The Effects of Nickel on the Structure and Functioning of a Freshwater Plankton Community Under High Dissolved Organic Carbon Conditions: A Microcosm Experiment.

In the present study, we aimed to test the protectiveness of the bioavailability-normalization procedure, with its associated hazardous concentrations for x% of the species (HCx), that is currently implemented to derive environmental threshold concentrations for nickel (Ni) in European environmental legislative frameworks. We exposed a natural plankton-dominated community to 3 constant Ni concentrations, that is, a control with no Ni added (background Ni of 1.2-4 µg/L) and the bioavailability-normalized HC5 and HC50 of 24 and 97 µg dissolved Ni/L, respectively, during a 56-d microcosm experiment under high dissolved organic carbon (DOC) conditions (DOC of 14 mg/L at test initiation). The effects of the bioavailability-normalized HC5 and HC50 values were evaluated at the levels of community structure (community composition and plankton group abundances), community functioning (measured as indirect physicochemical proxies for overnight respiration and carbon fluxes), and individual species abundances. The bioavailability-normalized HC50 treatment had clear effects (defined as effects occurring on at least 2 consecutive sampling days) on both the structure and functioning of the investigated aquatic community. Through its effect on community functioning (i.e., reduced pH and DOC), Ni also influenced its own bioavailability. Clear direct effects of Ni were observed for only 3 species (the Cyanobacteria Oscillatoria sp. 1 and the rotifers Asplanchna/Testidunela sp. and Trichocerca group similis). Most other effects occurring in the plankton community in the HC50 treatment were indirect and likely driven by the direct effect of Ni on the Cyanobacteria Oscillatoria sp. 1, which was the dominant phytoplankton species in the control microcosms. In contrast, the bioavailability-normalized HC5 did not induce clear effects on community structure and functioning endpoints: these were only affected on individual sampling days. Clear (direct) effects were observed for only 2 plankton species (the rotifer Trichocerca group similis and the Cyanobacteria Oscillatoria sp. 1), but their abundances recovered to control levels at the end of the study. In addition, a few species (1 phytoplankton and 3 zooplankton species) were affected in the HC5 treatment only on the last sampling day. It is uncertain whether these species would have shown clear effects over a longer exposure duration. Thus, our study shows that the bioavailability-normalized HC5 of Ni at high DOC induced clear effects on a few individual species. However, the overall conclusion is that the bioavailability-normalized HC5 of Ni as derived through the procedure that is currently implemented in European legislative frameworks protects against clear effects on community structure and function. Environ Toxicol Chem 2019;38:1923-1939. © 2019 SETAC.

The effects of a mixture of copper, nickel, and zinc on the structure and function of a freshwater planktonic community.

It is generally assumed that as long as the majority of species experiences no direct adverse effects attributable to a single substance (i.e., potentially affected fraction [PAF] <5%), no significant structural or functional effects at the community level are expected to occur. Whether this assumption holds for mixed metal contamination is not known. In the present study, we tested this by performing a microcosm experiment in which a naturally occurring freshwater planktonic community was exposed to a copper-nickel-zinc (Cu-Ni-Zn) mixture for 8 wk and various structural and functional community-level traits were assessed. In the low mixture concentration treatments (i.e., Ni-Zn mixtures, because there was no difference in Cu concentrations in these treatments with the control), community-level effects were relatively simple, only involving phytoplankton species groups. In the high mixture concentration treatments (Cu-Ni-Zn mixtures), community-level effects were more complex, involving several phytoplankton and zooplankton species groups. Multisubstance PAF (msPAF) values for all mixture treatments were calculated by applying the concentration addition model to bioavailability-normalized single-metal species sensitivity distributions (SSDs). Consistent effects on the structural traits community composition, abundance of zooplankton species groups, species diversity, and species richness and on the functional trait dissolved organic carbon (DOC) concentration (as a proxy for the microbial loop and pelagic food web interactions) were only observed at msPAF values >0.05 (i.e., in the Cu-Ni-Zn mixture). However, consistent effects on the abundance of various phytoplankton species groups (structural traits) and on 2 measures of community respiration, overnight Δ dissolved oxygen (ΔDO) and ΔpH (functional traits), were already observed at msPAF values of ≤0.05 (i.e., in the Ni-Zn mixture). This indicates that the threshold msPAF value of 0.05 was not protective against metal mixture exposure for all community-level structural and functional endpoints in the present study. A possible explanation for this result is the mismatch between the species in the SSD and those in our microcosm community. Indeed, our data suggest that the presence of one single dominant and very Zn- and/or Ni-sensitive species in the investigated community (i.e., a cyanobacteria of the genus Oscillatoria), which is not represented in the SSD of these metals, was probably the driver of all observed effects at or below an msPAF of 0.05. Overall, the present results show that SSDs are not necessarily a good predictor of community-level effects for all types of communities and that the presence of dominant sensitive species may result in significant, consistent effects on certain structural and functional community-level endpoints at msPAF values ≤0.05, which is generally considered protective in many regulatory frameworks. Environ Toxicol Chem 2018;37:2380-2400. © 2018 SETAC.

Mixtures of Cu, Ni, and Zn act mostly noninteractively on Pseudokirchneriella subcapitata growth in natural waters.

Freshwater biota are usually exposed to mixtures of different metals in the environment, which raises concern because risk-assessment procedures for metals are still mainly based on single-metal toxicity. Because microalgae are primary producers and therefore at the base of the food web, it is of utmost importance to understand the effects of metal mixtures on these organisms. Most studies that have investigated the combined interactive effects of mixtures on microalgae performed tests in only one specific water. The objective of the present study was to test if combined effects of mixtures to Pseudokirchneriella subcapitata were the same or different across natural waters showing diverse water-chemistry characteristics. This was done by performing experiments with ternary Cu-Ni-Zn mixtures in 3 natural waters and with binary Cu-Ni mixtures in 5 natural waters. We showed that the ternary mixture acted noninteractively on algal growth, except in one water in which the mixture acted antagonistically. We suggest that a low-cationic competition situation in the latter water could be the reason for the antagonistic interaction between the metals. On the other hand, the binary mixture acted noninteractively on algal growth in all tested waters. We showed that both the concentration addition and independent action models can serve as accurate models for toxicity of ternary Cu-Ni-Zn and binary Cu-Ni mixtures to P. subcapitata in most cases and as protective models in all cases. In addition, we developed a metal mixture bioavailability model, by combining the independent action model and the single-metal bioavailability models, that can be used to predict Cu-Ni-Zn and Cu-Ni toxicity to P. subcapitata as a function of metal concentration and water characteristics. Environ Toxicol Chem 2018;37:587-598. © 2017 SETAC.

A framework for ecological risk assessment of metal mixtures in aquatic systems.

Although metal mixture toxicity has been studied relatively intensely, there is no general consensus yet on how to incorporate metal mixture toxicity into aquatic risk assessment. We combined existing data on chronic metal mixture toxicity at the species level with species sensitivity distribution (SSD)-based in silico metal mixture risk predictions at the community level for mixtures of Ni, Zn, Cu, Cd, and Pb, to develop a tiered risk assessment scheme for metal mixtures in freshwater. Generally, independent action (IA) predicts chronic metal mixture toxicity at the species level most accurately, whereas concentration addition (CA) is the most conservative model. Mixture effects are noninteractive in 69% (IA) and 44% (CA) and antagonistic in 15% (IA) and 51% (CA) of the experiments, whereas synergisms are only observed in 15% (IA) and 5% (CA) of the experiments. At low effect sizes (∼ 10% mixture effect), CA overestimates metal mixture toxicity at the species level by 1.2-fold (i.e., the mixture interaction factor [MIF]; median). Species, metal presence, or number of metals does not significantly affect the MIF. To predict metal mixture risk at the community level, bioavailability-normalization procedures were combined with CA or IA using SSD techniques in 4 different methods, which were compared using environmental monitoring data of a European river basin (the Dommel, The Netherlands). We found that the simplest method, in which CA is directly applied to the SSD (CASSD ), is also the most conservative method. The CASSD has median margins of safety (MoS) of 1.1 and 1.2 respectively for binary mixtures compared with the theoretically more consistent methods of applying CA or IA to the dose-response curve of each species individually prior to estimating the fraction of affected species (CADRC or IADRC ). The MoS increases linearly with an increasing number of metals, up to 1.4 and 1.7 for quinary mixtures (median) compared with CADRC and IADRC , respectively. When our methods were applied to a geochemical baseline database (Forum of European Geological Surveys [FOREGS]), we found that CASSD yielded a considerable number of mixture risk predictions, even when metals were at background levels (8% of the water samples). In contrast, metal mixture risks predicted with the theoretically more consistent methods (e.g., IADRC ) were very limited under natural background metal concentrations (<1% of the water samples). Based on the combined evidence of chronic mixture toxicity predictions at the species level and evidence of in silico risk predictions at the community level, a tiered risk assessment scheme for evaluating metal mixture risks is presented, with CASSD functioning as a first, simple conservative tier. The more complex, but theoretically more consistent and most accurate method, IADRC , can be used in higher tier assessments. Alternatively, the conservatism of CASSD can be accounted for deterministically by incorporating the MoS and MIF in the scheme. Finally, specific guidance is also given related to specific issues, such as how to deal with nondetect data and complex mixtures that include so-called data-poor metals. Environ Toxicol Chem 2018;37:623-642. © 2017 SETAC.

Analyzing the capacity of the Daphnia magna and Pseudokirchneriella subcapitata bioavailability models to predict chronic zinc toxicity at high pH and low calcium concentrations and formulation of a generalized bioavailability model for D. magna.

Risk assessment in the European Union implements Zn bioavailability models to derive predicted-no-effect concentrations for Zn. These models are validated within certain boundaries (i.e., pH ≤ 8 and Ca concentrations ≥ 5mg/L), but a substantial fraction of the European surface waters falls outside these boundaries. Therefore, we evaluated whether the chronic Zn biotic ligand model (BLM) for Daphnia magna and the chronic bioavailability model for Pseudokirchneriella subcapitata could be extrapolated to pH > 8 and Ca concentrations < 5 mg/L. Results from D. magna experiments suggested that the BLM is not able to reflect the pH effect over a broad pH range (5.5-8.5). In addition, because of Ca deficiency of D. magna in the soft water tests, we cannot conclude whether the BLM is applicable below its Ca boundary. Results for P. subcapitata experiments showed that the bioavailability model can accurately predict Zn toxicity for Ca concentrations down to 0.8 mg/L and pH values up to 8.5. Because the chronic Zn BLM for D. magna could not be extrapolated beyond its validity boundaries for pH, a generalized bioavailability model (gBAM) was developed. Of 4 gBAMs developed, we recommend the use of gBAM-D, which combines a log-linear relation between the 21-d median effective concentrations (expressed as free Zn2+ ion activity) and pH, with more conventional BLM-type competition constants for Na, Ca, and Mg. This model is a first step in further improving the accuracy of chronic toxicity predictions of Zn as a function of water chemistry, which can decrease the uncertainty in implementing the bioavailability-based predicted-no-effect concentration in the risk assessment of high-pH and low-Ca concentration regions in Europe. Environ Toxicol Chem 2017;36:2781-2798. © 2017 SETAC.

Comparison of four methods for bioavailability-based risk assessment of mixtures of Cu, Zn, and Ni in freshwater.

Although chemical risk assessment is still mainly conducted on a substance-by-substance basis, organisms in the environment are typically exposed to mixtures of substances. Risk assessment procedures should therefore be adapted to fit these situations. Four mixture risk assessment methodologies were compared for risk estimations of mixtures of copper (Cu), zinc (Zn), and nickel (Ni). The results showed that use of the log-normal species sensitivity distribution (SSD) instead of the best-fit distribution and sampling species sensitivities independently for each metal instead of using interspecies correlations in metal sensitivity had little impact on risk estimates. Across 4 different monitoring datasets, between 0% and 52% of the target water samples were estimated to be at risk, but only between 0% and 15% of the target water samples were at risk because of the mixture of metals and not any single metal individually. When a natural baseline database was examined, it was estimated that 10% of the target water samples were at risk because of single metals or their mixtures; when the most conservative method was used (concentration addition [CA] applied directly to the SSD, i.e., CASSD ). However, the issue of metal mixture risk at geochemical baseline concentrations became relatively small (2% of target water samples) when a theoretically more correct method was used (CA applied to individual dose response curves, i.e., CADRC ). Finally, across the 4 monitoring datasets, the following order of conservatism for the 4 methods was shown (from most to least conservative, with ranges of median margin of safety [MoS] relative to CASSD ): CASSD > CADRC (MoS = 1.17-1.25) > IADRC (independent action (IA) applied to individual dose-response curves; MoS = 1.38-1.60) > IASSD (MoS = 1.48-1.72). Therefore, it is suggested that these 4 methods can be used in a general tiered scheme for the risk assessment of metal mixtures in a regulatory context. In this scheme, the CASSD method could serve as a first (conservative) tier to identify situations with likely no potential risk at all, regardless of the method used (the sum toxic unit expressed relative to the 5% hazardous concentration [SumTUHC5 ] < 1) and the IASSD method to identify situations of potential risk, also regardless of the method used (the multisubstance potentially affected fraction of species using the IASSD method [msPAFIA,SSD ] > 0.05). The CADRC and IADRC methods could be used for site-specific assessment for situations that fall in between (SumTUHC5 > 1 and msPAFIA,SSD < 0.05). Environ Toxicol Chem 2017;36:2123-2138. © 2017 SETAC.

Comparison of chronic mixture toxicity of nickel-zinc-copper and nickel-zinc-copper-cadmium mixtures between Ceriodaphnia dubia and Pseudokirchneriella subcapitata.

Although aquatic organisms in the environment are exposed to mixtures of metals, risk assessment for metals is most commonly performed on a metal-by-metal basis. To increase the knowledge about chronic mixture effects, the authors investigated whether metal mixture effects are dependent on the biological species, mixture composition, and metal concentration ratio. The authors evaluated the effects of quaternary Ni-Zn-Cu-Cd and ternary Ni-Zn-Cu mixtures on 48-h algal growth rate (Pseudokirchneriella subcapitata) and 7-d daphnid reproduction (Ceriodaphnia dubia) using a ray design. Single metals were 3-fold to 42-fold more toxic for C. dubia than for P. subcapitata, based on the 50% effective concentration expressed as free metal activity, the range representing different metals. Statistical analysis of mixture effects showed that the ternary and quaternary mixture effects were antagonistic on algal growth relative to the concentration addition (CA) model, when the analysis was based on dissolved concentrations and on free metal ion activities. Using the independent action (IA) model, mixture effects in both rays were statistically noninteractive for algal growth when the analysis was based on dissolved concentrations; however, the interactions shifted toward antagonism when based on free ion activities. The ternary Ni-Zn-Cu mixture acted antagonistically on daphnid reproduction relative to both reference models, either expressed as free ion activities or dissolved concentrations. When Cd was added to the mixture, however, the mixture effects shifted toward noninteractivity for daphnids. The metal concentration ratio did not significantly influence the magnitude of observed antagonistic effects. Regardless of statistical interactions observed, based on the present study, CA and in most instances also IA can serve as a protective model for ternary Ni-Zn-Cu and quaternary Ni-Zn-Cu-Cd toxicity to both species. Environ Toxicol Chem 2017;36:1056-1066. © 2016 SETAC.

The effects of zinc on the structure and functioning of a freshwater community: A microcosm experiment.

A major problem with risk assessment of chemicals is the extrapolation of laboratory single-species toxicity tests, which oversimplify the actual field situation by ignoring species interactions, to natural communities. The authors tested if the bioavailability-normalized 5% hazardous concentration (HC5) estimated from chronic planktonic single-species toxicity data (HC5plankton ) for zinc (Zn) is protective for a plankton community and investigated the direct and indirect effects of Zn (at HC5 and HC50) on a freshwater community's structure and function. Microcosms were exposed to 3 different Zn concentrations (background, HC5plankton = 75 µg Zn/L and HC50plankton = 300 µg Zn/L) for 5 wk. The planktonic groups revealed a consistent no-observed-effect concentration for the community of 75 µg Zn/L, similar to or higher than the HC5plankton , thus suggesting its protectiveness in the present study. At 300 µg Zn/L a significant reduction in cladocerans resulted in increases of rotifer, ciliate, and phytoplankton abundance. In addition, the phytoplankton community shifted in dominance from grazing-resistant to edible species. Contrary to the species sensitivity distribution (SSD) prediction, which identified phytoplankton as the most sensitive group, only the total chlorophyll and the abundance of 2 phytoplankton species were adversely affected at 300 µg Zn/L. Thus, although the HC5 estimated from the bioavailability-normalized SSD was overall protective for the plankton community, the SSD was not able to correctly predict the species sensitivity ranking within their community context at the HC50. Environ Toxicol Chem 2016;35:2698-2712. © 2016 SETAC.

Comparison of the capacity of two biotic ligand models to predict chronic copper toxicity to two Daphnia magna clones and formulation of a generalized bioavailability model.

Although it is increasingly recognized that biotic ligand models (BLMs) are valuable in the risk assessment of metals in aquatic systems, the use of 2 differently structured and parameterized BLMs (1 in the United States and another in the European Union) to obtain bioavailability-based chronic water quality criteria for copper is worthy of further investigation. In the present study, the authors evaluated the predictive capacity of these 2 BLMs for a large dataset of chronic copper toxicity data with 2 Daphnia magna clones, termed K6 and ARO. One BLM performed best with clone K6 data, whereas the other performed best with clone ARO data. In addition, there was an important difference between the 2 BLMs in how they predicted the bioavailability of copper as a function of pH. These modeling results suggested that the effect of pH on chronic copper toxicity is different between the 2 clones considered, which was confirmed with additional chronic toxicity experiments. Finally, because fundamental differences in model structure between the 2 BLMs made it impossible to create an average BLM, a generalized bioavailability model (gBAM) was developed. Of the 3 gBAMs developed, the authors recommend the use of model gBAM-C(uni), which combines a log-linear relation between the 21-d median effective concentration (expressed as free Cu(2+) ion activity) and pH, with more conventional BLM-type competition constants for sodium, calcium, and magnesium. This model can be considered a first step in further improving the accuracy of chronic toxicity predictions of copper as a function of water chemistry (for a variety of Daphnia magna clones), even beyond the robustness of the current BLMs used in regulatory applications.