Using tissue cysteine to predict the trophic transfer of methylmercury and selenium in lake food webs.

The biomagnification of toxic methylmercury (MeHg) and selenium (Se) through aquatic food webs using nitrogen stable isotopes (δ15N) varies among ecosystems but underlying mechanisms are yet unexplained. Given the strong links between MeHg and thiol-containing amino acids and proteins containing selenocysteine, our hypothesis was that cysteine content is a better predictor of MeHg and Se transfer through lake food webs than δ15N. Food web samples were collected from six lakes in Kejimkujik National Park, Nova Scotia, Canada, and the regression slopes of log MeHg or Se versus protein-bound cysteine or bulk δ15N were compared. Across all six lakes, MeHg varied by a factor of 10 among taxa and was significantly and positively related to both cysteine (R2 = 0.65-0.80, p < 0.001) and δ15N (R2 = 0.88-0.94, p < 0.001), with no among-system differences in these slopes. In contrast, total Se concentrations varied by less than a factor of 2 among taxa in four lakes and were significantly related to cysteine in only two food webs (R2 = 0.20 & 0.37, p = 0.014 & < 0.001); however, δ15N was not a predictor of Se in any lake (p = 0.052-0.777). Overall, these novel results indicate that cysteine content predicts MeHg, and sometimes Se, across trophic levels, providing a potential mechanism for among-system differences in their biomagnification.

Lifetime Chronicles of Selenium Exposure Linked to Deformities in an Imperiled Migratory Fish.

Aquatic ecosystems worldwide face growing threats from elevated levels of contaminants from human activities. Toxic levels of selenium (Se) shown to cause deformities in birds, fish, and mammals can transfer from parents to progeny during embryonic development or accumulate through Se-enriched diets. For migratory species that move across landscapes, tracking exposure to elevated Se is vital to mitigating vulnerabilities. Yet, traditional toxicological investigations resolve only recent Se exposure. Here, we use a novel combination of X-ray fluorescence microscopy and depositional chronology in a biomineral to reveal for the first time provenance, life stage, and duration of toxic Se exposure over the lifetime of an organism. Spinal deformities observed in wild Sacramento Splittail (Pogonichthys macrolepidotus), an imperiled migratory minnow, were attributed to elevated Se acquired through maternal transfer and juvenile feeding on contaminated prey. This novel approach paves the way for diagnosing sources, pathways, and potential for a cumulative exposure of Se relevant for conservation.

Resolving selenium exposure risk: Spatial, temporal, and tissue-specific variability of an endemic fish in a large, dynamic estuary.

Estuaries provide critical habitat for a vast array of fish and wildlife but are also a nexus for core economic activities that mobilize and concentrate contaminants that can threaten aquatic species. Selenium (Se), an essential element and potent reproductive toxin, is enriched in parts of the San Francisco Estuary (SFE) to levels known to cause toxicity, yet the risk of Se to species that inhabit the SFE is not well understood. We quantified Se concentrations in muscle, liver and ovary of the demersal cyprinid Sacramento Splittail from six regions in the SFE at three time points to evaluate Se exposure risk. Selenium levels exceeded proposed EPA criteria in ovary and thresholds of concern for liver in 15% and 20%, respectively, of fish collected in the fall of 2010, preceding the discovery of juvenile Splittail displaying a high incidence (>40%) of spinal deformities characteristic of Se toxicity, and again in 2011. No exceedances were detected in muscle tissue. Selenium concentrations varied significantly among regions for muscle (F5,113 = 20.49, p < 0.0001), liver (F5,113 = 28.4, p < 0.0001) and ovary (F5,112 = 19.3, p < 0.0001) but did not vary between the wet and dry years, nor were they influenced by foraging trophic level or prey selection. Foraging location along the salinity gradient, defined by δ34S values, explained regional Se exposures in Splittail. Relationships between tissues varied among regions for muscle and liver and muscle and ovary, but a single global relationship could be defined for ovary and liver Se concentrations. Our results suggest that the proposed EPA Se criteria for muscle tissue in Splittail may be under-protective as it would not have predicted exceedances in liver or ovary tissue and that the relationship between muscle tissue and ovary and liver may be Se concentration and seasonal dependent.

Nutrients mediate the effects of temperature on methylmercury concentrations in freshwater zooplankton.

Methylmercury (MeHg) bioaccumulation in freshwater aquatic systems is impacted by anthropogenic stressors, including climate change and nutrient enrichment. The goal of this study was to determine how warmer water temperatures and excess nutrients would alter zooplankton communities and phytoplankton concentrations, and whether those changes would in turn increase or decrease MeHg concentrations in freshwater zooplankton. To test this, we employed a 2 × 2 factorial experimental design with nutrient and temperature treatments. Mesocosms were filled with ambient water and plankton from Cottage Grove Reservoir, Oregon, U.S.A., a waterbody that has experienced decades of elevated MeHg concentrations and corresponding fish consumption advisories due to run-off from Black Butte Mine tailings, located within the watershed. Treatment combinations of warmer temperature (increased by 0.7 °C), nutrient addition (a single pulse of 10× ambient concentrations of nitrogen and phosphorous), control, and a combination of temperature and nutrients were applied to mesocosms. The individual treatments altered phytoplankton densities and community structure, but alone the effects on MeHg concentrations were muted. Importantly, we found a significant interactive effect of nutrients and temperature: the nutrient addition appeared to buffer against increased MeHg concentrations associated with elevated temperature. However, there was variability in this response, which seems to be related to the abundance of Daphnia and edible phytoplankton. Nutrients at low temperature were associated with marginal increases (1.1×) in zooplankton MeHg. Our findings suggest that global change drivers that influence community composition and ecosystem energetics of both zooplankton and phytoplankton can alter MeHg pathways through food webs.

Food web pathway determines how selenium affects aquatic ecosystems: a San Francisco Bay case study.

Chemical contaminants disrupt ecosystems, but specific effects may be under-appreciated when poorly known processes such as uptake mechanisms, uptake via diet, food preferences, and food web dynamics are influential. Here we show that a combination of food web structure and the physiology of trace element accumulation explain why some species in San Francisco Bay are threatened by a relatively low level of selenium contamination and some are not. Bivalves and crustacean zooplankton form the base of two dominant food webs in estuaries. The dominant bivalve Potamocorbula amurensis has a 10-fold slower rate constant of loss for selenium than do common crustaceans such as copepods and the mysid Neomysis mercedis (rate constant of loss, ke = 0.025, 0.155, and 0.25 d(-1), respectively). The result is much higher selenium concentrations in the bivalve than in the crustaceans. Stable isotope analyses show that this difference is propagated up the respective food webs in San Francisco Bay. Several predators of bivalves have tissue concentrations of selenium that exceed thresholds thought to be associated with teratogenesis or reproductive failure (liver Se >15 microg g(-1) dry weight). Deformities typical of selenium-induced teratogenesis were observed in one of these species. Concentrations of selenium in tissues of predators of zooplankton are less than the thresholds. Basic physiological and ecological processes can drive wide differences in exposure and effects among species, but such processes are rarely considered in traditional evaluations of contaminant impacts.