Beate I. Escher

The Challenge of Micropollutants in Aquatic Systems

The increasing worldwide contamination of freshwater systems with thousands of industrial and natural chemical compounds is one of the key environmental problems facing humanity. Although most of these compounds are present at low concentrations, many of them raise considerable toxicological concerns, particularly when present as components of complex mixtures. Here we review three scientific challenges in addressing water-quality problems caused by such micropollutants. First, tools to assess the impact of these pollutants on aquatic life and human health must be further developed and refined. Second, cost-effective and appropriate remediation and water-treatment technologies must be explored and implemented. Third, usage and disposal strategies, coupled with the search for environmentally more benign products and processes, should aim to minimize introduction of critical pollutants into the aquatic environment.

Recent advances in environmental risk assessment of transformation products

When micropollutants degrade in the environment, they may form persistent and toxic transformation products, which should be accounted for in the environmental risk assessment of the parent compounds. Transformation products have become a topic of interest not only with regard to their formation in the environment, but also during advanced water treatment processes, where disinfection byproducts can form from benign precursors. In addition, environmental risk assessment of human and veterinary pharmaceuticals requires inclusion of human metabolites as most pharmaceuticals are not excreted into wastewater in their original form, but are extensively metabolized. All three areas have developed their independent approaches to assess the risk associated with transformation product formation including hazard identification, exposure assessment, hazard assessment including dose-response characterization, and risk characterization. This review provides an overview and defines a link among those areas, emphasizing commonalities and encouraging a common approach. We distinguish among approaches to assess transformation products of individual pollutants that are undergoing a particular transformation process, e.g., biotransformation or (photo)oxidation, and approaches with the goal of prioritizing transformation products in terms of their contribution to environmental risk. We classify existing approaches for transformation product assessment in degradation studies as exposure- or effect-driven. In the exposure-driven approach, transformation products are identified and quantified by chemical analysis followed by effect assessment. In the effect-driven approach, a reaction mixture undergoes toxicity testing. If the decrease in toxicity parallels the decrease of parent compound concentration, the transformation products are considered to be irrelevant, and only when toxicity increases or the decrease is not proportional to the parent compound concentration are the TPs identified. For prioritization of transformation products in terms of their contribution to overall environmental risk, we integrate existing research into a coherent model-based, risk-driven framework. In the proposed framework, read-across from data of the parent compound to the transformation products is emphasized, but limitations to this approach are also discussed. Most prominently, we demonstrate how effect data for parent compounds can be used in combination with analysis of toxicophore structures and bioconcentration potential to facilitate transformation product effect assessment.

Comparative analysis of estrogenic activity in sewage treatment plant effluents involving three in vitro assays and chemical analysis of steroids

In this study, we assessed and compared the suitability of three in vitro screening tools for the measurement of estrogenic activity in sewage treatment plant effluents (STPEs). These assays were the yeast estrogen screen (YES), production of zona radiata proteins (ZRPs) in trout hepatocytes, and the induction of reporter gene expression in the transfected rainbow trout gonad cell line RTG-2. Data obtained with the YES were additionally compared with calculated estrogenicity, based on steroid analysis data of the effluents. For comparison purposes, the response of the in vitro systems toward the estrogenic chemicals beta-estradiol, ethinyl estradiol, bisphenol-A, nonylphenol, and octylphenol was assessed. All three assays showed sensitivities in the same order of magnitude in response to the steroid compounds tested, with ZRP production being the least sensitive. Regarding the estrogenic environmental chemicals tested, the RTG-2 assay was more than an order of magnitude more sensitive than the other two assays. Despite their different sensitivities toward selected test chemicals, the three in vitro systems indicated estrogenic activity in the same concentration range for the tested STPEs. Calculated estrogenicity (chemical analysis) and measured estrogenicity (YES) were of the same order of magnitude for the STPEs tested. The present study indicates that all three in vitro systems, with the yeast-based system being the easiest and most robust, are applicable for the screening of estrogenic activity in effluent samples.

Toxic equivalent concentrations (TEQs) for baseline toxicity and specific modes of action as a tool to improve interpretation of ecotoxicity testing of environmental samples.

The toxic equivalency concept is a widely applied method to express the toxicity of complex mixtures of compounds that act via receptor-mediated mechanisms such as induction of the arylhydrocarbon or estrogen receptors. Here we propose to extend this concept to baseline toxicity, using the bioluminescence inhibition test with Vibrio fischeri, and an integrative ecotoxicity endpoint, algal growth rate inhibition. Both bioassays were validated by comparison with literature data and quantitative structure-activity relationships (QSARs) for baseline toxicity were developed for all endpoints. The novel combined algae test, with Pseudokirchneriella subcapitata, allows for the simultaneous evaluation of specific inhibition of photosynthesis and growth rate. The contributions of specific inhibition of photosynthesis and non-specific toxicity could be differentiated by comparing the time and endpoint pattern. Photosynthesis efficiency, measured with the saturation pulse method after 2 h of incubation, served as indicator of specific inhibition of photosynthesis by photosystem II inhibitors. Diuron equivalents were defined as toxicity equivalents for this effect. The endpoint of growth rate over 24 h served to derive baseline toxicity equivalent concentrations (baseline-TEQ). By performing binary mixture experiments with reference compounds and complex environmental samples from a sewage treatment plant and a river, the TEQ concept was validated. The proposed method allows for easier interpretation and communication of effect-based water quality monitoring data and provides a basis for comparative analysis with chemical analytical monitoring.

Mechanistic studies on baseline toxicity and uncoupling of organic compounds as a basis for modeling effective membrane concentrations in aquatic organisms

Abstract. Mechanistic studies on membrane toxicity are reviewed and linked to effects observed in-vivo. Time-resolved spectroscopy on energy-transducing membranes is an in-vitro test method that provides information on the toxicodynamics of membrane toxicity, namely baseline toxicity and uncoupling. Without blurring effects of the toxicokinetic phase, the intrinsic potency of membrane toxicants can directly be determined thus allowing the development of a classification system to distinguish between baseline toxicity and uncoupling. Toxicity data of known baseline toxicants and of substituted phenols from literature are reevaluated in light of the information obtained from the mechanistic test system. Exposure-based effect data (aqueous effect concentrations) of substituted phenols (baseline toxicants and uncouplers) from various aquatic organisms (bacteria, protozoa, algae, daphnids, fish) are compared to the corresponding effect concentrations expected from baseline toxicity. The results of the comparison support the view that the membrane is the primary target site for acute toxicity of substituted phenols in aquatic organisms. The classification into baseline toxicants and uncouplers based upon the criteria derived from the mechanistic test system is correct. Nevertheless, the acute toxicity in-vivo cannot be correlated quantitatively to the mechanistic data presumably because bioaccumulation is not directly proportional to intrinsic toxicity and the metabolic transformation is variable within the test set of substituted phenols. The pH-dependence of acute toxicity is mainly determined by the pH-dependence of bioccumulation, the internal effect concentrations are virtually independent on external pH. The internal effect concentrations, also called lethal body burdens or critical body residues, which are taken from the literature or calculated from aqueous effect concentrations and bioconcentration factors, are related to membrane concentrations using a simple three-compartment equilibrium partitioning model. The modeled membrane concentrations of non-polar and polar narcotics turn out to be statistically indistinguishable, which is consistent with the findings from the mechanistic test system. The toxic ratios, i.e., the excess toxicity of uncouplers in relation to their baseline toxicity, agree for most compounds upon comparison of the mechanistic test system with the aquatic organisms thus confirming that uncoupling is the dominant mode of action responsible for lethality.

Mixture Toxicity Revisited from a Toxicogenomic Perspective

The advent of new genomic techniques has raised expectations that central questions of mixture toxicology such as for mechanisms of low dose interactions can now be answered. This review provides an overview on experimental studies from the past decade that address diagnostic and/or mechanistic questions regarding the combined effects of chemical mixtures using toxicogenomic techniques. From 2002 to 2011, 41 studies were published with a focus on mixture toxicity assessment. Primarily multiplexed quantification of gene transcripts was performed, though metabolomic and proteomic analysis of joint exposures have also been undertaken. It is now standard to explicitly state criteria for selecting concentrations and provide insight into data transformation and statistical treatment with respect to minimizing sources of undue variability. Bioinformatic analysis of toxicogenomic data, by contrast, is still a field with diverse and rapidly evolving tools. The reported combined effect assessments are discussed in the light of established toxicological dose-response and mixture toxicity models. Receptor-based assays seem to be the most advanced toward establishing quantitative relationships between exposure and biological responses. Often transcriptomic responses are discussed based on the presence or absence of signals, where the interpretation may remain ambiguous due to methodological problems. The majority of mixture studies design their studies to compare the recorded mixture outcome against responses for individual components only. This stands in stark contrast to our existing understanding of joint biological activity at the levels of chemical target interactions and apical combined effects. By joining established mixture effect models with toxicokinetic and -dynamic thinking, we suggest a conceptual framework that may help to overcome the current limitation of providing mainly anecdotal evidence on mixture effects. To achieve this we suggest (i) to design studies to establish quantitative relationships between dose and time dependency of responses and (ii) to adopt mixture toxicity models. Moreover, (iii) utilization of novel bioinformatic tools and (iv) stress response concepts could be productive to translate multiple responses into hypotheses on the relationships between general stress and specific toxicity reactions of organisms.

Passive sampling combined with ecotoxicological and chemical analysis of pharmaceuticals and biocides - evaluation of three Chemcatcher configurations.

Passive sampling is a tool to monitor the presence and concentrations of micropollutants in the aquatic environment. We investigated the duration of integrative sampling and the effects of flow rate on the performance of three configurations of the Chemcatcher - a sampler for polar organic compounds. Chemcatchers were fitted with Empore styrenedivinylbenzene (SDB) XC disks (XC), SDB-RPS disks (RPS) or SDB-RPS disks covered with a polyethersulfone membrane (RPS-PES). Samplers were either exposed to treated sewage effluent for 5 days at various flow rates, or at a single flow rate with overlapping exposures of 3-24 days. Chemical analysis focused on a set of pharmaceuticals and biocides and ecotoxicological analysis measured inhibition of photosystem II in algae. For compounds with logK(OW)>2, both XC and RPS disks respond dynamically to higher flow rates; uptake increased up to five-fold when flow increased from 0.03 to 0.37ms(-1). At a flow rate of 0.13ms(-1) the integrative window of SDB disks approached 6 days for more hydrophobic compounds (logK(OW)>3.5). The RPS-PES configuration was less affected by flow and also showed an extended integrative window (up to 24 days). The membrane causes a lag phase of up to 2.3 days which thwarts a sound interpretation of data from sampling periods of less than 10 days.

Rapid exposure assessment of PSII herbicides in surface water using a novel chlorophyll a fluorescence imaging assay

Recently a new Maxi-Imaging-PAM (Max-I-PAM) instrument for phytotoxicity assessment via chlorophyll fluorescence imaging was introduced. This new instrument allows rapid detection of the effects of PS II inhibiting herbicides which are high use agricultural chemicals frequently detected in surface waters in Australia and elsewhere. Several studies have applied the new instrument for detection of phytotoxicants in water using microalgae suspensions; however, these use preliminary protocols and to date no validated method is available for high throughput testing of environmental samples in 96-well plates. Here we developed and applied a new protocol allowing dose-response assessment of four samples within 2 h (8 dilutions in duplicate). The technique was found to be sensitive, with a detection limit of 2.3 ng l(-1) for the herbicide diuron when testing solid phase extracts (SPE) of 1000 ml water samples, and reproducible both between experiments (coefficient of variation (CV)=0.30) and within the 96-well plate (CV=0.06). Relative potencies were determined for four reference PS II impacting herbicides (diuron>hexazinone>atrazine>simazine). Extracts from 1000 ml environmental samples and diuron spiked ultrapure water as well as passive sampler extracts were evaluated and good agreement was found between diuron equivalent concentrations calculated from bioassay results (DEQ(IPAM)) and DEQ(CHEM) values calculated from LCMS chemical analysis of the four reference compounds in the same samples. Overall, the technique provides a valuable bioanalytical tool for rapid and inexpensive effects-based assessment of PS II impacting herbicides in environmental mixtures.

Effect of pulse herbicidal exposure on Scenedesmus vacuolatus: A comparison of two photosystem II inhibitors

Herbicide concentrations fluctuate in rivers following crop application and can reach high levels after rain events, yet the duration of these pulses is short. In the present study, we assessed the effect of atrazine and isoproturon pulse exposure on Scenedesmus vacuolatus (Chlorophyceae; strain 211-8b, Kessler) as well as the recovery in the postexposure period. We further explored whether the time-dependent toxicity is similar for herbicides inhibiting the photosystem II (PSII). The growth rate was assessed for different exposure durations, and in addition the inhibition of the effective quantum yield of PSII was measured to monitor the response at the target site. Atrazine and isoproturon did not have similar time-dependent effects on growth rate, despite their same primary mode of action on PSII. Atrazine was less toxic than isoproturon after 10 h of exposure, but the toxicity of both herbicides was similar after 48 h of exposure. However, both compounds inhibited the PSII effective quantum yield within 1 h following exposure. Similarly, the effective quantum yield recovered completely within 4 h after removal of the toxicants, leading to rapid recovery of algal growth. The rapid onset of effects of isoproturon on the growth of the alga during exposure suggests that a single pulse to this herbicide is likely to induce greater effects than an atrazine pulse at the same concentration, even if these effects are reversible. The information gained in the present study should support the effect assessment of sequential exposures as well as the risk evaluation of fluctuating herbicidal exposure.

Toxicokinetic and Toxicodynamic Modeling Explains Carry-over Toxicity from Exposure to Diazinon by Slow Organism Recovery

Carry-over toxicity occurs when organisms exposed to an environmental toxicant survive but carry some damage resulting in reduced fitness. Upon subsequently encountering another exposure event stronger effects are possible if the organisms have not yet fully recovered. Carry-over toxicity was observed after exposure of the freshwater amphipod Gammarus pulex to repeated pulses of diazinon with varying intervals. Uptake, biotransformation and depuration kinetics were determined. Metabolites were identified and quantified (diazoxon, 2-isopropyl-6-methyl-4-pyrimidinol, one nonidentified metabolite). Parameters of a process-based toxicokinetic-toxicodynamic model were determined by least-squares fitting followed by Markov Chain Monte Carlo parameter estimation. Model parametrization was based on the time-course of measured internal concentrations of diazinon and its metabolite diazoxon in combination with the pulsed toxicity experiment. Prediction intervals, which take the covariation between parameters into account, were calculated for bioaccumulation factors, organism recovery time and simulations of internal concentrations as well as the time-course of survival under variable exposure. Organism recovery time was 28 days (95% prediction interval 25-31 days), indicating the possibility for carry-over toxicity from exposure events several weeks apart. The slow organism recovery and carry-over toxicity was caused by slow toxicodynamic recovery; toxicokinetic processes alone would have resulted in a recovery time of only 1-2 days.