No abstract

The bioconcentration factor (BCF) and bioaccumulation factor (BAF) are used as the criteria for bioaccumulation in the context of identifying and classifying substances that are hazardous to the aquatic environment. The BCF/BAF criteria, while developed as surrogates for chronic toxicity and/or biomagnification of anthropogenic organic substances, are applied to all substances including metals. This work examines the theoretical and experimental basis for the use of BCF/BAF in the hazard assessment of Zn, Cd, Cu, Pb, Ni, and Ag. As well, BCF/BAFs for Hg (methyl and inorganic forms) and hexachlorobenzene (HCB) were evaluated. The BCF/BAF data for Zn, Cd, Cu, Pb, Ni, and Ag were characterized by extreme variability in mean BCF/BAF values and a clear inverse relationship between BCF/BAF and aqueous exposure. The high variability persisted when even when data were limited to an exposure range where chronic toxicity would be expected. Mean BCF/BAF values for Hg were also variable, but the inverse relationship was equivocal, in contrast with HCB, which conformed to the BCF model. This study illustrates that the BCF/BAF criteria, as currently applied, are inappropriate for the hazard identification and classification of metals. Furthermore, using BCF and BAF data leads to conclusions that are inconsistent with the toxicological data, as values are highest (indicating hazard) at low exposure concentrations and are lowest (indicating no hazard) at high exposure concentrations, where impacts are likely. Bioconcentration and bioaccumulation factors do not distinguish between essential mineral nutrient, normal background metal bioaccumulation, the adaptive capabilities of animals to vary uptake and elimination within the spectrum of exposure regimes, nor the specific ability to sequester, detoxify, and store internalized metal from metal uptake that results in adverse effect. An alternative to BCF, the accumulation factor (ACF), for metals was assessed and, while providing an improvement, it did not provide a complete solution. A bioaccumulation criterion for the hazard identification of metals is required, and work directed at linking chronic toxicity and bioaccumulation may provide some solutions.

Some physiological aspects of five fish anaesthetics in rainbow trout were investigated. The effects of benzocaine, 2-phenoxyethanol, MS-222 (Sandoz), metomidate, and carbon dioxide gas (CO2) on acid–base regulation, hematocrit, blood gases, and cortisol and adrenaline concentrations were determined in resting rainbow trout fitted with chronic catheters in the dorsal aorta. A severe hypoxia developed with the cessation of breathing in deep anaesthesia. This was accompanied by a rise in blood [Formula: see text] and adrenaline concentration, and a fall in blood pH. Blood bicarbonate concentrations remained unchanged and cortisol concentrations declined with time. There was a transient increase in hematocrit coinciding with the increase in adrenaline concentrations.

No abstract

As part of a SETAC Pellston Workshop, we evaluated the potential use of metal tissue residues for predicting effects in aquatic organisms. This evaluation included consideration of different conceptual models and then development of several case studies on how tissue residues might be applied for metals, assessing the strengths and weaknesses of these different approaches. We further developed a new conceptual model in which metal tissue concentrations from metal-accumulating organisms (principally invertebrates) that are relatively insensitive to metal toxicity could be used as predictors of effects in metal-sensitive taxa that typically do not accumulate metals to a significant degree. Overall, we conclude that the use of tissue residue assessment for metals other than organometals has not led to the development of a generalized approach as in the case of organic substances. Species-specific and site-specific approaches have been developed for one or more metals (e.g., Ni). The use of gill tissue residues within the biotic ligand model is another successful application. Aquatic organisms contain a diverse array of homeostatic mechanisms that are both metal- and species-specific. As a result, use of whole-body measurements (and often specific organs) for metals does not lead to a defensible position regarding risk to the organism. Rather, we suggest that in the short term, with sufficient validation, species- and site-specific approaches for metals can be developed. In the longer term it may be possible to use metal-accumulating species to predict toxicity to metal-sensitive species with appropriate field validation.

No abstract

Abstract-The influence of different water Cl Ϫ (50-600 M), Ca 2ϩ (50-1,500 M), Na ϩ (50-1,500 M), or dissolved organic carbon (DOC, 0.31-5 mg/L) levels on silver-induced physiological and biochemical perturbations of rainbow trout were investigated. Fish were acclimated to soft water (50 M; Cl Ϫ , Ca 2ϩ , and Na ϩ ), then exposed to 3.7 g/L Ag (as AgNO 3 ) for 6 h, which resulted in a reduction in Na ϩ influx from the water, an inhibition of gill sodium-and potassium-activated adenosine triphosphatase (Na ϩ / K ϩ -ATPase) activity, and an accumulation of silver on the gills. Increasing the water Cl Ϫ or DOC levels ameliorated the silver toxicity. However, increasing water Ca 2ϩ or Na ϩ concentration did not reduce the silver-induced physiological and biochemical perturbations. The free silver ion (Ag ϩ ) concentrations (calculated from MINEQLϩ, a geochemical speciation computer program) showed a negative correlation with the Na ϩ influx rates and gill Na ϩ /K ϩ -ATPase activity. However, gill silver levels did not correlate to Ag ϩ concentrations and no correlation was found between gill silver levels and either Na ϩ influx rates or gill Na ϩ /K ϩ -ATPase activity. These results support the notion that the [Ag ϩ ] concentration is of major importance when assessing silver toxicity in fish, and that this should be taken into account in regulatory strategies for silver in the natural environment.

An early silver-gill binding model using conditional equilibrium binding constants (K) was fitted to actual toxicity data for rainbow trout (Oncorhynchus mykiss) and subsequently modified to produce a mechanistically based acute toxicity model for predicting silver toxicity. The model used an "off the shelf" aquatic geochemistry software program (MINEQL + ) and physiologically based log K values to predict the acute effects of waterborne silver in rainbow trout. The final version of the model does not predict total gill-silver loading, as the early model did, but rather predicts the binding of Ag + to key toxic sites on the gill and incorporates the effects of cation competition at these sites. The acute toxicity model for Ag + provided the best fit to toxicity data when a log K value for the affinity of these sites was 7.6 with cationic competition log K values for Na + and Ca 2+ of 2.9 and 2.3, respectively. A log K for Ag-DOM of 9.0 was used representing strong Ag + binding to dissolved organic matter. The model we present is easy to use and provides a good match with previously published acute AgNO 3 toxicity data for rainbow trout from 31 data sets in 10 studies. The modified model is now ready for full verification with a greater range of laboratory and natural waters.