Validation of Bioavailability-Based Toxicity Models for Metals.

Regulatory jurisdictions worldwide are increasingly incorporating bioavailability-based toxicity models into development of protective values (PVALs) for freshwater and saltwater aquatic life (e.g., water quality criteria, standards, and/or guidelines) for metals. Use of such models for regulatory purposes should be contingent on their ability to meet performance criteria as specified through a model-validation process. Model validation generally involves an assessment of a model's appropriateness, relevance, and accuracy. We review existing guidance for validation of bioavailability-based toxicity models, recommend questions that should be addressed in model-validation studies, discuss model study type and design considerations, present several new ways to evaluate model performance in validation studies, and suggest a framework for use of model validation in PVAL development. We conclude that model validation should be rigorous but flexible enough to fit the user's purpose. Although a model can never be fully validated to a level of zero uncertainty, it can be sufficiently validated to fit a specific purpose. Therefore, support (or lack of support) for a model should be presented in such a way that users can choose their own level of acceptability. We recommend that models be validated using experimental designs and endpoints consistent with the data sets that were used to parameterize and calibrate the model and validated across a broad range of geographically and ecologically relevant water types. Environ Toxicol Chem 2019;39:101-117. © 2019 SETAC.

Modeling the Fate of Metal Concentrates in Surface Water.

Metals present in concentrates are in a solid form and are not bioavailable, but they can dissolve or potentially transform to more soluble forms. Transformation/dissolution laboratory protocols have been developed to assess the importance of dissolution of sparingly soluble metal substances in the context of hazard classification; however, these tests represent worst-case scenarios for metal bioavailability because attenuation mechanisms such as complexation, sorption, and transport to the sediment are not considered. A unit world model (UWM) for metals in lakes, tableau input coupled kinetics equilibrium transport (TICKET)-UWM, has been developed that considers key processes affecting metal transport, fate, and toxicity including complexation by aqueous inorganic and ligands, partitioning to dissolved organic carbon (DOC) and particulate organic carbon (POC), precipitation, and transport of dissolved metals and solids between the water column and sediment. The TICKET-UWM model was used to assess the fate of a metal concentrate and dissolved metal ions released from the concentrate following an instantaneous input to a generalized lake. Concentrate dissolution rates in the water column were parameterized using results from batch transformation/dissolution tests for 2 specific concentrates containing lead (Pb), copper (Cu), and cobalt (Co). The TICKET-UWM results for a generalized lake environment showed that water column concentrations of metals in the lake environment after 28 d were several orders of magnitude lower than the 28-d concentration from the transformation/dissolution tests because Pb, Cu, and Co partitioned to POC in the water column and were subsequently removed due to settling. Resuspension of sediment served to increase total metal in the water column, but the resulting concentrations were still much lower than the 28-d concentrations from the transformation/dissolution tests. Information from TICKET-UWM could be used to refine the environmental hazard profiles of metals. Environ Toxicol Chem 2019;38:1256-1272. © 2019 SETAC.

Metal mixtures modeling evaluation project: 1. Background.

Despite more than 5 decades of aquatic toxicity tests conducted with metal mixtures, there is still a need to understand how metals interact in mixtures and to predict their toxicity more accurately than what is currently done. The present study provides a background for understanding the terminology, regulatory framework, qualitative and quantitative concepts, experimental approaches, and visualization and data-analysis methods for chemical mixtures, with an emphasis on bioavailability and metal-metal interactions in mixtures of waterborne metals. In addition, a Monte Carlo-type randomization statistical approach to test for nonadditive toxicity is presented, and an example with a binary-metal toxicity data set demonstrates the challenge involved in inferring statistically significant nonadditive toxicity. This background sets the stage for the toxicity results, data analyses, and bioavailability models related to metal mixtures that are described in the remaining articles in this special section from the Metal Mixture Modeling Evaluation project and workshop. It is concluded that although qualitative terminology such as additive and nonadditive toxicity can be useful to convey general concepts, failure to expand beyond that limited perspective could impede progress in understanding and predicting metal mixture toxicity. Instead of focusing on whether a given metal mixture causes additive or nonadditive toxicity, effort should be directed to develop models that can accurately predict the toxicity of metal mixtures.

Metal mixture modeling evaluation project: 3. Lessons learned and steps forward.

A comparison of 4 metal mixture toxicity models (that were based on the biotic ligand model [BLM] and the Windermere humic aqueous model using the toxicity function [WHAM-FTOX ]) was presented in a previous paper. In the present study, a streamlined version of the 4 models was developed and applied to multiple data sets and test conditions to examine key assumptions and calibration strategies that are crucial in modeling metal mixture toxicity. Results show that 1) a single binding site on or in the organism was a useful and oftentimes sufficient framework for predicting metal toxicity; 2) a linear free energy relationship (LFER) for bidentate binding of metals and cations to the biotic ligand provided a good first estimate of binding coefficients; 3) although adjustments in metal binding coefficients or adjustments in chemical potency factors can both be used in model calibration for single-metal exposures, changing metal binding coefficients or chemical potency factors had different effects on model predictions for metal mixtures; and 4) selection of a mixture toxicity model (based on concentration addition or independent action) was important in predicting metal mixture toxicity. Moving forward, efforts should focus on reducing uncertainties in model calibration, including development of better methods to characterize metal binding to toxicologically active binding sites, conducting targeted exposure studies to advance the understanding of metal mixture toxicity, and further developing LFERs and other tools to help constrain the model calibration.

Metal mixture modeling evaluation project: 2. Comparison of four modeling approaches.

As part of the Metal Mixture Modeling Evaluation (MMME) project, models were developed by the National Institute of Advanced Industrial Science and Technology (Japan), the US Geological Survey (USA), HDR|HydroQual (USA), and the Centre for Ecology and Hydrology (United Kingdom) to address the effects of metal mixtures on biological responses of aquatic organisms. A comparison of the 4 models, as they were presented at the MMME workshop in Brussels, Belgium (May 2012), is provided in the present study. Overall, the models were found to be similar in structure (free ion activities computed by the Windermere humic aqueous model [WHAM]; specific or nonspecific binding of metals/cations in or on the organism; specification of metal potency factors or toxicity response functions to relate metal accumulation to biological response). Major differences in modeling approaches are attributed to various modeling assumptions (e.g., single vs multiple types of binding sites on the organism) and specific calibration strategies that affected the selection of model parameters. The models provided a reasonable description of additive (or nearly additive) toxicity for a number of individual toxicity test results. Less-than-additive toxicity was more difficult to describe with the available models. Because of limitations in the available datasets and the strong interrelationships among the model parameters (binding constants, potency factors, toxicity response parameters), further evaluation of specific model assumptions and calibration strategies is needed.

Modeling water column partitioning of polychlorinated biphenyls to natural organic matter and black carbon.

High volume in situ surface water samples were collected from a tidal tributary of the Delaware Estuary using an Infiltrex sampling system equipped with a 1 µm particle filter and a XAD-2 resin column. Particulate and dissolved phase polychlorinated biphenyl (PCB) congeners were analyzed using high resolution gas chromatography/high resolution mass spectrometry to obtain detection levels in the femtograms per liter range. The data were fit to a four-phase equilibrium partitioning model including freely dissolved PCB, PCB bound to particulate organic carbon (POC), PCB bound to dissolved organic carbon (DOC), and PCB bound to black carbon (BC). Isotherms were assumed to be linear for POC and DOC and nonlinear for BC. The partition coefficient between BC and dissolved PCB was assumed to depend on the dihedral angle between the phenyl rings. Following parameter optimization, the correlation coefficient between the log of the modeled and measured apparent distribution coefficient Kp,app was 0.94, and the RMSE was 0.189 log units. Including BC in the model reduces the dissolved PCB phase concentration in the water column for all congeners, especially for the non-ortho and mono-ortho substituted congeners.

Oxidation of Cr(III) to Cr(VI) during chlorination of drinking water.

Drinking water treatment typically uses strong oxidants such as chlorine which are capable of converting Cr(III) to Cr(VI). The rates and extent of Cr(III) oxidation by chlorine are not well established. Cr(III) oxidation experiments were therefore conducted in distilled deionized water and New York City tap water dosed initially with Cr(III) and supplemented with sodium hypochlorite to increase free chlorine residual. Reaction progress was monitored using capillary electrophoresis which quenched reactions and allowed for quantification of Cr(VI). Three different forms of Cr(III) were used as reactants: a Cr(III) nitrate salt, Cr(III)-EDTA, and Cr(III) hydroxide. Rates of Cr(VI) production for all three forms of Cr(III) were rapid, on the order of hours. However, oxidation rates slowed and a plateau in Cr(VI) concentrations was reached. This resulted in less than 100% conversion of Cr(III) to Cr(VI) even at relatively high chlorine doses (10 to 100 mg L(-1) as Cl(2)). The loss of free chlorine due to a non-Cr chlorine demand, the precipitation of Cr(III) to Cr(OH)(3)(s), and the partial oxidation of Cr(III) to intermediate oxidation states (i.e. Cr(IV) and Cr(V)) were examined and eliminated as possible explanations for this behavior. Consumption of chlorine via reaction with intermediate oxidation states of Cr is therefore offered as a possible explanation for the plateau in Cr(VI) concentrations.

Comparing laboratory- and field-measured biota-sediment accumulation factors.

Standardized laboratory protocols for measuring the accumulation of chemicals from sediments are used in assessing new and existing chemicals, evaluating navigational dredging materials, and establishing site-specific biota-sediment accumulation factors (BSAFs) for contaminated sediment sites. The BSAFs resulting from the testing protocols provide insight into the behavior and risks associated with individual chemicals. In addition to laboratory measurement, BSAFs can also be calculated from field data, including samples from studies using in situ exposure chambers and caging studies. The objective of this report is to compare and evaluate paired laboratory and field measurement of BSAFs and to evaluate the extent of their agreement. The peer-reviewed literature was searched for studies that conducted laboratory and field measurements of chemical bioaccumulation using the same or taxonomically related organisms. In addition, numerous Superfund and contaminated sediment site study reports were examined for relevant data. A limited number of studies were identified with paired laboratory and field measurements of BSAFs. BSAF comparisons were made between field-collected oligochaetes and the laboratory test organism Lumbriculus variegatus and field-collected bivalves and the laboratory test organisms Macoma nasuta and Corbicula fluminea. Our analysis suggests that laboratory BSAFs for the oligochaete L. variegatus are typically within a factor of 2 of the BSAFs for field-collected oligochaetes. Bivalve study results also suggest that laboratory BSAFs can provide reasonable estimates of field BSAF values if certain precautions are taken, such as ensuring that steady-state values are compared and that extrapolation among bivalve species is conducted with caution.

Comparing laboratory and field measured bioaccumulation endpoints.

An approach for comparing laboratory and field measures of bioaccumulation is presented to facilitate the interpretation of different sources of bioaccumulation data. Differences in numerical scales and units are eliminated by converting the data to dimensionless fugacity (or concentration-normalized) ratios. The approach expresses bioaccumulation metrics in terms of the equilibrium status of the chemical, with respect to a reference phase. When the fugacity ratios of the bioaccumulation metrics are plotted, the degree of variability within and across metrics is easily visualized for a given chemical because their numerical scales are the same for all endpoints. Fugacity ratios greater than 1 indicate an increase in chemical thermodynamic activity in organisms with respect to a reference phase (e.g., biomagnification). Fugacity ratios less than 1 indicate a decrease in chemical thermodynamic activity in organisms with respect to a reference phase (e.g., biodilution). This method provides a holistic, weight-of-evidence approach for assessing the biomagnification potential of individual chemicals because bioconcentration factors, bioaccumulation factors, biota-sediment accumulation factors, biomagnification factors, biota-suspended solids accumulation factors, and trophic magnification factors can be included in the evaluation. The approach is illustrated using a total 2393 measured data points from 171 reports, for 15 nonionic organic chemicals that were selected based on data availability, a range of physicochemical partitioning properties, and biotransformation rates. Laboratory and field fugacity ratios derived from the various bioaccumulation metrics were generally consistent in categorizing substances with respect to either an increased or decreased thermodynamic status in biota, i.e., biomagnification or biodilution, respectively. The proposed comparative bioaccumulation endpoint assessment method could therefore be considered for decision making in a chemicals management context.

TICKET-UWM: a coupled kinetic, equilibrium, and transport screening model for metals in lakes.

The tableau input coupled kinetic equilibrium transport-unit world model (TICKET-UWM) has been developed as a screening model for assessing potential environmental risks associated with the release of metals into lakes. The model is based on a fully implicit, one-step solution algorithm that allows for simultaneous consideration of dissolved and particulate phase transport; metal complexation to organic matter and inorganic ligands; precipitation of metal hydroxides, carbonates, and sulfides; competitive interactions of metals and major cations with biotic ligands; a simplified description of biogeochemical cycling of organic carbon and sulfur; and dissolution kinetics for metal powders, massives, and other solid forms. Application of TICKET-UWM to a generalized lake in the Sudbury area of the Canadian Shield is presented to demonstrate the overall cycling of metals in lakes and the nonlinear effects of chemical speciation on metal responses. In addition, the model is used to calculate critical loads for metals, with acute toxicity of Daphnia magna as the final endpoint. Model results show that the critical loads for Cu, Ni, Pb, and Zn varied from 2.5 to 39.0 g metal/m(2) -year and were found to be one or more orders of magnitude higher than comparable loads for pesticides (lindane, 4,4'-DDT) and several polyaromatic hydrocarbon (PAH) compounds. In sensitivity calculations, critical metal-loading rates were found to vary significantly as a function of the hydraulic detention time, water hardness, and metal dissolution kinetic rates.

Tableau input coupled kinetic equilibrium transport (TICKET) model.

TICKET is a general-purpose, multispecies reactive transport model that is based on the tableau structure in MINEQL. The model can be used in solving problems from simple chemical equilibrium calculations for batch systems to complex one-dimensional, reactive transport computations for surface water, groundwater, and water treatment systems. To streamline model input and model formulation, specifications of equilibrium speciation (including homogeneous precipitation, solid solutions, adsorption, and ion exchange) and linear and nonlinear kinetic reactions are defined directly in the tableau. In addition, the burden of accounting for appearing and disappearing solid phases is circumvented by approximating homogeneous precipitation as a solid solution (with an insoluble seed). The solution technique for the model is based on a one-step algorithm and can be applied to both steady-state and fully implicit, time-variable problems. This approach is particularly well-suited in handling chemical speciation-transport problems with fast, nonlinear reaction kinetics and transient chemical intermediates. TICKET model simulations are presented for several test cases to verify the computational scheme. A model application, which examines the effects of sorption and overlying dissolved oxygen concentration on Fe(II) and As(III) oxidation in a sediment column, is also presented to demonstrate the utility of TICKET in examining complex chemical speciation-transport behavior.

Iron(II)-catalyzed oxidation of arsenic(III) in a sediment column.

Arsenic contamination in aquatic systems is a worldwide concern. Understanding the redox cycling of arsenic in sediments is critical in evaluating the fate of arsenic in aquatic environments and in developing sediment quality guidelines. The direct oxidation of inorganic trivalent arsenic, As(III), by dissolved molecular oxygen has been studied and found to be quite slow. A chemical pathway for As(III) oxidation has been proposed recently in which a radical species, Fe(IV), produced during the oxidation of divalent iron, Fe(II), facilitates the oxidation of As(III). Rapid oxidation of As(III) was observed (on a time scale of hours) in batch systems at pH 7 and 7.5, but the extent of As(III) oxidation was limited. The Fe(II)-catalyzed oxidation of As(III) is examined in a sediment column using both computational and experimental studies. A reactive-transport model is constructed that incorporates the complex kinetics of radical species generation and Fe(II) and As(III) oxidation that have been developed previously. The model is applied to experimental column data. Results indicate that the proposed chemical pathway can explain As(III) oxidation in sediments and that transport in sediments plays a vital role in increasing the extent of As(III) oxidation and efficiency of the Fe(II) catalysis.

Thermodynamic analysis of arsenic methylation.

The Challenger mechanism for the methylation of arsenic is a repeating sequence of a two-electron reduction of pentavalent arsenic As(V) species to trivalent arsenic As(III) species followed by a methylation-oxidation reaction forming the successive methyl As(V) species. This unusual oxidation-reduction sequence prompted an examination of the thermodynamics of these reactions. Quantum chemical methods are employed to estimate the thermodynamic parameters for the methyl arsenic species. The sequence is thermodynamically favored at neutral pH for redox potentials with pe < 0 and methyl cation activities pCH3+ < -3 to -7 depending on the precise situation analyzed. The observed distribution of methyl arsenic species in human urine, which is remarkably constant across many studied populations, can be reproduced using an equilibrium model if the formation of TMA species is prevented. The estimated thermodynamic parameters are sufficiently accurate to evaluate questions of thermodynamic plausibility but not the precise details of speciation.

Effect of thioarsenite formation on arsenic(III) toxicity.

Soluble arsenic(III)-sulfide complexes (thioarsenites) play a significant role in the chemistry of arsenic in reducing, sulfidic environments at circumneutral pH. Chemical equilibrium calculations using thioarsenite thermodynamic data from the literature indicate that the formation of a dithioarsenite complex, AsS(OH)(SH)(-1), reduces the concentration of the uncomplexed inorganic As(III) species present (defined sigma H3AsO3, where sigma H3AsO3 = AsO3(-3) + HAsO3(-2) + H2AsO3(-1) + H3AsO3). With enough sulfide present, soluble As(III) is dominated by this complex. Therefore, it is of interest to examine the effect of dithioarsenite formation on As(III) toxicity. The Microtox acute toxicity test was used for this purpose. Tests performed on solutions with varying S:As ratios indicate that As(III) toxicity is a function of the uncomplexed As(III) concentration rather than the total As(III) concentration. This suggests that the dithioarsenite species is not bioavailable and that its formation reduces As(III) toxicity. Chemical equilibrium calculations and sediment pore-water field data from various sources indicate that, in many sediments, dithioarsenite formation can reduce toxicity.