Environmental Risk Assessment of Time-Variable Toxicant Exposure with Toxicokinetic-Toxicodynamic Modeling of Sublethal Endpoints and Moving Time Windows: A Case Study with Ceriodaphnia dubia.

Toxicokinetic-toxicodynamic (TKTD) modeling has received increasing attention in terms of the regulatory environmental risk assessment of chemicals. This type of mechanistic model can integrate all available data from individual-level bioassays into a single framework and enable refined risk assessments by extrapolating from laboratory results to time-variable exposure scenarios, based, for instance, on surface water exposure modeling (e.g., FOCUS). Dynamic energy budget (DEB) models coupled with TKTD modules (DEB-TKTD) constitute the leading approach to assess and predict sublethal effects of chemicals on individual organisms. However, thorough case studies are rare. We provide a state-of-the-art example with the standard aquatic test species Ceriodaphnia dubia and the fungicide azoxystrobin, including all steps, from bespoke laboratory toxicity tests to model calibration and validation, through to environmental risk assessment. Following the framework proposed in the European Food Safety Authority Scientific Opinion from 2018, we designed bespoke good laboratory practice-compliant laboratory toxicity studies based on test guideline 211 of the Organisation for Economic Co-operation and Development and then identified robust parameter values from those data for all relevant model parameters through model calibration. The DEB-TKTD model, DEBtox2019, then informed the design of the validation experiment. Once validated, the model was used to perform predictions for a time-variable exposure scenario generated by FOCUS. A moving time-window approach was used to perform the environmental risk assessment. This assessment method reduces uncertainty in the risk assessment while maintaining consistency with the traditional measures of risk. Environ Toxicol Chem 2024;00:1-13. © 2024 Syngenta Crop Protection AG. ibacon GmbH and The Author(s). Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Development of a mechanistic model for analyzing avian reproduction data for pesticide risk assessment.

Mechanistic effect models are increasingly recommended as tools for refining evaluations of risk from exposure to pesticides. In the context of bird and mammal risk assessments, DEB-TKTD models have been recommended for characterizing sublethal effects at lower tiers. However, there are currently no such models. Currently, chronic, multi-generational studies are performed to characterize potential effects of pesticides on avian reproduction, but it is has not been established to what extent results from these studies can inform effect models. Here, a standard Dynamic Energy Budget (DEB) model was extended to account for the avian toxicity endpoints observed in regulatory studies. We linked this new implementation to a toxicological module to capture observed pesticide effects on reproduction via a decreased efficiency of egg production. We analysed ten reproduction studies with five different pesticides conducted with the mallard (Anas platyrhynchos) and the northern bobwhite (Colinus virginianus). The new model implementation accurately distinguished between effects on egg production from direct mechanism of toxicity and from food avoidance. Due to the specific nature of regulatory studies, model applicability for risk refinement is currently limited. We provide suggestions for next steps in model development.

Fish Species Sensitivity Ranking Depends on Pesticide Exposure Profiles.

In the regulatory environmental risk assessment of plant protection products, the exposure tested in standard toxicity tests assumes simple exposure dynamics, such as constant exposure at the first stage of testing. However, environmental exposure can be highly dynamic. A species response to exposure is governed by toxicokinetics (TK) and toxicodynamics (TD). Therefore, it can be expected that the sensitivity of a species to a substance is dependent on the interplay of TKTD processes with the dynamics of the exposure. We investigated whether exposure dynamics affects species sensitivity of five fish species and if their sensitivity rankings differ among exposure profiles. We analyzed individual survival under projected surface water exposure to benzovindiflupyr. For this purpose, we calibrated compound- and species-specific reduced general unified threshold models of survival (GUTS-RED) models from standard laboratory toxicity data with the assumptions of stochastic death and individual tolerance. Using the calibrated models, we generated species sensitivity distributions based on median lethal profile multiplication factors for three characteristic exposure profiles. The analysis was performed using different GUTS-RED implementations: openGUTS (MATLAB® and Windows® versions) and the R package morse. The sensitivity rankings of the fish species changed as a function of exposure profile. For a multiple-peak scenario, rainbow trout was the most sensitive species. For a single peak followed by a slow concentration decline the most sensitive species was the fathead minnow (GUTS-RED-stochastic death) or the common carp (GUTS-RED-individual tolerance). Our results suggest that a single most sensitive species cannot be defined for all situations, all exposure profiles, and both GUTS-RED variants. Environ Toxicol Chem 2022;41:1732-1741. © 2022 Syngenta. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Application of General Unified Threshold Models of Survival Models for Regulatory Aquatic Pesticide Risk Assessment Illustrated with an Example for the Insecticide Chlorpyrifos.

Mathematical models within the General Unified Threshold models of Survival (GUTS) framework translate time-variable chemical exposure information into expected survival of animals. The GUTS models are species and compound specific and explicitly describe the internal exposure dynamics in an organism (toxicokinetics) and the related damage and effect dynamics (toxicodynamics), thereby connecting the external exposure concentration dynamics with the simulated mortality or immobility over time. In a recent scientific opinion on toxicokinetic-toxicodynamic (TKTD) models published by the European Food Safety Authority (EFSA), the GUTS modeling framework was considered ready for use in the aquatic risk assessment for pesticides and aquatic fauna. The GUTS models are suggested for use in risk assessment, if they are sufficiently validated for a specific substance-species combination. This paper aims to illustrate how they can be used in the regulatory environmental risk assessment for pesticides for a specific type of refinement, that is, when risks are triggered by lower tiers in acute as well as in chronic risk assessment and mortality or immobility is the critical endpoint. This approach involves the evaluation of time-variable exposure regimes in a so-called "Tier-2C" assessment. The insecticide chlorpyrifos was selected as an example compound because a large data set was available. The GUTS models for 13 different freshwater arthropods and 8 different theoretical aquatic exposure profiles were used to calculate a series of GUTS-based risk estimates, including exposure profile-specific multiplication factors leading to 50% mortality or immobility at the end of the tested profile (LP50/EP50) as "margins of safety." To put the use of GUTS models within the tiered aquatic risk assessment into perspective, GUTS models for the 13 aquatic arthropods were also used to predict the environmental risks of a measured chlorpyrifos exposure profile from an experimental ditch study (Tier-3 approach), and the results are discussed in the context of calibration of the tiered approach. Integr Environ Assess Manag 2021;17:243-258. © 2020 The Authors. Integrated Environmental Assessment and Management published by Wiley Periodicals LLC on behalf of Society of Environmental Toxicology & Chemistry (SETAC).

An energy-based model to analyze growth data of earthworms exposed to two fungicides.

The pesticide risk assessment for earthworms is currently performed using standardized tests, the model species Eisenia fetida, and the analyses of the data obtained are performed with ad hoc statistical tools. We assessed the impact of two fungicides on the entire growth pattern of the earthworm species Aporrectodea caliginosa, which is highly representative of agricultural fields. Individuals of three different ages (from hatching to 56 days old) were exposed to Cuprafor micro® (copper oxychloride) and Swing® Gold (dimoxystrobin and epoxiconazole). Data were analyzed with an energy-based toxicodynamic model coupled with a toxicokinetic model. The copper fungicide caused a drastic growth inhibition once the no effect concentration (NEC), estimated at 65 mg kg-1 of copper, was exceeded. The Swing® Gold negatively affected the growth with NEC values estimated at 0.387 mg kg-1 and 0.128 mg kg-1 for the dimoxystrobin and the epoxiconazole in this fungicide formulation, respectively. The time-profile of the effects on A. caliginosa individuals was fully accounted for by the model, whatever their age of exposure. Furthermore, toxicity data analyses, supported by measurements of fungicide concentrations in earthworm at the end of the experiment, allowed bettering understanding of the mechanisms of action of the fungicides towards earthworm growth.

Building and Applying Quantitative Adverse Outcome Pathway Models for Chemical Hazard and Risk Assessment.

An important goal in toxicology is the development of new ways to increase the speed, accuracy, and applicability of chemical hazard and risk assessment approaches. A promising route is the integration of in vitro assays with biological pathway information. We examined how the adverse outcome pathway (AOP) framework can be used to develop pathway-based quantitative models useful for regulatory chemical safety assessment. By using AOPs as initial conceptual models and the AOP knowledge base as a source of data on key event relationships, different methods can be applied to develop computational quantitative AOP models (qAOPs) relevant for decision making. A qAOP model may not necessarily have the same structure as the AOP it is based on. Useful AOP modeling methods range from statistical, Bayesian networks, regression, and ordinary differential equations to individual-based models and should be chosen according to the questions being asked and the data available. We discuss the need for toxicokinetic models to provide linkages between exposure and qAOPs, to extrapolate from in vitro to in vivo, and to extrapolate across species. Finally, we identify best practices for modeling and model building and the necessity for transparent and comprehensive documentation to gain confidence in the use of qAOP models and ultimately their use in regulatory applications. Environ Toxicol Chem 2019;38:1850-1865. © 2019 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals, Inc. on behalf of SETAC.

Toxicity-based toxicokinetic/toxicodynamic assessment of bioaccumulation and nanotoxicity of zerovalent iron nanoparticles in Caenorhabditis elegans.

Elucidating the relationships between the toxicity-based-toxicokinetic (TBTK)/toxicodynamic (TD) properties of engineered nanomaterials and their nanotoxicity is crucial for human health-risk analysis. Zerovalent iron (Fe0) nanoparticles (NPs) are one of the most prominent NPs applied in remediating contaminated soils and groundwater. However, there are concerns that Fe0NP application contributes to long-term environmental and human health impacts. The nematode Caenorhabditis elegans is a surrogate in vivo model that has been successfully applied to assess the potential nanotoxicity of these nanomaterials. Here we present a TBTK/TD approach to appraise bioaccumulation and nanotoxicity of Fe0NPs in C. elegans. Built on a present C. elegans bioassay with estimated TBTK/TD parameters, we found that average bioconcentration factors in C. elegans exposed to waterborne and food-borne Fe0NPs were ~50 and ~5×10-3, respectively, whereas 10% inhibition concentrations for fertility, locomotion, and development, were 1.26 (95% CI 0.19-5.2), 3.84 (0.38-42), and 6.78 (2.58-21) µg·g-1, respectively, implicating that fertility is the most sensitive endpoint in C. elegans. Our results also showed that biomagnification effects were not observed in waterborne or food-borne Fe0NP-exposed worms. We suggest that the TBTK/TD assessment for predicting NP-induced toxicity at different concentrations and conditions in C. elegans could enable rapid selection of nanomaterials that are more likely to be nontoxic in larger animals. We conclude that the use of the TBTK/TD scheme manipulating C. elegans could be used for rapid evaluation of in vivo toxicity of NPs or for drug screening in the field of nanomedicine.

A toxicokinetic study of specifically acting and reactive organic chemicals for the prediction of internal effect concentrations in Scenedesmus vacuolatus.

The toxic potency of chemicals is determined by using the internal effect concentration by accounting for differences in toxicokinetic processes and mechanisms of toxic action. The present study examines toxicokinetics of specifically acting and reactive chemicals in the green algae Scenedesmus vacuolatus by using an indirect method. Concentration depletion in the exposure medium was measured for chemicals of lower (log KOW < 3: isoproturon, metazachlor, paraquat) and moderate (log KOW 4-5: irgarol, triclosan, N-phenyl-2-naphthylamine) hydrophobicity at 7 to 8 time points over 240 min or 360 min. Uptake and overall elimination rates were estimated by fitting a toxicokinetic model to the observed concentration depletions. The equilibrium of exposure concentrations was reached within minutes to hours or was even not observed within the exposure time. The kinetics of bioconcentration cannot be explained by the chemical's hydrophobicity only, but influential factors such as ionization of chemicals, the ion trapping mechanism, or the potential susceptibility for biotransformation are discussed. Internal effect concentrations associated with 50% inhibition of S. vacuolatus reproduction were predicted by linking the bioconcentration kinetics to the effect concentrations and ranged from 0.0480 mmol/kg wet weight to 7.61 mmol/kg wet weight for specifically acting and reactive chemicals. Knowing the time-course of the internal effect concentration may promote an understanding of toxicity processes such as delayed toxicity, carry-over toxicity, or mixture toxicity in future studies.

Linking survival and biomarker responses over time.

The practical usefulness of biomarkers is limited by the complexity of linking their responses to life-history traits of the organisms (e.g., survival, growth, reproduction) over time. Here the authors present a first attempt to model biomarker responses and survival over time simultaneously with a toxicokinetic-toxicodynamic approach. Even though more work is clearly needed, the present study provides a novel direction for interpreting biomarker responses and dynamically linking them to life-history traits.