormasjärvi, the sedimentary basin of the latter being mixed on the spot, and from one non-mining-affected Lake Ukonjärvi. The sediments showed Hg accumulation at 106.9 (L. Kalliojärvi), 293.1 (L. Jormasjärvi) and 372.1 (L. Ukonjärvi) ng Total Hg g-1 dw with MeHg proportions being 3.8, 0.6 and 1.2 % of Total Hg (THg), respectively. We incubated the sediments at laboratory set-ups for 14 days with their natural bacteria present with two treatments: 1) normoxia and 2) hypoxia in the sediment overlaying water, measured the THg-MeHg proportions in the sediments thereafter and compared them to various water and sediment characteristics. The results indicated our hypotheses that 1) hypoxia enhances Hg biomethylation compared to normoxia in the non-mining-affected sediment, and that 2) the mining-affected anomaly-sediments likely show deviation from this phenomenon, were true. In the heavily contaminated L. Kalliojärvi sediment the highest proportions of MeHg were measured at 14-d with some but not as evident hypoxia effect as in the non-mining-affected L. Ukonjärvi sediment. In the moderately mining-affected L. Jormasjärvi sediment there was no difference between the treatments in Hg biomethylation. Methylmercury in sediment correlated negatively with redox potential in the overlaying water, sediment dry matter content and iron concentration (Pearson’s r-values: -0.31 - -0.55), and positively with water sulfate and iron concentrations (0.59-0.69). Mercury biomethylation can be enhanced in mining-affected sediments. Yet the conditions favorable for Hg methylation are complex, and enhancement of Hg biomethylation will probably not always be the prevailing case in mining-affected freshwaters.

Permeable reactive barriers (PRBs) remove nitrogen from groundwater by enhancing denitrification. The PRBs consist of woodchips that provide a virtually unlimited carbon source for heterotrophic denitrifiers. This carbon also supports other anaerobic bacteria, some of which, including sulfate-reducing bacteria (SRB), have the ability to methylate inorganic mercury that occurs in groundwater in industrialized areas. We examined microbial community composition and geochemistry in flow-through PRB mesocosms, half of which were spiked with sulfate throughout the experiment to simulate a salt water environment. Halfway through the experiment, we began spiking all replicates daily with mercuric chloride. We hypothesized that mercury addition would alter community composition to favor higher abundances of known methylating taxa, and that the sulfate mesocosms would produce more methylmercury than the non-sulfate mesocosms due mainly to an increase in abundance of SRB. We found that taxa of SRB, including desulfobacterales, were more abundant after spiking with mercury, while taxa inhibited by Hg, like arcobacter, decreased. We also found a period of adjustment after the start of Hg spiking, where nitrate removal became incomplete and methylmercury production was low, but after two days, nitrate removal became complete again and methylmercury production rose. Overall, however, there was no net increase in methylmercury production over the course of the experiment, and no significant difference in production between sulfate and non-sulfate treatments. This could be due to inhibition of methylation from high sulfide concentrations, or demethylation by microbes possessing the mer operon, which were also more abundant after spiking with mercuric chloride.

Microbial methylation and demethylation are two competing processes controlling the net production and bioaccumulation of neurotoxic methylmercury (MeHg) in natural aquatic environments. Although mercury (Hg) methylation by anaerobic microorganisms and demethylation by aerobic Hg-resistant bacteria have both been extensively studied, little attention has been given to microbial degradation of MeHg, particularly by anaerobic iron reducers and aerobic methanotrophs, despite their ubiquitous presence in the environment. We report that the iron-reducing bacterium Geobacter bemidjiensis Bem can both methylate inorganic Hg and degrade MeHg concurrently under anoxic conditions. A reductive demethylation pathway is likely utilized by G. bemidjiensis to degrade MeHg, with elemental Hg(0) as the major reaction product, possibly due to the presence of homologs encoding both organo-mercurial lyase (MerB) and mercuric reductase (MerA) in this organism. Additionally, G. bemidjiensis Bem cells can mediate multiple reactions including Hg sorption, reduction and oxidation, resulting in both time and concentration dependent Hg species transformations. For the first time, we also demonstrate that some methanotrophs (e.g., Methylosinus trichosporium OB3b) can take up and degrade MeHg rapidly despite they do not possess merB or merA in their genome. Demethylation by M. trichosporium OB3b increased with increasing MeHg concentrations (up to ~75 nM), but a high MeHg concentration (125 nM) decreased demethylation due to MeHg toxicity. Unlike many known Hg-methylators, all methanotrophs are found to take up substantial amounts of MeHg, likely as a one-carbon (C1) growth substrate and energy source. These findings indicate a cycle of methylation and demethylation among anaerobic and aerobic bacteria, and suggest that both anaerobic iron reducers and aerobic methanotrophs may play an important role in the net balance of MeHg production in the environment.

Tests on natural sediments were performed to measure the effect of ferric oxide on mercury methylation and mobility, by shifting the microbial community and increasing the sorption capacity of the sediment. Iron oxide treatment was tested on replicate chambers having either oxic or anoxic overlying water, and untreated controls were tested on replicate chambers simultaneously. A single treatment of slurried ferric oxide was injected into the top 2 cm of sediment at a density of 0.50 moles/m2 (28 grams as Fe+3), and the chambers allowed to run for 6 months. Overlying water and sediment were periodically sampled to track the evolution of the chambers.

After 6 months, the mass of methylmercury in the top 2 cm of iron-treated sediment decreased by an average of 40% compared to controls, while the mass increased in the 3-5 cm interval by an average of 20%. The center of mercury methylation deepened from 1 to 4 cm, which decreased the availability of sulfate to SRB and increased the upward diffusion distance of methylmercury to the overlying water by a factor of four. The concentration of methylmercury in the top 3 cm of porewater of iron-treated sediment was below detection (<20 pg/L), while the untreated anoxic control averaged 350 pg/L. The concentration of methylmercury in overlying water of both oxic and anoxic iron-treated sediments was below 20 pg/L, while the anoxic control was 270 pg/L. Upward transport and release of methylmercury to the overlying water was limited by sorption onto the iron oxide, which increased the bulk partition coefficient from 3,000 in the anoxic control to 30,000 in the anoxic iron-treated sediment. The results show that iron oxide treatment can be effective in suppressing mercury methylation and release even in systems with anoxic water columns.

Mercury (Hg) is a global pollutant and potent neurotoxin that bioaccumulates in aquatic and terrestrial foodwebs as monomethylmercury (MMHg). Microbial activity is the main driver of MMHg production, with sulfate reducing bacteria being important contributors. As such, predicting MMHgs fate in the environment is important for addressing ecosystem and human health concerns. The roasting of arsenopyrite at Giant Mine in Yellowknife, Canada, has created strong environmental gradients of sulfate in lakes in the surrounding area with distance from the mine. Whereas total Hg levels remain constant with increasing distance from the mine, the ratio of MMHg relative to total Hg increases with proximity to the stack. We hypothesized that the sulfate gradient is responsible for the pattern of MMHg concentrations around Giant Mine. As such, a factorial sampling design was developed to resolve whether sulfate or other environmental variables were affecting the production of MMHg in the lakes surrounding Giant Mine. To test our hypothesis, we sampled water and sediments from various lakes around Giant Mine. Using stable isotope analysis we determined simultaneous methylation and demethylation potentials, and defined the environmental variables affecting these rates of mercury cycling. Furthermore, we characterized the sediment microbial community structure using high throughput sequencing of 16S rRNA genes and physical chemical variables of the lakes sampled.

We are using an integrative geochemical and molecular approach to characterize microbial communities associated with methylmercury (MeHg) production in freshwater sediments impacted by sulfate from mine water discharges on Minnesota's Iron Range. Prior research in the region has shown that MeHg does not accumulate significantly more rapidly in systems heavily impacted by sulfate, but the microbial processes responsible methylation in sulfate-impacted and sulfate-unimpacted ecosystems are presently unknown. Both methylation rates (enriched stable isotopes) and culture-independent techniques (hgcA gene and transcript sequencing, full-cycle rRNA methods, and metagenomics) were applied to the water column and sediment of two sulfate-impacted lakes in Northern Minnesota, and to sulfate-amended wetland sediment mesocosms.

Methylation rates and MeHg accumulation were not clearly related to sulfate concentration or sulfide accumulation in either the lakes or the sulfate-amended mesocosm sediment. In the lakes, hgcA gene cloning shows that while the potential methylating communities differ among the sites, more than 50% of the hgcA gene sequences at all locations are affiliated with the Geobacteraceae (Deltaproteobacteria), Methanomicrobia (Euryarchaeota), or unknown populations. hgcA transcripts affiliated with methanogenic Archaea, iron-reducing Deltaproteobacteria, and unknown populations indicate that these organisms in addition to sulfate reducers may be important contributors to methylmercury production in the least sulfate-impacted sediments studied. Metagenomic analysis, in progress, will be used to determine the taxonomy and metabolic potential of the unknown hgcA-expressing populations in the lakes.

In the sulfate-amended mesocosm sediment, methylmercury production rates vary over two orders of magnitude across the experiment but do not show a clear relationship with sulfate amendment level. Both MeHg and inorganic Hg increased in the overlying water of mesocosms with increasing sulfate amendment, though this increase was not related to an increase in MeHg or methylation potential in the sediment, and may be related to the influence of sulfide accumulation on Hg partitioning. Microbial diversity decreases with elevated sulfate concentration, but cell counts indicate approximately similar cell densities across all sulfate amendments. Based on 16S rRNA gene amplicon sequencing, certain groups of sulfate-reducing Deltaproteobacteria increase with increasing sulfate concentration, other sulfate-reducers show no change, and iron-reducers (Geobacteraceae) and methanogens (Methanomicrobia) decrease. We will also report hgcAB and dsrB amplicon libraries from across the experiment that will reveal relationships between the methylating communities and mercury biogeochemistry. These collective results expand our knowledge of the clades and environmental influences important for understanding mercury methylation in sulfate-impacted freshwater ecosystems.

Mercury (Hg) methylation is as a key area of research necessary to understand spatial and temporal variability of toxic methylmercury (MeHg) on the landscape. Numerous factors affect MeHg production, the most important being those that affect inorganic Hg(II) bioavailability (e.g., Hg(II) concentration and ligand composition), and those that affect microbial community composition and activity. The principal goal of this project is to decipher the relative importance of aqueous biogeochemistry versus microbial community on MeHg production in Alaskan peatlands exhibiting a range of trophic status, including those lacking electron acceptors that support the traditional respiratory pathway of MeHg production (e.g., sulfate reduction). MeHg production is carried out by different groups of microorganisms that possess the hgcAB gene cluster, including the well-studied sulfate- and iron-reducing bacteria (SRB and FeRB). However, less well known bacteria also possess the hgcAB genes, including: syntrophs, methanogens, acetogens, and fermenters. Field sites for this study (ombrotrophic bogs to mineral-rich fens) were purposely chosen to encounter this full range microbial community assemblage.

Our experimental approach employed intact peat cores, site porewater, and enriched isotopes of inorganic Hg(II) and MeHg tracers (198Hg and Me204Hg). The tracers were added to anoxic porewater from each site and allowed to equilibrate for four hours. Next, amended porewater solutions were injected into peat cores, held at room temperature for 24 hours and then frozen. In addition, several other well-known and characterized dissolved organic matter (DOM) solutions (two from the Florida Everglades, Williams Lake in Minnesota, and cysteine) that have been shown to increase bioavailability of inorganic Hg for methylation in pure culture, as well as a no-DOM control, were also equilibrated with the enriched Hg isotope solution and injected into peat cores. Our results show that ambient MeHg concentrations in peat and porewater are significantly greater (15-30x) in fens compared to bogs. Likewise, methylation rates using site-specific porewater were also greater at fen sites. Interestingly, however, when bog porewater containing the 198Hg tracer was injected into fen peat cores, a 3-5x increase in methylation rate was observed when compared to replicate cores injected with 198Hg amended fen porewater. Our overall results suggest that the intact microbial community is of paramount importance for controlling MeHg differences among wetland types. Paradoxically, bog water DOM serves to enhance the bioavailability of Hg(II) for methylation better than fen water, however, bogs are consistently the lowest MeHg containing sites.

Wetlands, which are anoxic and rich in organic matter, have been recognized as globally important sources of methylmercury (MeHg). MeHg can then be transferred to their hydrologically coupled lakes and streams, resulting in risk of human exposure to this potent neurotoxin. The formation of MeHg is mediated by different groups of anaerobic microorganisms. The main groups are believed to be sulfate-reducing bacteria (SRB), iron-reducing bacteria (FeRB) and methanogens. However, the specific roles of these different groups with regards to methylation in wetlands are incomplete and there is a paucity of data on distribution patterns, diversity and interactions. In this study, we explore biogeochemical processes controlling MeHg formation in a chronosequence of boreal wetlands, with a focus on the interaction between electron acceptors (e.g. sulfate), and electron donors in the form of labile carbon provided by vascular plants. The chronosequence was formed by post-glacial land uplift along the Bothnian Bay in northern Scandinavia. This has created a unique natural age gradient that spans around 4000 years within < 10 km, with large variations in catchment hydrogeochemistry and nutrient availability that alter the presence of electron acceptors under the same climatic conditions. For this purpose, we sampled 5 wetlands in each of three age groups. For each wetland, we subjected 5 sampling sites with samples collected at two soil depths to DNA- and RNA-based analyses to determine the combined microbial community composition, potential methylators and active methylators. To experimentally influence the electron donors, we trimmed vascular plants to see how a reduction in root exudates would impact the formation of MeHg and corresponding changes in microbial community composition. The older wetlands had the highest total mercury (THg) concentrations but also the lowest net MeHg production and ratio of (MeHg/THg), which is an indicator of net mercury methylation. The microbial analyses support the findings that the amount of Hg is not the main factor controlling net MeHg production, but rather the interplay between microbes and the availability of electron donors and acceptors.

Acknowledgements: This work was supported by the Sino-Swedish Mercury Management Research Framework (SMaReF: VR2013-6978) and China Scholarship council.