Acute and Chronic Toxicity of Cobalt to Freshwater Organisms: Using a Species Sensitivity Distribution Approach to Establish International Water Quality Standards.

Water quality standards for cobalt (Co) have not been developed for the European Union or United States. The objective of the present study was to produce freshwater Co toxicity data that could be used by both the European Union and the United States to develop appropriate regulatory standards (i.e., environmental quality standards or predicted-no-effect concentrations in Europe and ambient water quality criteria or state water quality standards in the United States). Eleven species, including algae, an aquatic plant, and several invertebrate and fish species, were used in the performance of acute and chronic Co toxicity tests. Acute median lethal or median effective concentration (LC50 or EC50) values ranged from 90.1 µg Co/L for duckweed (Lemna minor) to 157 000 µg Co/L for midges (Chironomus tentans). Chronic 10% effect concentration (EC10) values ranged from 4.9 µg Co/L for duckweed to 2170 µg Co/L for rainbow trout (Oncorhynchus mykiss). Chronic 20% effect concentration (EC20) values ranged from 11.1 µg Co/L for water flea (Ceriodaphnia dubia) to 2495 µg Co/L for O. mykiss. Results indicated that invertebrate and algae/plant species are more sensitive to chronic Co exposures than fish. Acute-to-chronic ratios (derived as acute LC50s divided by chronic EC20s) were lowest for juvenile O. mykiss (0.6) and highest for the snail Lymnaea stagnalis (2670). Following the European-based approach and using EC10 values, species sensitivity distributions (SSDs) were developed and a median hazardous concentration for 5% of the organisms of 1.80 µg Co/L was derived. Chronic EC20 values were used, also in an SSD approach, to derive a US Environmental Protection Agency-style final chronic value of 7.13 µg Co/L. Environ Toxicol Chem 2020;39:799-811. © 2020 SETAC.

The acute toxicity of nickel to Daphnia magna: predictive capacity of bioavailability models in artificial and natural waters.

The effects of Ca, Mg, Na and pH on the acute toxicity of Ni to Daphnia magna were investigated in a series of 48-h immobilization assays in synthetic test solutions. Both Ca and Mg reduced Ni toxicity, while Na did not. Ni toxicity was not affected in the pH range of 5.7-7.5, but a further increase of pH up to 8.1 resulted in an increase of toxicity of the free Ni2+ ion. Based on the results of these experiments, a biotic ligand model (BLM) was developed in which the effects of Ca and Mg were modeled as single-site competition effects. Stability constants representing the binding strength between Ca2+ and Mg2+ and the biotic ligand (BL) were logK(CaBL)=3.10 and logK(MgBL)=2.47, respectively. The effect of pH could not be appropriately described by single-site competition between Ni2+a nd H+. Since the overall variation of toxicity within the tested pH range was relatively small, we decided not to incorporate the effect of pH in the current model. The model was able to predict 48-h EC50s in all synthetic test solutions by an error less than factor 2. The model's predictive capacity was also evaluated using results of toxicity tests in Ni-spiked natural surface waters. For 15 out of 16 tested waters, 48-h EC50s were predicted by an error less than factor 2. Additionally, after calibration to account for interclonal or interspecies sensitivity differences, the model was able to accurately predict earlier published 48-h EC50s for another D. magna clone as well as for Ceriodaphnia dubia. Finally, the predictive capacity of the model was demonstrated to be better than that of previously proposed models that include a logK(NaBL), a logK(HBL) and a logK(CaBL), but did not incorporate a logK(MgBL). An in-depth comparison of these models learned that (i) there is no need to incorporate a logK(NaBL), (ii) it is important to recognize the protective effect of Mg, and (iii) the incorporation of a logK(HBL) does not adequately describe the effect of pH. Although our model seems very promising, further research, especially into the effects of elevated pH and alkalinity levels, is needed to allow further refinement.

Reduction of growth and haemolymph Ca levels in the freshwater snail Lymnaea stagnalis chronically exposed to cobalt.

The ecological risk assessment and the development of water-quality criteria for Co are currently still hampered by insufficient knowledge about the toxicity of Co to freshwater organisms. A relevant group of organisms, for which no toxicity data with Co are available, is the class of the herbivorous pulmonate freshwater snails, which fulfil a pivotal role in the consumption and decomposition of aquatic plants and epihyton. We measured the growth rate of the pond snail Lymnaea stagnalis chronically exposed for 28 days to a series of Co concentrations. The no observed effect concentration (NOEC) and the lowest observed effect concentration (LOEC) for growth rate were 26 and 79 microg Co/L, respectively. Growth rate of snails exposed to 79 microg Co/L and higher concentrations was more impaired in the final 2 weeks of exposure than in the first 2 weeks of exposure. The reduced growth rate at 79 microg Co/L was accompanied by a reduced concentration of Ca in the haemolymph at the end of the exposure. Possible mechanisms of toxicity of Co to snail growth were suggested to be an impairment of Ca uptake and homeostasis and/or feeding inhibition. Although additional research is needed to investigate the relative importance of these mechanisms, as well as the interrelatedness between them, the toxicity data currently presented can assist in risk assessment and water-quality criteria development.

Ambient copper concentrations in agricultural and natural European soils: an overview.

Copper concentrations in soil are affected by a large number of processes related to the natural spatial variability (geochemistry), the amount released, the spatial and temporal distributions of these releases, and the large number of transportation, complexation, and dissolution processes. The present study reports the generated country-specific and land use-specific environmental concentration distributions of ambient copper exposure levels in European soils that were used for the derivation of "reasonable worst-case" predicted environmental copper concentrations (RWC-ambient copper PEC) for three types of soil uses: Agricultural soils, forest soils, and undefined grassland soils. Only recent and high-quality monitoring data (Q1) that comply with a number of criteria (i.e., sampling strategy, land use, digestion method, and absence of point sources) were selected for this purpose. Data treatment procedures used in the present study were based on the methods and concepts laid down in the European Union Technical Guidance Document on Risk Assessment and in the "combined monitoring-based and modeling-based priority setting" procedure. The derived median RWC-ambient copper PEC for European agricultural soils is 31.1 mg/kg dry weight, with the lowest and highest RWC-ambient copper PEC found in Belgium (16.1 mg/kg dry wt) and northern Italy (57.5 mg/kg dry wt), respectively. The high value for Italian soils probably is related to the (recent) volcanic origin of these soils. Similarly, RWC-ambient copper PECs were derived for forest soils and undefined grassland soils (24.4 and 35.3 mg/kg, respectively). Observed differences between the diifferent soil uses could be related to various parameters and processes that determine the copper levels in soil (e.g., soil type associated with specitic soil uses, addition of fertilizers and pesticides, and presence of cattle).

Cross-phylum comparison of a chronic biotic ligand model to predict chronic toxicity of copper to a freshwater rotifer, Brachionus calyciflorus (Pallas).

Short chronic 48-h toxicity tests with the freshwater rotifer Brachionus calyciflorus (Pallas) were conducted to assess the modifying effects of pH and natural dissolved organic carbon (DOC) concentration on reproductive toxicity of Cu. Toxicity tests were carried out in four test waters according to a 2 x 2 design, in which pH (6 and 7.8) and DOC (5 and 15 mg C/L) were the test variables. Concentrations of dissolved Cu with no observed effect at 48 h (NOEC) varied 12-fold between 8.2 and 103 microg/L. Higher DOC and higher pH resulted in a reduction of toxicity, which is in line with the concepts of the biotic ligand model (BLM). A chronic Cu-BLM, originally developed for the cladoceran Daphnia magna, was calibrated to the rotifer dataset and was found to be able to predict all rotifer NOECs with an error factor of less than 1.6. This finding may be of great interest for risk assessment and the establishment of water quality criteria, as it suggests that chronic Cu-BLMs are comparable across phyla (i.e., arthropoda to rotifera).

Development of a chronic zinc biotic ligand model for Daphnia magna.

The individual effects of the cations Ca(2+), Mg(2+), Na(+), and H(+) on the chronic toxicity of Zn to the waterflea Daphnia magna were investigated in different series of univariate experiments, resulting in the development of a chronic Zn biotic ligand model (BLM) for this species. Using the mathematical approach based on a linear relationship between cation activity and metal activity at the EC(x) level, the following stability constants for binding of competing cations to the biotic ligand (BL) were derived: logK(CaBL) = 3.22, logK(MgBL) = 2.69, logK(NaBL) = 1.90, and logK(HBL) = 5.77. With the derived constants and a logK(ZnBL) of 5.31, two different BLMs that predict chronic zinc toxicity (EC(50), no observed effect concentration (NOEC)) for D. magna as a function of water characteristics were developed. Fractions of binding sites occupied by Zn at the considered effect levels EC(50) and NOEC were 0.127 and 0.084, respectively. The NOEC-based model predicts the Zn toxicity within a factor of 2, while the chronic EC(50) could be predicted within a factor of 1.5. In the future, these chronic Zn BLMs for D. magna can improve the ecological relevancy of zinc risk assessments by decreasing the bioavailability-related uncertainty of zinc toxicity.

Development and field validation of a predictive copper toxicity model for the green alga Pseudokirchneriella subcapitata.

In this study, the combined effects of pH, water hardness, and dissolved organic carbon (DOC) concentration and type on the chronic (72-h) effect of copper on growth inhibition of the green alga Pseudokirchneriella subcapitata were investigated. Natural dissolved organic matter (DOM) was collected at three sites in Belgium and The Netherlands using reverse osmosis. A full central composite test design was used for one DOM and a subset of the full design for the two other DOMs. For a total number of 35 toxicity tests performed, 72-h effect concentration resulting in 10% growth inhibition (EbC10s) ranged from 14.2 to 175.9 micrograms Cu/L (factor 12) and 72-h EbC50s from 26.9 to 506.8 micrograms Cu/L (factor 20). Statistical analysis demonstrated that DOC concentration, DOM type, and pH had a significant effect on copper toxicity; hardness did not affect toxicity at the levels tested. In general, an increase in pH resulted in increased toxicity, whereas an increase of the DOC concentration resulted in decreased copper toxicity. When expressed as dissolved copper, significant differences of toxicity reduction capacity were noted across the three DOM types tested (up to factor 2.5). When expressed as Cu2+ activity, effect levels were only significantly affected by pH; linear relationships were observed between pH and the logarithm of the effect concentrations expressed as free copper ion activity, that is, log(EbC50Cu2+) and log(EbC10Cu2+): (1) log(EbC50Cu2+)= - 1.431 pH + 2.050 (r2 = 0.95), and (2) log(EbC10cu2+) = -1.140 pH -0.812 (r2 = 0.91). A copper toxicity model was developed by linking these equations to the WHAM V geochemical speciation model. This model predicted 97% of the EbC50dissolved and EbC10dissolved values within a factor of two of the observed values. Further validation using toxicity test results that were obtained previously with copper-spiked European surface waters demonstrated that for 81% of tested waters, effect concentrations were predicted within a factor of two of the observed. The developed model is considered to be an important step forward in accounting for copper bioavailability in natural systems.

Algal toxicity tests for environmental risk assessments of metals.

Current regulatory methods for assessing the effects of contaminants, and metals in particular, rely mainly on a limited number of standardized test methods and test species (OECD, ISO, ASTM, USEPA). However, these test protocols allow a certain degree of freedom in relation to physicochemical parameters or biological aspects, which may lead to large variability in test results. The current review, based on effects data and theoretical considerations reported in the literature, tried to determine and quantify the effect of variation of these factors on the outcome of metal toxicity tests with algae. Major physicochemical parameters that affect metal toxicity to algae are hardness, pH, preculture conditions, type of test medium, and presence of chelating agents: Literature data also clearly demonstrate the importance of test species or strain selection (inter- and intraspecies sensitivity variability) on the outcome of algal toxicity tests. For Zn, a factor of 8.3 is observed between the NOEC for Selenastrum capricornutum (currently renamed Pseudokirchneriella subcapitata) and Croococcus paris. An intraspecies difference for S. capricornutum of a factor of 60 is observed between various reported EC50S for Cd. Next to differences in physicochemical test conditions, possible adaptation or acclimation to deficient/elevated metal concentrations add to the reported differences: S. capricornutum became three times less sensitive to Zn when acclimated to 65 microg Zn/L compared to cultures in ISO medium. This review has revealed that currently accepted standard protocols used in regulatory frameworks contain a number of major shortcomings on the physicochemical and biological aspects of algal toxicity testing with metals. These shortcomings are summarized in Table 5, together with a number of suggestions that could help to modify and improve standard test protocols for evaluating metal toxicity to algae. Until now, important factors such as pH control during test performance, selection of test medium, test species, and the effects of possible adaptation/acclimation to natural metal concentrations have not been considered, which could have serious implications when the resulting unsuitable or irrelevant toxicity data are subsequently used for setting environmental management policies. These findings also have their consequences when extrapolating laboratory data to the field as the complexity of natural waters currently is not reflected in laboratory standard media. These media contain no dissolved organic matter, have a relatively high pH, and contain large amounts of essential nutrients. In addition, the limited number of laboratory test species do not reflect natural phytoplankton communities. Test procedures for assessing the environmental impact of metal contamination in a specified ecoregion should therefore be based on performing a battery of algal tests with species adapted to and tested under the specific natural conditions of the region.

Refinement and field validation of a biotic ligand model predicting acute copper toxicity to Daphnia magna.

A previously developed biotic ligand model (BLM) was validated for its capacity to predict acute 48-h EC(50) values of copper to Daphnia magna in 25 reconstituted media with different pH values and concentrations of artificial dissolved organic carbon, Ca, Mg and Na. Before the BLM validation, fitting of measured (with a copper ion-selective electrode) and calculated (with the BLM) Cu(2+)-activity was performed by adjusting the WHAM model V (i.e. the metal-organic speciation part of the BLM) copper-proton exchange constant to pK(MHA)=1.9. Using this value, the 48-h EC(50) values observed agreed very well with BLM-predicted EC(50) values for tests performed at pH<8, but not at all for tests performed at pH>8. Additional experiments demonstrated that this was due to toxicity of the CuCO(3) complex, which is the most abundant inorganic copper species at pH>8. This was incorporated into the initial BLM by allowing the binding of CuCO(3) (next to Cu(2+) and CuOH(+)) to the biotic ligand of D. magna. The affinity of CuOH(+) and CuCO(3) for the biotic ligand was approximately five- and 10-fold lower than that of Cu(2+), respectively. With the refined BLM, 48-h EC(50) values could be accurately predicted within a factor of two not only in all 25 reconstituted media, but also in 19 natural waters. This validated and refined BLM could support efforts to improve the ecological relevance of risk assessment procedures applied at present.

Predicting acute zinc toxicity for Daphnia magna as a function of key water chemistry characteristics: development and validation of a biotic ligand model.

The individual effect of different major cations (Ca2+, Mg2+, Na+, K+, and H+) on the acute toxicity of zinc to the waterflea Daphnia magna was investigated. The 48-h median effective concentration (EC50) in the baseline test medium (i.e., a standard medium with very low ion concentrations) was about 6 microM (Zn2+). An increase of Ca2+ (from 0.25 mM to 3 mM), Mg2+ (from 0.25 mM to 2 mM), and Na+ activity (from 0.077 mM to 13 mM) reduced zinc toxicity by a factor of 6.3, 2.1, and 3.1, respectively. No further toxicity reduction was observed when Ca2+ and Mg2+ activities exceeded 3.0 and 2.0 mM, respectively. Both K+ and H+ did not significantly alter zinc toxicity (expressed as Zn2+ activity). From these data, conditional stability constants for Ca2+ (log K = 3.24), Mg2+ (log K = 2.97), Na+ (log K = 2.16), and Zn2+ (log K = 5.31) were derived and incorporated into a biotic ligand model (BLM) predicting acute zinc toxicity to D. magna in surface waters with different water quality characteristics. Validation of the developed BLM using 17 media with different pH, hardness, and dissolved organic carbon (DOC) content resulted in a significant correlation coefficient (R2 = 0.76) between predicted and observed 48-h EC50. Eighty-eight percent of the predictions were within a factor of 1.3 of the observed 48-h EC50.