Annika Agatz

Toxicokinetic‐toxicodynamic modeling of quantal and graded sublethal endpoints: A brief discussion of concepts

We report on the advantages and problems of using toxicokinetic-toxicodynamic (TKTD) models for the analysis, understanding, and simulation of sublethal effects. Only a few toxicodynamic approaches for sublethal effects are available. These differ in their effect mechanism and emphasis on linkages between endpoints. We discuss how the distinction between quantal and graded endpoints and the type of linkage between endpoints can guide model design and selection. Strengths and limitations of two main approaches and possible ways forward are outlined.

Feeding inhibition explains effects of imidacloprid on the growth, maturation, reproduction, and survival of Daphnia magna.

Effects of some xenobiotics on aquatic organisms might not be caused directly by the compound but rather arise from acclimation of the organism to stress invoked by feeding inhibition during exposure. Experiments were conducted to identify effects of imidacloprid on individual performance (feeding, growth, maturation, reproduction, and survival) of Daphnia magna under surplus and reduced food availability. Concentrations inhibiting feeding by 5, 50, and 95% after one day of exposure were 0.19, 1.83, and 8.70 mg/L, respectively. Exposure with imidacloprid at ≥ 3.7 mg/L reduced growth by up to 53 ± 11% within one week. Surplus food availability after inhibition allowed recovery from this growth inhibition, whereas limited food supply eliminated the potential for recovery in growth even for exposure at 0.15 mg/L. A shift in the distribution of individual energy reserves toward reproduction rather than growth resulted in increased reproduction after exposure to concentrations ≤ 0.4 mg/L. Exposure to imidacloprid at ≥ 4.0 mg/L overwhelmed this adaptive response and reduced reproduction by up to 57%. We used the individual based Daphnia magna population model IDamP as a virtual laboratory to demonstrate that only feeding was affected by imidacloprid, and that in turn this caused the other impacts on individual performance. Consideration of end points individually would have led to a different interpretation of the effects. Thus, we demonstrate how multiple lines of evidence linked by understanding the ecology of the organism are necessary to elucidate xenobiotic impacts along the effect cascade.

Imidacloprid perturbs feeding of Gammarus pulex at environmentally relevant concentrations

Changes in food uptake by detritivorous macroinvertebrates could disrupt the ecosystem service of leaf litter breakdown, necessitating the study of shredding under anthropogenic influences. The impact of the neonicotinoid insecticide imidacloprid on the feeding rate of individual Gammarus pulex was measured at a daily resolution both during and after a 4-d exposure period. The authors found that imidacloprid inhibits feeding of G. pulex during exposure at concentrations ≥ 30 µg/L and that there was no recovery in feeding on transfer into clean media for 3 d. Exposure to imidacloprid at concentrations ≥ 0.81 µg/L and ≤ 9.0 µg/L resulted in increased feeding after exposure even though there was no significant effect on feeding during the exposure itself. Comparison with the literature shows that concentrations found to influence feeding lie within the range of estimated and measured environmental concentrations. Additionally, effects on feeding rate were observed at concentrations 2 orders of magnitude lower than those causing mortality. The lethal concentration for 50% of test organisms after 4 d of exposure (270 µg/L, literature data) and the effect concentration for a reduction in feeding by 50% (5.34 µg/L) were used for this comparison. The present study discusses the potential that effects on feeding may evoke effects at the population level or disturb leaf litter breakdown in the environment.

Promoting effects on reproduction increase population vulnerability of Daphnia magna.

Environmental risk assessment of chemicals is based on single species tests at the individual level with single compounds. However, the protection goal is the sustainability of a population, which faces several natural stressors and mixtures of chemicals in the environment. Therefore, experiments were undertaken to quantify the combined effects of chemicals with different modes of action on Daphnia magna populations. Populations continuously exposed to dispersogen A and at abundance equilibrium were treated with a 2-d pulse of p353-nonylphenol. In previous studies, dispersogen A was shown to act as a natural info-chemical, promoting the reproduction of daphnids (higher offspring quantity) coupled with reduced offspring fitness, whereas nonylphenol in pulsed-exposure caused size-selective mortality. Dispersogen A caused accelerated population growth to maximum abundance, shifted the population structure towards smaller individuals, and increased the population sensitivity to nonylphenol. The authors showed that a positive effect observed at the individual level can be transposed to a negative effect when monitored at the population level. So far, positive effects are not addressed in environmental risk assessment, and even in higher-tier testing, population structure is not quantified. Both factors indicate a potential mismatch between protection aim and risk assessment practice.

Family-portraits for daphnids : scanning living individuals and populations to measure body length

A method has been developed and tested to determine the body length of living daphnids. The purpose of the method was the simple, accurate, repeatable, quick, and to the living organism, harmless measurement of body length of all individuals in a population to enhance the capability of observing population development over time. Generally, organisms are transferred to a petri dish and temporarily fixed by removing access medium. A picture of the petri dish is taken using an ordinary flatbed scanner. Pictures are manually analysed with purposely developed software. We found no significant impact of the method on either individual performance (growth and reproduction) or population development (abundance and structure) of daphnids in comparison to the previously used method for data gathering (sieving, counting and length measurement of a subsample via microscopy). The disadvantage of our method, an increased demand in time for picture analysis, is negligible compared to the advantages this method has. Data generated with the new method do represent the population structure more accurately than those data generated with the previously used method. Scanning organisms does also allow a retrospective quality control for generated data as pictures can securely be stored. The quality of the pictures is furthermore sufficient to include additional endpoints to the analysis (e.g., number and size of aborts, number and size of eggs in the brood pouch, spine length). Here, we present, test and discuss an alternative approach to automated image analysis for data gathering in single and multiple individual and species experiments.

Variability in feeding of Gammarus pulex: moving towards a more standardised feeding assay

BackgroundFocusing on feeding as an endpoint in ecotoxicological studies is a useful and sensitive tool to detect sub-lethal impacts on individual organisms with relevance to higher levels of organisation (i.e. population and ecosystem levels). We conducted a series of experiments to identify and quantify the influence of parasite infection and food source, food quality, body size and acclimation time prior to testing on the feeding rate of individual Gammarus pulex. Our aim was to assess the variability in feeding rate associated with these factors to support design of feeding assays with individual organisms at a daily resolution.ResultsOverall, feeding rates varied enormously across experiments, and all factors were confirmed to have a significant impact on feeding rates. Reducing the intra-specific variability by using a particular sub-group within each tested factor (except acclimation time) was found to be indispensable for a successful feeding assay with individual organisms. Focusing on organisms of a sub-group in terms of parasite infection and body mass resulted in a reduction in intra-specific variability of up to 50% and 57%, respectively. Using a food source of particular quality reduced the variability by up to 38%.ConclusionsWe presented a list of factors that naturally have an impact on feeding rates of Gammarus, quantified their impact on the variability in feeding rates, discussed their importance for consideration when planning a feeding assay and suggested some additional measurements alongside the feeding assay to improve data comparison between studies.

Evidence for links between feeding inhibition, population characteristics, and sensitivity to acute toxicity for Daphnia magna.

A population experiment with Daphnia magna tested the hypothesis that short-term feeding inhibition provokes a shift in population structure that will vary with conspecific pressure (e.g., pressure occurring from individuals of the same species due to competition for food and space) and increases population sensitivity to a xenobiotic exposure due to size-dependent toxicity (e.g., decreasing sensitivity with increasing body length). Populations were exposed for one week to a feeding inhibitor (imidacloprid, 0.15 or 12.0 mg/L) followed by one week of recovery and one day of exposure to an acute toxin (carbaryl, 0.0098 mg/L). Identical exposure under low and high conspecific pressure was studied by delaying the start of exposure for half of the populations by two weeks; thus populations were in a different stage of population development when exposure occurred. Feeding inhibition of 97% (12.0 mg/L imidacloprid) caused a shift in population structure toward smaller individuals but also reduced population abundance by up to 56 ± 7% with a strong influence of conspecific pressure. Increased population sensitivity to carbaryl was observed after feeding inhibition of 97% as hypothesized. Carbaryl exposure for one day resulted in population decline of up to 23 ± 6% when populations were not previously exposed to imidacloprid. Identical carbaryl exposure provoked a four times stronger decline in population abundance when exposure occurred following feeding inhibition of 97%. In conflict with the hypothesis, this was at least in part due to changes in the reproductive strategy of daphnids following exposure to imidacloprid rather than driven by the shift in population structure. The differences in population sensitivity to additional stress (carbaryl) occurring one week after feeding inhibition caused by exposure to imidacloprid adds a further challenge to understanding potential impacts from multiple stressors as occurring in the field at the population level.

Introducing the 2-DROPS model for two-dimensional simulation of crop roots and pesticide within the soil-root zone

Mathematical models of pesticide fate and behaviour in soils have been developed over the last 30years. Most models simulate fate of pesticides in a 1-dimensional system successfully, supporting a range of applications where the prediction target is either bulk residues in soil or receiving compartments outside of the soil zone. Nevertheless, it has been argued that the 1-dimensional approach is limiting the application of knowledge on pesticide fate under specific pesticide placement strategies, such as seed, furrow and band applications to control pests and weeds. We report a new model (2-DROPS; 2-Dimensional ROots and Pesticide Simulation) parameterised for maize and we present simulations investigating the impact of pesticide properties (thiamethoxam, chlorpyrifos, clothianidin and tefluthrin), pesticide placement strategies (seed treatment, furrow, band and broadcast applications), and soil properties (two silty clay loam and two loam top soils with either silty clay loam, silt loam, sandy loam or unconsolidated bedrock in the lower horizons) on microscale pesticide distribution in the soil profile. 2-DROPS is to our knowledge the first model that simulates temporally- and spatially-explicit water and pesticide transport in the soil profile under the influence of explicit and stochastic development of root segments. This allows the model to describe microscale movement of pesticide in relation to root segments, and constitutes an important addition relative to existing models. The example runs demonstrate that the pesticide moves locally towards root segments due to water extraction for plant transpiration, that the water holding capacity of the top soil determines pesticide transport towards the soil surface in response to soil evaporation, and that the soil type influences the pesticide distribution zone in all directions. 2-DROPS offers more detailed information on microscale root and pesticide appearance compared to existing models and provides the possibility to investigate strategies targeting control of pests at the root/soil interface.

Assessment of acute toxicity tests and rhizotron experiments to characterize lethal and sublethal control of soil-based pests: Acute toxicity tests and rhizotrons

BACKGROUND Characterizing lethal and sublethal control of soil-based pests with plant protection products is particularly challenging due to the complex and dynamic interplay of the system components. Here, we present two types of studies: acute toxcity experiments (homogenous exposure of individuals in soil) and rhizotron experiments (heterogeneous exposure of individuals in soil) to investigate their ability to strengthen our understanding of mechanisms driving the effectivness of the plant protection product. Experiments were conducted using larvae of the western corn rootworm Diabrotica virgifera LeConte and three pesticide active ingredients: clothianidin (neonicotinoid), chlorpyrifos (organophosphate) and tefluthrin (pyrethroid). RESULTS The order of compound concentrations needed to invoke a specific effect intensity (EC50 values) within the acute toxicity tests was chlorpyrifos > tefluthrin > clothianidin. This order changed for the rhizotron experiments because application type, fate and transport of the compounds in the soil profile, and sublethal effects on larvae also influence their effectiveness in controlling larval feeding on corn roots. CONCLUSION Beyond the pure measurement of efficacy through observing relative changes in plant injury to control plants, the tests generate mechanistic understanding for drivers of efficacy apart from acute toxicity. The experiments have the potential to enhance efficacy testing and product development, and might be useful tools for assessing resistance development in the future.

A standardized tritrophic small‐scale system (TriCosm) for the assessment of stressor‐induced effects on aquatic community dynamics

Chemical impacts on the environment are routinely assessed in single-species tests. They are employed to measure direct effects on nontarget organisms, but indirect effects on ecological interactions can only be detected in multispecies tests. Micro- and mesocosms are more complex and environmentally realistic, yet they are less frequently used for environmental risk assessment because resource demand is high, whereas repeatability and statistical power are often low. Test systems fulfilling regulatory needs (i.e., standardization, repeatability, and replication) and the assessment of impacts on species interactions and indirect effects are lacking. In the present study we describe the development of the TriCosm, a repeatable aquatic multispecies test with 3 trophic levels and increased statistical power. High repeatability of community dynamics of 3 interacting aquatic populations (algae, Ceriodaphnia, and Hydra) was found with an average coefficient of variation of 19.5% and the ability to determine small effect sizes. The TriCosm combines benefits of both single-species tests (fulfillment of regulatory requirements) and complex multispecies tests (ecological relevance) and can be used, for instance, at an intermediate tier in environmental risk assessment. Furthermore, comparatively quickly generated population and community toxicity data can be useful for the development and testing of mechanistic effect models. Environ Toxicol Chem 2018;37:1051-1060.