G8 (I) Mercury bioaccumulation and trophic transfer

Thiols such as cysteine and glutathione, are found in water columns of oceans and estuaries where they form charged complexes with inorganic mercury (Hg^II) and methylmercury (CH₃Hg^II). In this study, we used a mer-lux bioreporter to decipher the mechanisms of Hg^II and CH₃Hg^II uptake into Escherichia coli cells when these ions are bound to glutathione and cysteine. We investigated the impact of three amino acids (cysteine, methionine and leucine) on the uptake of Hg^II and CH₃Hg^II. We also compared uptake rates of Hg^II and CH₃Hg^II into an E. coli strain that has a known glutathione transporter to an E. coli strain where this transporter system had been disabled. This experiment was conducted in both a minimal and enriched medium. We found that complexation with cysteine resulted in increased uptake of both mercury species while complexation with methionine and leucine had little or no effect on uptake. The mechanism of uptake of the mercurial-cysteine complexes is likely not passive diffusion but could result from the activities of a cysteine transport system. In enriched medium, there was increased uptake of Hg^II and CH₃Hg^II with increasing glutathione concentration in the strain with the functioning transport system but not in the strain where this capability had been disabled. In a minimum medium, uptake rates decreased with increasing glutathione concentration. Overall, our studies demonstrate that uptake of Hg^II and CH₃Hg^II bound to thiols is due to active transport mechanisms in E. coli if the appropriate genes are activated.

Wetlands are particularly important areas of methylmercury (MeHg) production, partly because they support sulphate-reducing bacteria, microbial mediators of methylation. Not all wetlands have equal MeHg production, and there are large differences among wetlands’ spatial and hydrological connections to aquatic systems supporting fish. Peatlands are not fish habitat, but often have hydrological linkages to nearby lakes. Although considerations of risk from peatland-derived MeHg are usually focused on lakes, little research has been conducted on risks to peatland invertebrates, which may have critical ecological linkages to small mammals, amphibians, and birds. A key question is whether soil and pore water MeHg concentrations in peatlands drive the load of mercury observed in peatland invertebrates.

This study was part of a large-scale peatland manipulation experiment at the Marcell Experimental Forest in Minnesota. Annual wet sulphate deposition was increased four-fold above background in one half of the peatland through a piped sprinkler system for four years. In 2006, this experimental half was subdivided into portions where sulphate deposition was either continued until the end of 2008 (experimental) or ceased in 2006 (recovery). Peatland invertebrates were sampled in spring 2009 by hand picking and netting.

The most abundant and ubiquitous invertebrate genera across sub-habitats and treatments were Culex (mosquito), Limnephilus (caddisfly), and Dytiscus (predaceous diving beetle) larvae. Pore water and solid-phase MeHg concentrations followed the pattern experimental > recovery > control, with some spatial dependencies. Mercury concentrations (means in ng/g dry weight in experimental, recovery, and control, respectively) in mosquito (149, 92, 79), caddisfly (105, 50, 59), and diving beetle (154, 61, 53) larvae followed nearly the exact same patterns. Methylmercury analysis is ongoing to confirm the percentages of total Hg as MeHg in the different taxa. The study demonstrates that peatland invertebrate loads of mercury reflect gradients in MeHg accumulation in water and peat. Also, soil, water, and biotic concentrations of MeHg decreased significantly after only two years of lessened sulphate deposition, suggesting that further controls on sulphur emissions to the environment could have significant and rapid effects on lowering mercury burdens in peatland biota.

The bioaccumulation of methylmercury in freshwater fish is a wide-spread problem, even in lightly contaminated, seemingly pristine areas, such as national parks. The United States National Park Service manages more than 18,200 km² of aquatic resources. Sampling to monitor mercury in fish across these ecosystems requires specialized gear, personnel with technical expertise, and the ability to execute sampling protocols in remote areas. We and others have observed correlations between methylmercury concentrations in larval dragonflies and total mercury in axial muscle of game fish. Sampling larval dragonflies as sentinel organisms for methylmercury contamination requires substantially less effort than sampling fish. Larval dragonflies are commonly encountered among the national park units in the western Great Lakes region, yet we know little about their taxonomic, behavioral, or spatial variability--all of which are critical in evaluating their efficacy as biosentinels for monitoring mercury in food webs. We sampled pre-emergent dragonflies in late spring 2008 and 2009 from selected aquatic ecosystems in six parks of the western Great Lakes region, including Indiana Dunes National Lakeshore, Sleeping Bear Dunes National Lakeshore, Pictured Rocks National Lakeshore, Isle Royale National Park, Voyageurs National Park, and Grand Portage National Monument. Larval dragonflies were sampled with hand-held nets in nearshore habitats, identified to species and sex, measured (length), lyophilized to a constant dry weight, and analyzed for total mercury (THg) and methylmercury (MeHg). We initially focused on larvae in the family Gomphidae, which burrow in loose substrates within a centimeter of the water-sediment interface--where microbial methylation rates tend to be great. We analyzed over 370 gomphid larvae, which, as a group, averaged (±1 SD) 78 ± 26 percent MeHg as a fraction of THg. Mean MeHg concentrations among gomphid species were correlated with mean THg concentrations in predatory fish (primarily northern pike, large yellow perch, and largemouth bass) across sampled lakes. Gomphus spicatus, a widespread species found in 13 of 25 water bodies, ranged in mean (±1 SD) MeHg concentration from 19 ± 11 to 121 ± 47 ng/g dry weight. Variation in mean MeHg concentration among sympatric gomphid species was generally low, which suggests that sentinel groups may be effectively monitored in aggregate, possibly at the genus or guild level.

Although mercury has been well studied in temperate and arctic aquatic systems that often are characterized by low primary productivity, relatively little is known about the behaviour of mercury in tropical eutrophic lakes. The distribution and trophic transfer of mercury was characterized in several Ugandan lakes including the East African great lakes Victoria, Albert and Edward as well as four smaller lakes. The large lakes sampled all support substantial commercially important fisheries, while the smaller Ugandan lakes support subsistence fisheries that provide a critically important source of protein and income for riparian communities. These lakes ranged in trophic status from mesotrophic to hypereutrophic. Consistent biomagnification of mercury was observed at all study sites; however, total mercury concentrations in fish were generally low, and would not be expected to pose a risk to even the most frequent fish consumers. Mercury dynamics were strongly linked to lake trophic status; with biomagnification rates and mercury concentrations in fish lower at the hypereutrophic study sites than at the mesotrophic and eutrophic study sites. Year-round high phytoplankton biomass and growth rates in highly productive tropical lakes may act to mitigate the potential for high mercury concentrations in fish, whereby growth and biomass dilution at the base of the food web can act to reduce mercury concentrations entering the food web, and growth dilution of mercury at consumer trophic levels can act to reduce the realized biomagnification rate of mercury. In these tropical lakes the highest mercury concentrations in fish were observed in the lakes with the lowest phytoplankton biomass (and often the lowest mercury concentrations in water). These results may demonstrate the critical role of lake trophic status in modifying bioaccumulation and biomagnifications rates in food webs.

Methylation and biomagnification of mercury are well documented in aquatic ecosystems of the Amazon. But their occurrence at the high altitudes of the arid and cold Altiplano is still poorly known. Recent researches on two Bolivian lakes have demonstrated their importance.

We compared the trophic structure and mercury transfer in aquatic communities of two bolivian lakes (Uru Uru and Poopo). Both lakes are located in a semiarid region and form part of the endoreic bolivian-peruvian Titicaca catchment located at 3800m. Lake Uru-Uru is directly connected to lake Titicaca by the Desaguadero River, it also has additional tributaries polluted by effluents of mining and urban origin. Lake Poopo located downstream of lake Uru Uru is only connected to lake Titicaca and lake Uru-Uru during the wet season and present a north-south salinity gradient.

In lake Uru-Uru, stable isotope (d¹³C and d¹⁵N) analysis of the different compartments allowed us to discriminate the different trophic sources involved. In general, the bottom sediment (organic matter) was the most important carbon source for the food webs, followed by an aquatic plant (Myriophyllum sp.) and the periphyton (biofilm). Macroinvertebrates were primary consumers, showing intermediate mercury concentrations (0,1 – 0,2 mg/Kg). The higher trophic positions were represented by fishes (Orestia spp.) and a grebe (Rollandia Rolland), with maximum mercury concentrations higher than 0,5 mg/Kg. In lake Poopo, we studied five sites according to the salinity gradient from north to south. The diversity of aquatic organisms was inversely correlated with salinity, showing a shorter and simpler food chain than in lake Uru-Uru. The bottom sediment was the unique source for the food webs and fish top predator (Odontesthes bonariensis) showed Hg concentrations higher than 0,6 mg/Kg.

In both lakes, Hg transfer and biomagnification along the trophic structure (from source to top predator) was observed. These results showed that mercury methylation processes were significant in these two lakes.

Mercury has been largely introduced in the environment by human activities and the complexity of its biogeochemical cycle makes difficult the complete understanding of its sources and transformations in a specific target organism or food chain. Hg stable isotopes were demonstrated to be useful to track its sources and transformations using both mass dependent fractionation (MDF) and “odd isotope anomaly” (or MIF). This study presents the comparison of speciation and isotopic signatures of Hg in different compartments of an aquatic ecosystem of Russia: the pristine Lake Baikal located upstream to the Bratsk-Water Reservoir (B-WR), contaminated by Hg from chlor-alkali plant.

Sediments, water, zooplankton, herbivorous and carnivorous fish muscle (roach, perch, ...), and endemic Baikal seal samples have been analysed for Hg speciation by GC-ICP-MS and Hg stable isotope (sediment, biota) by cold-vapor MC-ICP-MS. Determination of diet, mass, length, sex, age and carbon and nitrogen stable isotope analysis has allowed the characterisation of the biota samples.

Hg species concentration in sediments, water, plankton and fishes at both sites showed significant higher amounts (2 to 40 times) in B-WR than in Lake Baikal. In the contaminated reservoir (B-WR), Hg isotopic signature recorded in all trophic levels exhibits a close influence from the isotopic composition of the contaminated sediment which is characterised by typical direct anthropogenic pollution (i.e. no MIF anomaly). These results suggest that biota from contaminated sites may preserve the Hg isotopic signature of the anthropogenic source, especially when feeding in benthic environment. In the pristine Lake Baikal, Hg species concentration ratio (MeHg/IHg) depends both on the trophic level and the age of the specimen in the case of seals. Stable isotope Hg signature of Lake Baikal fish and seal tissues showed positive correlation between d²⁰²Hg (from -0.90 to 3.00 ‰) and trophic level. Furthermore, the anomaly of Hg odd isotope (MIF) observed in tissues of biota (d¹⁹⁹Hg from 0.20 to 6.25 ‰), suggests the bioaccumulation of MeHg originating from the pelagic environment, in which it has been exposed to MIF photochemical pathways.

The comparison of the results obtained for both contaminated and pristine sites suggests that Hg isotopic signature in aquatic organisms reveals both MeHg pathways in aquatic environments and trophic bioaccumulation routes.

Salt marshes are important areas for coastal fish production, and can fuel the transfer of Hg to open water marine food webs through the nekton trophic relay. Mercury methylation and bioaccumulation of MeHg in food webs can be influenced by both environmental conditions and food web structure. Gulf of Maine salt marsh ecosystems exhibit landscape-scale patterns of habitat heterogeneity. These marshes are an integration of geomorphologically complex surface water features within a biogenic landscape of sediment accumulating and peat forming vegetation zones. We present results regarding habitat-specific concentrations of Hg and MeHg, and their tissue concentrations in associated nekton (fish and shrimp) for channels, creeks and pools in a typical Gulf of Maine back-barrier salt marsh. Sediment Hg concentrations show a strong increasing gradient from channels to pools to creeks. Hg tissue concentrations were higher for shrimp (Crangon septemspinosa, Palaeomonetes pugio) in creeks than in pools, but ratios of MeHg to Hg tissue concentrations were higher in pools than creeks. Hg tissue concentrations for the highly abundant marsh minnow Fundulus heteroclitus were higher in pools than creeks. MeHg concentrations in Fundulus heteroclitus were also higher in pools than creeks, and declined with increasing body size, consistent with observed changes in diet from pelagic to benthic prey. Planktivorous fish feeding in channels (Menidia menidia) showed the highest tissue concentrations of Hg and MeHg, further indicating that MeHg bioaccumulation is driven by the pelagic food web. The ratio of MeHg to Hg tissue concentration in Menidia was similar to that for shrimp in pools, and much higher than that for shrimp in creeks, while ratios for Fundulus in creeks and pools were similar to or higher than that for Menidia. The highest MeHg:Hg ratio was found for small Fundulus from pools. Variations in tissue MeHg:Hg ratios may be explained by the influence of habitat-specific variation in sediment organic matter on Hg methylation. Our findings contribute to an understanding of spatial variation in trophic transfer of Hg and MeHg in salt marsh food webs. As we refine our understanding with continued study, our goal is to develop a spatially explicit model of whole-salt marsh nekton MeHg bioaccumulation and trophic relay to the greater Gulf of Maine food web.

Input of nutrients and allochthonous dissolved organic carbon (ADOC) control the pelagic food web structure and productivity in estuarine ecosystems, which in turn control the type and supply rate of organic material deposited to the benthic zone. The deposition of organic material is important for Hg biogeochemical processes as it affects e.g. redox conditions, activity of Hg methylating bacteria and Hg speciation in sediments. In this study estuarine model ecosystems, with different pelagic food web structure and productivity, were constructed to establish the relation between biogeochemical processes controlling net formation of methyl mercury (CH₃Hg) and the type and supply rate of organic material to the benthic zone. Furthermore the issues of how this relation affects the relative CH₃Hg formation from simulated fresh and aged Hg depositions, and how ecological changes in the pelagic food web (e.g. due to climate change or eutrophication scenarios) may alter CH₃Hg formation rate, were addressed.

Model ecosystems were constructed utilizing a mesocosm facility containing 12 isolated tubes of 70 cm id and 5 m height with intact sediment cores (65 cm diameter) and brackish water sampled from an estuary located in the Bothnian bay. Five different Hg isotope tracers were added to the systems: solid-phase tracers (black cinnabar (ß-HgS) as well as inorganic mercury (Hg^II) and CH₃Hg bound to organic matter) were injected 0.5 cm below the sediment surface and aqueous Hg^II and CH₃Hg tracers were added to the water column. By additions of nutrient and ADOC three different treatment regimens were constructed in triplicates; two treatments with phytoplankton dominated pelagic food webs but with different productivity and one with bacteria dominated pelagic food web and elevated concentration of ADOC. These treatments conceptually represent effects on the pelage from eutrophication and climate change scenarios, respectively, and resulted in substantial differences in deposition of OM to the sediments. Formation and degradation of CH₃Hg were monitored for the respective tracers during an incubation period of 8 weeks, and the systems were characterized with respect to important chemical and biological parameters, including solid and aqueous phase speciation of Hg, bacterial communities and organic material, to establish their importance for Hg reactivity.