G15 (II) Mercury in Fish

Mercury (Hg) is a widespread environmental contaminant known for the neurotoxicity of its methylated form, which bioaccumulates and biomagnifies in aquatic food webs. Recent studies have shown elevated Hg concentrations in fish from some remote areas, even when background levels are low and point source inputs are not apparent. Additionally, the bioaccumulation and biomagnification of Hg is known to vary among species utilizing different food webs within and between systems. We used stable isotopes of carbon and nitrogen to determine if trophic based differences in Hg levels between species are also found within a trophically-divergent species, using benthic and limnetic morphotypes of threespine stickleback fish (Gasterosteus aculeatus) from Benka Lake Alaska. The average total Hg concentration in the limnetic morphotype was nearly double that of the benthic morphotype and total Hg concentration was strongly correlated with the proportion of benthic carbon in the diet of individual fish. Limnetic fish had a higher average trophic position than benthic fish and calculated biomagnification factors were also greater in the limnetic food web. In addition, significant sex effects were observed with female fish of each morphotype having lower Hg concentrations than their male counterparts, and Hg was present in eggs of both benthic and limnetic females. These results indicate that both trophic ecology and sex are important determinants of Hg contamination even within a single species and lake. Since stickleback are an ecologically important consumer and a primary prey item for many species of birds and sportfish, the observed differences between benthic and limnetic morphotypes likely have important implications for Hg contamination at higher trophic levels.

The Aleutian Archipelago, Alaska, is an isolated arc of over 300 volcanic islands stretching 1600 km across the Bering Sea. Although remote, some Aleutian Islands were heavily impacted by military activities from World War II until the present and were exposed to high levels of contaminants used by the military, including mercury. Atmospheric and oceanic transport of mercury in the Aleutian Islands is also a concern due to global atmospheric deposition and prevailing ocean currents. Recent research has documented high levels of mercury in marine fishes, seabirds, and marine mammals throughout the Aleutian Archipelago, but mercury contamination of freshwater ecosystems is poorly understood. Total mercury concentrations were measured in threespine stickleback (Gasterosteus aculeatus) fish collected from eight freshwater lakes at Adak Island. Mean total mercury concentrations for whole-body homogenates for all lakes ranged from 0.314 to 0.560 mg/kg dry weight and differed significantly among lakes. Stickleback collected from seabird associated lakes had significantly higher concentrations of total mercury compared to non-seabird lakes and lakes with formerly used defense sites. The d¹³C and d¹⁵N stable isotope ratios of stickleback collected from seabird lakes suggest a marine input of nutrients and contaminants. These findings indicate that seabirds are transporting marine derived mercury to freshwater lakes via guano and their tissues, and that the marine environment is a more important contributor of mercury contamination than the local military sites.

Total mercury and nitrogen stable isotopes (d15N), as an indicator of trophic level, were measured in 320 tissue samples from 57 species of invertebrate (17 species) and finfish (40 species) collected from the coastal waters off Southwest Florida. Additionally, methylmercury was determined in representative samples of invertebrates. Average Hg concentrations for species ranged from 0.004 mg/kg in a sample from an unidentified species of brittlestar (Ophiuroidea) to 2.839 ± 1.39 mg/kg in king mackerel (Scomberomorus cavalla). In general, fish were more enriched in d15N (‰) than invertebrates; exceptions to this general pattern included an unidentified species of isopod (Isopoda), the calico crab (Hepatus epheliticus) and stone crab (Menippe spp.) that had higher d15N than several species of small fish. The log 10 transformed average Hg concentration in the various species was positively correlated with d15N (p<0.001, r² = 0.66) and had a slope of 0.21. When the d15N of the calico scallop (Argopecten gibbus), as a primary consumer, was used to convert d15N to trophic level (TL), they ranged from 1.5 for conch (Strombus alatus) to 4.15 for both spanish mackerel (Scomberomorus maculatus) and crevalle jack (Caranx hippos). A food web magnification factor (FWMF) of 5.05 was calculated from the relationship between log 10 transformed average Hg concentration and TL. This FWMF for mercury transfer through this sub-tropical marine system will be compared to published values from other areas and habitats.

As part of a larger project examining mercury transfer through a sub-tropical marine food web in the coastal waters off Southwest Florida, total mercury (THg) concentrations were determined in biopsy samples collected while sharks were being tagged with satellite or conventional tags. Additionally, stable isotopes of carbon and nitrogen were measured in fin clips as indicators of carbon source (d13C) and trophic position (d15N). As of this writing, a total of 51 samples were collected from five species of sharks in this on-going project: blacknose (Carcharhinus acronotus), blacktip (C. limbatus), bull (C. leucas), great hammerheads (Sphyrna mokarran), and tiger (Galeocerdo cuvier). Thus far, 39 of the biopsy samples have been analyzed for Hg using a Nippon MA2000 direct Hg analyzer and 26fin clips have been analyzed for carbon and nitrogen isotopes by UC Davis Stable Isotope Facility (SIF). THg concentration ranged from 0.188 mg/kg in a tiger shark to 4.52 mg/kg in a blacktip. With a mean concentration of 3.06 ± 0.66 mg/kg, blacktip sharks on average contained significantly higher levels of Hg than all other species (p<0.001; Holm-Sidak post-hoc test p<0.05). The blacktip and blacknose sharks showed an obvious increase in THg levels with increasing size; however, they contained higher concentrations than the much larger hammerhead, bull and tiger sharks. Further, d15N suggested that most of the tigers at one time had fed at a slightly higher trophic position than the backtip and blacknose. While among-species differences in d15N were statistically significant (p = 0.014), between-species comparisons were not (Dunn’s Methods p>0.05). When shark data were overlaid on top of similar data from invertebrates and bony fish caught off SW Florida, bull and tiger sharks fit nicely into the THg-d15N relationship. Yet several smaller, bony fish appeared to be feeding at a higher trophic position as suggested by the higher d15N and higher THg (e.g., king mackerel, Scomberomorus cavalla). Equally interesting, the blacktip, blacknose and hammerhead did not fit as nicely into the THg-d15N relationship, having higher Hg levels than predicted based on their d15N. These data will be discussed in the context of ontogenetic dietary shifts, possible differences in isotope turnover versus Hg clearance rates in metabolically distinct tissues, and the on-going debate regarding comparisons of diet-tissue discrimination factors between sharks and bony fish.

Eighty four caudal peduncle muscle samples of bigeye tuna, Thunnus obesus, collected onboard of Taiwanese fishing vessels at Pacific Oceans from Aug. 2005 to Jul. 2007, and were used to analyze total Hg (THg) and organic Hg (OHg) concentrations. The THg and OHg ranged from 0.158 to 3.324 (0.935 ± 0.655) and 0.111 to 1.965 (0.544 ± 0.396) mg kg^-1 flesh wt. that were similar to previous studies of Atlantic and Pacific Oceans in recent years, but showing five-times elevated levels of the tuna from the Ocean in recent three decades. The relationships between the log THg and log OHg concentrations and log fork length (cm) and log body weight (kg) were well established and revealed a significantly elevated slop than that of Atlantic Ocean, of which indicating the blooming of industrial development around the Pacific region with high Hg emission. In comparison with the OHg vs THg from three Oceans found that the demethylation of the muscle commence at the THg = 0.65. Six of 84 muscle samples had levels of OHg above 1 mg kg^-1 wet weight set by US FDA and WHO for predatory fish.

The fish of the cod family (Gadidae) include several species that are important sources of seafood, such as Atlantic cod (Gadhus morhua) and haddock (Melanogrammus aeglefinus). Tusk (Brosme brosme) is a less known, relatively large, gadoid fish species, also popular as seafood. It appears on both sides of the North Atlantic at depths between 150 and 800 m,. In Norway the tusk is found in the open oceans, in the Norwegian coastal current (NCC) as well as in the deep fjords.

While the Atlantic cod is shown to have low concentrations of mercury in fillet except at local hot spot sites, there is a lack of knowledge on mercury accumulation in the tusk. To gain more data on mercury levels in this species, tusk were sampled at seven localitites in open oceans, six in the NCC and two in a fjord system with some known mercury pollutant problems. Length and weight were determined for individual fish before fillet samples were taken of all fish and liver samples were taken of fish from coastal stations. The fillet samples were homogenised and freeze-dried, and liver samples were homogenised, prior to analyses for metals. For the determination of mercury subsamples of approximately 0.2 g dry weight were used for microwave-assisted wet digestion. The mercury determinations were carried out on an Agilent quadrupole ICPMS.

The tusk weighed between 0.32 and 10 kg, and mercury concentration correlated positively with fish size. There was a clear pattern of lower mercury concentration in fillet of tusk from the open oceans than in those from the NCC. Also, mercury levels were higher in fillets of fish captured closer to the population dense southern areas than in the less populated northern areas. This was the case both for the ocean and the NCC localities. The mean fillet concentration of mercury in the southernmost NCC locality was 0.49 mg/kg wet weight, which is approximately at the EU and Norway’s upper limit for sale of 0.5 mg/kg wet weight, while in the northernmost locality close to Svalbard the mean concentration was as low as 0.08 mg/kg wet weight. In the more polluted fjord areas the mercury levels were about three times the legal limit. The results show that this gadoid species can potentially accumulate substantial amounts of mercury, and closer monitoring has to be in place to secure the public health against possible mercury contamination.

Mercury is found in both organic and inorganic forms in the atmosphere, sediment and water bodies due to natural and anthropogenic sources. The WHO estimates that humans ingest 2.4ug organic mercury and 4.3ug inorganic mercury daily. Limited information is available on the uptake and accumulation of mercury in developing organisms, particularly as it relates to chemical form of the metal. To address this problem we utilized synchrotron X-ray fluorescence imaging in zebrafish, an increasingly well studied model vertebrate for investigating mechanisms of chemical toxicity. We were interested in mapping the localization of inorganic mercury in zebrafish in conjunction with a previous study by our group on the localization of organic mercury in zebrafish (Korbas et al. (2008)PNAS 105: 12108-12112).

Zebrafish larvae were exposed to one of four forms of mercury (methyl mercury chloride, methyl mercury L-cysteine, mercuric chloride, and mercury bis-L-cysteineate). Adjacent serial sections were utilized for synchrotron imaging and histological staining, respectively.

Significant variations in mercury accumulation were found in fish exposed to organic mercury compared to fish exposed to inorganic mercury. Larvae exposed to inorganic mercury exhibited accumulation in the kidneys and olfactory epithelium. However, only the fish exposed to organic mercury exhibited preferential accumulation in the lens epithelium, gut tube, and skeletal muscles. These variations demonstrate the importance of considering chemical form when considering the uptake and accumulation of mercury.

Ongoing research investigates the efficacy of chelating agents. Although dimercaptosuccinic acid (DMSA) and alpha lipoic acid (ALA) are not true chelators, both are currently used to treat mercury exposure. As DMSA and ALA are poorly optimized for this role they are not always effective and often cause a number of adverse effects.

A preliminary test involved exposing fish to methyl mercury L-cysteine or mercuric chloride followed by a short treatment with either ALA or DMSA. Our results indicate that neither ALA nor DMSA are proper cheltors. Eventually, we hope to test a custom chelator that can bind and excrete different forms of mercury more effectively.

Mercury (Hg) is a toxic metal that is found in aquatic food webs and is hazardous to humans. An emerging conceptual model predicts that the areas of the landscape that have the potential to contain food webs with elevated concentrations of Hg are those that receive high amounts of Hg and sulfate deposition and have high coverage of forests and wetlands and low coverage of agriculture. The objective of our study was to test this conceptual model using concentrations of Hg in largemouth bass (Micropterus salmoides) from 145 reservoirs in four ecoregions of North Texas. The highest level of Hg contamination in fish was in the South Central Plains, the ecoregion that receives the highest levels of Hg and sulfate deposition and contains extensive forest and wetland habitat and little agriculture. This study has important implications for other areas of the US because the South Central Plains extends into parts of Oklahoma, Louisiana and Arkansas, covering a total area of 152,132 km² of the southern US.

Lake St. Francis is a fluvial lake (254.2 km²) in the St. Lawrence River with shores in Quebec and Ontario, Canada and New York, USA. Here the consumption of large fish has been restricted for decades as a result of legacy contamination present in the sediments from industrial discharge and non-point sources (tributaries). The influence of tributaries on mercury loading and cycling remains unclear. In this study, we explored the relationships between total and methyl mercury to the water chemistry to identify pathways of fish mercury contamination. Chromophoric dissolved organic material (CDOM) was measured continuously using a Turner Designs 10-AU fluorometer along the 2m contour. Values for CDOM were compared with direct measurements of dissolved organic carbon (DOC) and values were more than 10 times higher in tributaries with much of the colour from altered wetlands in the headwater region. Along both the north and south shores, values for total and methyl mercury were correlated to CDOM and DOC and were much higher in the tributary water. Measurements were made in May, August and October 2009. Chlorophyll a was measured by filtration in ecologically significant size classes of pico (0.2 to 2 µm), nano (2-20 µm), and micro (>20 µm). Specific conductivity was a useful tracer for streams with high conductivity along the northern shore and low conductivity for streams from the Adirondack Mountains, USA flowing into the south shore. Our objective was to determine the relative importance of the high load from the St. Lawrence River(129,000 g THg y^-1 and 7000 g MeHg y^-1) resulting from the high volume (annual discharge 6,900 m³ s^-1 yet low concentrations (ca 500 pg L^-1 THg and 50 pg L^-1 MeHg) along with low concentrations of DOC and CDOM. These values compare with the loading from north shore tributaries of about 280 g THg y^-1 and 50 g MeHg y^-1 and loadings from south shore tributaries 10 times higher. Although about 4 times higher concentrations of total and methyl mercury, loadings from the tributaries are much lower than in the St. Lawrence River. However, tributary mouths as preferred fish habitat suggest a larger influence on fish concentrations than mass balance calculations would suggest. In addition, the loading of the St Lawrence is channelled through the dredged ship channel and is restricted from circulation with either the north or south parts of Lake St. Francis.

In aquaculture, the inclusion of marine protein (fishmeal) in fish feed is the dominant source of mercury (Hg), mainly methylmercury (MeHg), in farmed fish. In Europe, the levels of contaminants in feed and seafood are regulated through the European feed and food legislation, which set statutory limits for a wide range of contaminants in feed ingredients, feed and food, including seafood. The current maximum level for Hg in fish feed is 0.2 mg/kg feed (Commission Directive 2010/6/EC), and for fish the maximum level is 0.5 mg/kg wet weight (Commission Regulation (EC) no 1881/2006) for most species while it is 1.0 mg/kg wet weight (Commission regulation (EC) no 629/2008) for a limited list of species including Atlantic halibut, sword fish and tuna.. The European Food Safety Authority (EFSA) states in their scientific opinion of the Scientific Panel on Contaminants in the Food Chain “Mercury as undesirable substance in animal feed” (2008) that the current maximum level for mercury in fish feed is sufficient to ensure a mercury level in farmed salmon fillet that does not pose a risk to consumers. They also state that the validity of the maximum level needs to be ascertained for other species of farmed fish.

In a series of trials we have investigated the assimilation and depuration of dietary MeHg in farmed species (Atlantic cod and Atlantic salmon), as well as in a model organism (zebrafish). In general, a continuous accumulation of mercury in muscle was seen during the exposure periods, and the elimination of mercury from muscle was slow and incomplete within applied depuration periods. Atlantic cod were fed a MeHg enriched diet (1 mg Hg/kg feed) for three months followed by a three months depuration period. The carry-over of MeHg from feed to muscle, described by the estimated absorption efficiency, was 38%. The final mercury concentration in muscle was 0.38 ± 0.04 mg Hg/kg ww (n = 6), where more than 90% of the mercury was in the form of MeHg, and more than 99% was found in the protein fraction. The results from these trials show that fish readily take up MeHg from feed and that the MeHg is efficiently accumulated in the muscle, where it is incorporated into larger peptides and/or proteins. An overview of these studies will be given, and the implications for seafood safety will be discussed.