Chemical and bioanalytical characterisation of PAHs in risk assessment of remediated PAH-contaminated soils

Polycyclic aromatic hydrocarbons (PAHs) are common contaminants in soil at former industrial areas; and in Sweden, some of the most contaminated sites are being remediated. Generic guideline values for soil use after so-called successful remediation actions of PAH-contaminated soil are based on the 16 EPA priority pollutants, which only constitute a small part of the complex cocktail of toxicants in many contaminated soils. The aim of the study was to elucidate if the actual toxicological risks of soil samples from successful remediation projects could be reflected by chemical determination of these PAHs. We compared chemical analysis (GC-MS) and bioassay analysis (H4IIE-luc) of a number of remediated PAH-contaminated soils. The H4IIE-luc bioassay is an aryl hydrocarbon (Ah) receptor-based assay that detects compounds that activate the Ah receptor, one important mechanism for PAH toxicity. Comparison of the results showed that the bioassay-determined toxicity in the remediated soil samples could only be explained to a minor extent by the concentrations of the 16 priority PAHs. The current risk assessment method for PAH-contaminated soil in use in Sweden along with other countries, based on chemical analysis of selected PAHs, is missing toxicologically relevant PAHs and other similar substances. It is therefore reasonable to include bioassays in risk assessment and in the classification of remediated PAH-contaminated soils. This could minimise environmental and human health risks and enable greater safety in subsequent reuse of remediated soils.     

Ecorisk evaluation and treatability potential of soils contaminated with petroleum hydrocarbon-based fuels

We used a series of toxicity tests to monitor oil degradation in the Kuwaiti oil lakes. Three soils from different locations with a history of hydrocarbon contamination were treated in bench-scale microcosms with controlled nutrient amendments, moisture content, and temperature that had promoted mineralization of total hydrocarbon and oil and grease in a preliminary study. Two hundred days of bioremediation treatment lowered hydrocarbon concentration to below 2 and 5 mg g(-1) for soils A and B, respectively, while in soil C hydrocarbon concentration remained at 12 mg g(-1). Although 85% of the total petroleum hydrocarbons (TPHs) in soil A were reduced 50d after treatment, results of the seed germination and Microtox tests suggested an initial increase in toxicity, indicating that toxic intermediary metabolites may have formed during biodegradation. Also, the significant decrease of TPHs and corresponding high toxicity levels were noted in soil B 200d after bioremediation. Clearly, toxicity values, and not just hydrocarbon concentration, are a key factor in assessing the effectiveness of bioremediation techniques. Field chemistry data showed a significant reduction in hydrocarbon levels after the biological treatment. We concluded that the toxicity assessment of the contaminated soil with a battery of toxicity bioassays could provide meaningful information regarding a characterization procedure in ecological risk assessment.     

Relationship between hydrocarbon measurements and toxicity to a chironomid, fish larva and daphnid for oils and oil spill chemical treatments in laboratory freshwater marsh microcosms

This research investigated the extent to which various common hydrocarbon measures can be used to predict toxicity to freshwater aquatic organisms due to fouling by oil. Actual toxicity results, on laboratory freshwater marsh microcosms using two water-column species and a benthic species, were described earlier. The hydrocarbon measures used were TPH(g), TPH(FID), TPH(MS), TTAH (sum of 41 target aromatic hydrocarbons), principal components of 41 TAHs, and each individual TAH. In general, toxicity was more closely related to TPH(MS) levels than to TPH(FID) and (especially) TPH(g) levels. The strongest relationships were found for TTAH levels and for the principal components of the TAHs. Regressions of toxicity on many individual TAHs were also strong, with a single group of compounds explaining as much as 59% of the variation in survival. While the various regressions were highly significant statistically and at times able to accurately predict broad differences in toxicity, the high variation in survival at a specific hydrocarbon concentration indicates that these hydrocarbon measures can not substitute for actual toxicity determinations in accurately ranking the toxicity of samples from oiled freshwater marshes.     

Predicting the sensitivity of fishes to dioxin-like compounds: possible role of the aryl hydrocarbon receptor (AhR) ligand binding domain

Dioxin-like compounds are chronically toxic to most vertebrates. However, dramatic differences in sensitivity to these chemicals exist both within and among vertebrate classes. A recent study found that in birds, critical amino acid residues in the aryl hydrocarbon receptor (AhR) ligand binding domain are predictive of sensitivity to dioxin-like compounds in a range of species. It is currently unclear whether similar predictive relationships exist for fishes, a group of animals at risk of exposure to dioxin-like compounds. Effects of dioxin-like compounds are mediated through the AhR in fishes and birds. However, AhR dynamics are more complex among fishes. Fishes possess AhRs that can be grouped within at least three distinct clades (AhR1, AhR2, AhR3) with each clade possibly containing multiple isoforms. AhR2 has been shown to be the active form in most teleosts, with AhR1 not binding dioxin-like compounds. The role of AhR3 in dioxin-like toxicity has not been established to date and this clade is only known to be expressed in some cartilaginous fishes. Furthermore, multiple mechanisms of sensitivity to dioxin-like compounds that are not relevant in birds could exist among fishes. Although, at this time, deficiencies exist for the development of such a predictive relationship for application to fishes, successfully establishing such relationships would offer a substantial improvement in assessment of risks of dioxin-like compounds for this class of vertebrates. Elucidation of such relationships would provide a mechanistic foundation for extrapolation among species to allow the identification of the most sensitive fishes, with the ultimate goal of the prediction of risk posed to endangered species that are not easily studied.     

Mixed-function oxidase-specific activity in wild and caged speckled sanddabs Citharichthys stigmaeus in Elkhorn Slough, Moss Landing Harbor and nearshore Monterey Bay, California

The goal of this study was to characterize the spatial, seasonal and annual hepatic activities of mixed-function oxidase (MFO) in the speckled sanddab Citharichthys stigmaeus, the most common fish in the Moss Landing area. In addition, techniques to monitor MFO activities in caged speckled sanddabs were developed and tested. Once the relationship between MFO activities in caged and wild fish populations is determined, caged fish could be used to monitor potential hydrocarbon impacts at Moss Landing, or other marine sites. During each of the spatial, seasonal and annual sediment samplings conducted in 1985-1987 as part of a separate hydrocarbon variability study at Moss Landing, 12 wild speckled sanddabs were collected from Moss Landing Harbor, Elkhorn Slough and nearshore Monterey Bay sites. In addition, four locations were chosen for a 14-day field caged fish experiment. The caged fish experiments successfully demonstrated the feasibility of using caged sanddabs as indicators of hydrocarbon exposure. The major source of variability in hepatic aryl hydrocarbon hydroxylase activity in wild speckled sanddabs from the Moss Landing area is due to seasonal rather than site differences. Significant relationships between caged fish MFO response and sediment hydrocarbon concentrations were found. It is possible that caged fish could be used in place of costly sediment sampling and analysis, and provide a more direct method to assess biological impacts. Many of the caging techniques demonstrated in this study can easily be transferred to other benthic flatfish, and other marine and freshwater environments.     

The aryl receptor inhibitor epigallocatechin-3-gallate protects INS-1E beta-cell line against acute dioxin toxicity

The aim of this research was to investigate the mechanism(s) underlying the acute toxicity of dioxin in pancreatic beta cells and to evaluate the protective effects of epigallocatechin-3-gallate (EGCG), the most abundant of the green tea’s catechins and a powerful inhibitor of the aryl hydrocarbon receptor (AhR). Using the insulin-secreting INS-1E cell line we have explored the effect of 1 h exposure to different concentrations of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), alone or in the presence of EGCG, on: (a) cell survival; (b) cellular ultrastructure; (c) intracellular calcium levels; (d) mitochondrial membrane potential; (e) glucose-stimulated insulin secretion and (f) activation of MAP kinases. Our results demonstrate that TCDD is highly toxic for INS-1E cells, suggesting that pancreatic beta cells should be considered a relevant and sensitive target for dioxin acute toxicity. EGCG significantly protects INS-1E cells against TCDD-induced toxicity in terms of both cell survival and preservation of cellular ultrastructure. The mechanism of this protective effect seems to be related to: (a) the ability of EGCG to preserve the mitochondrial function and thus to prevent the TCDD-induced inhibition of glucose-stimulated insulin secretion and (b) the ability of EGCG to inhibit the TCDD-induced activation of selected kinases, such as e.g. ERK 1/2 and JNK. Our results clearly show that EGCG is able to protect pancreatic beta cells against dioxin acute toxicity and indicate the mitochondrion as the most likely target for this beneficial effect.     

Assessment of the toxic potential of hydrocarbon containing sludges

A short-time period microbial toxicity test-battery was used for the investigation of acute toxicity and genotoxicity of five hydrocarbon containing sludges. Four sludges were obtained from a petrochemical industry and the fifth from a petroleum refinery. Some of the sludges had been stored for long periods. Bioremediation potential assays for soils polluted with each of the sludges were also considered. The sludges did not show acute toxicity in any of the microbial tests performed. However, when the diethylether soluble fractions of these sludges were analyzed some of them showed acute toxicity, for which the clearest results were obtained with the resazurin reduction method. The greatest toxicity detected with the Resazurin based method was found in the diethylether extracts of the freshly collected (not stored) sludges. On the other hand, the diethylether soluble fraction of those sludges that had been stored showed genotoxicity when analyzed with the Salmonella/microsome assay. After the incorporation of the sludges into the soil, increased bacterial counts were noted and substantial hydrocarbon elimination was achieved in 30 days, showing that bioremediation may be a possible technology for cleaning soils polluted with these sludges.     

Polychlorinated naphthalenes: an environmental update

Polychlorinated naphthalenes (PCNs; CNs) form a complex mixture of up to 75 congeners containing from one to eight chlorine atoms per naphthalene molecule. Chloronaphthalenes are widespread global environmental pollutants which accumulate in biota. All chloronaphthalenes are planar compounds and can contribute to the aryl hydrocarbon (Ah) receptor-mediated mechanism of toxicity with a combination of various 2,3,7,8-tetrachlorodibenzo-p-dioxin like toxic responses. There are three known main sources of environmental pollution with PCNs: technical PCN formulations, technical polychlorinated biphenyl formulations, and thermal and other processes in the presence of chlorine. The purpose of this paper is to briefly review the most recent data on environmental pollution, chemistry, analysis, sources, formation, persistence, toxicity and behavior of PCNs.     

A study of 2,4,6-trinitrotoluene inhibition of benzo[a]pyrene uptake and activation in a microbial mutagenicity assay

A number of in vitro and in vivo studies have determined that binary and complex mixtures may interact to produce a toxicity that could not be predicted based on the individual chemicals. The present study was conducted with a binary mixture of model compounds to investigate possible interactions affecting their mutagenicity. The compounds included Benzo[a]pyrene (BAP), a polycyclic aromatic hydrocarbon that is an indirect-acting mutagen of great environmental concern, and 2,4,6-Trinitrotoluene (TNT), a nitro-aromatic compound that is a direct-acting mutagen frequently found as a soil contaminant at munitions sites. This study indicated that a binary mixture of BAP and TNT failed to induce the positive mutagenic response in Salmonella typhimurium strain TA98 characteristic of either compound alone. Spectrofluorometric analysis of BAP, and kinetic analyses of ³HBAP uptake in the presence or absence of TNT using TA98 cells that were treated or untreated with activated rat liver microsomes were performed. In cells preloaded with BAP, cellular BAP fluorescence was rapidly suppressed in the presence of TNT. Mass spectroscopy of BAP and TNT mixtures revealed a number of products, believed to be the result of complexation and nitration, that may account for the antagonistic action of TNT on BAP-induced mutagenicity in TA98 cells. Further, kinetic studies indicated that TNT inhibited the incorporation of BAP into cells.     

The effect of EDTA and fulvic acid on Cd, Zn, and Cu toxicity to a bioluminescent construct (pUCD607) of Escherichia coli

The hypothesis, that metal toxicity is dominated by free ion activity, was tested by comparing calculated metal activities with measured toxic responses to a genetically modified, luminescent bacterium, Escherichia coli. The toxicity of Cd, Cu, and Zn sulphate salts in the presence of EDTA and fulvic acid in well-defined solutions was measured. Good agreement between free metal activity and toxicity was found for Cu but not for Zn and Cd. The toxicity relationships were altered by glucose addition to the organism. Stable chloride complexes may have contributed to the toxicity of Cd under the test conditions. The results suggest that there is not always a simple relationship between toxicity and free-ion metal concentration and that further account should be taken of competitive interactions between living cells and ligands and the physiological status of the organism.     

A technical mixture of 2,2',4,4'-tetrabromo diphenyl ether (BDE47) and brominated furans triggers aryl hydrocarbon receptor (AhR) mediated gene expression and toxicity

Polybrominated diphenyl ethers (PBDE) are found as ubiquitous contaminants in the environment, e.g., in sediments and biota as well as in human blood samples and mother's milk. PBDEs are neuro- and developmental toxins, disturb the endocrine system and some are even carcinogenic. Structural similarities of PBDEs with dioxin-like compounds, e.g., 2,3,7,8-tetrachloro-dibenzodioxin (TCDD), have raised concern about a possible "dioxin-like" action of PBDEs. TCDD exerts its toxicity via binding to and activation of the aryl hydrocarbon receptor (AhR). AhR ligands are in contrast to PBDEs usually coplanar compounds. Thus, PBDEs are not likely to be strong AhR agonists. The aim of this study was to analyze the effects of the most abundant PBDE congener, 2,2',4,4'-tetrabromo diphenyl ether (BDE47), on AhR activity and signaling. Initially, we measured cytochrome P450 1A1 (Cyp1A1) induction as a readout for AhR activation by BDE47. Low grade purified BDE47 increased CYP1A1 levels in transformed and primary rat hepatocytes and human hepatoma cells. Chemical analysis of the BDE47 sample identified trace contaminations with brominated furans such as 2,3,7,8-tetrabromo dibenzodioxin (TBDF), which most likely were responsible for the observed activation of AhR. Subsequently, the BDE47 mixture was studied for its effect on AhR mediated toxicity and global gene expression. Indeed, in rat hepatoma cells and in zebrafish embryos the BDE47 mixture provoked changes in gene expression and toxicity similar to known AhR agonists. In addition to the dioxin-like actions, the BDE47 sample enhanced Cyp2B and Cyp3A expression suggesting that commercial PBDE mixtures, which also often contain brominated furans, may disturb cellular homeostasis at multiple levels.     

Current state of yusho and prospects for therapeutic strategies

The mass food poisoning incident yusho occurred in Japan in 1968. It was caused by the ingestion of rice bran oil contaminated with polychlorinated biphenyls and various dioxins and dioxin-like compounds including polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDFs). Notably, PCDFs were found to contribute to approximately 65% of the total toxicity equivalent in the blood of yusho patients. Lipophilic dioxins are retained in the body for a longer period than previously estimated. Victims suffered from characteristic skin manifestations associated with non-specific systemic symptoms, neurological symptoms, and respiratory symptoms. The severe symptoms seen in the initial phase subsequently faded, but recently, improvements have scarcely been observed. The Yusho Group has been researching treatments for this condition. Several clinical trials with chelating agents or dietary fibers aimed at accelerating the excretion of compounds. While some treatments increased dioxin excretion, none provided satisfactory symptom relief. Concurrently, various phytochemicals and herbal extracts have been found to possess biological activities that suppress dioxin-induced toxicity via aryl hydrocarbon receptor or activate the antioxidant nuclear factor-erythroid 2-related factor-2 (NRF2) signal pathway, making them promising therapeutic candidates. Here, we summarize the current status of yusho and findings of clinical trials for yusho patients and discuss the treatment prospects.

     

In vitro AHH induction by polychlorinated biphenyl and dibenzofuran mixtures: Additive effects

The activities of several individual polychlorinated biphenyls (PCBs) and dibenzofurans (PCDFs) and several environmentally significant reconstituted mixtures of these compounds as inducers of aryl hydrocarbon hydroxylase (AHH) and ethoxyresorufin O-deethylase (EROD) in rat hepatoma H-4-II E cells were determined. The observed AHH and EROD induction EC₅₀S for the mixtures were compared with the calculated values, which were based on the summation of the relative per cent contributions of the individual components of the reconstituted PCB and PCDF mixtures. The results show that the differences between the observed and calculated EC₅₀s for these mixtures were minimal or not significant and the data supports the use of the rat hepatoma H-4-II E cell system as a bioassay for toxic halogenated aryl hydrocarbons.     

水体沉积物毒性鉴别与评价研究进展

综述了水体沉积物毒性鉴别与评价的主流方法以及研究进展,指出毒性测试和致毒因子鉴别方法是限制水体沉积物污染生态风险评价的关键因素,认为发展和应用以生物标记物和生物效应为导向,尤其是各种具有污染专一性指示作用的生物效应标志测试方法的建立和应用,并结合具有选择性的样品分级技术和先进的仪器进行定性定量分析,将是以生物效应为导向、以化学分析为基础的复杂水体沉积物毒性鉴别与评价的重要发展方向.     

Effect of inducers of P-450 cytochrome isoenzymes on TCDD immunosuppressive activity

Drugs inducing different forms of P-450 cytochrome isoenzymes and binding the Ah receptor or not were investigated for their ability to modify 2,3,7,8, tetrachlorodibenzo-p-dioxin (TCDD) immunotoxicity in mice. 3-Methyl-cholanthrene (3MC) and β-naphthoflavone (BNF) administered to TCDD-treated mice caused additive inhibition of humoral antibody production and of responsiveness to concanavallin A (ConA) but not to phytohemagglutinin (PHA) and lipopolisaccharide (LPS) while phenobarbital (PB) never modified TCDD immunosuppression. Natural killer (NK) cell activity was not reduced by single drug treatment or by combined treatments. Hepatic aryl hydrocarbon hydroxylase (AHH) induction by TCDD was not significantly modified by PB, 3MC or BNF.     

The evaluation of dioxin and dioxin-like contaminants in selected food samples obtained from the Belgian market: comparison of TEQ measurements obtained through the CALUX bioassay with congener specific chemical analyses

A limited number of different foods were analysed for dioxin-like compounds by the CALUX bioassay which is an in vitro luciferase reporter gene assay measuring chemical activation of the aryl hydrocarbon receptor. Sixty-two milk samples were obtained from a surveillance campaign, 34 meat samples and 34 fishery products were purchased from the Belgian market. Bio-analytical and chemo-analytical dioxin toxicity equivalents (TEQ) values of the same milk samples were compared. Spearman's Rank correlation coefficients of 0.72, 0.67, 0.73 were obtained respectively between CALUX-TEQ and PCDD/F-TEQ, DL-PCB-TEQ and PCDD/F+DL-PCB-TEQ. The bioassay limit of detection was 0.1 pg TEQ from 1 g animal lipid, the limit of quantification was 0.4 pg TEQ. The repeatability of the CALUX bioassay (variability of butter fat samples analysed in the same run) showed a coefficient of variation (CV) of 10%, intra laboratory reproducibility based on independent runs of the same butter fat samples showed more variation (CV of 26% for samples above 2 pg TEQ/g lipid). All milk samples with a chemical TEQ value above the current limit value in Belgium showed an elevated CALUX-TEQ concentration, above 6 pg TEQ/g lipid. No false negative results were obtained. Based on the good correlation between CALUX-TEQ and chemically measured TEQ levels, the CALUX bioassay can be recommended as a screening tool for routine measurement of potentially toxic PHAHs in milk samples. Chemical analyses could then largely be restricted to positive samples, in order to identify the nature and to quantify the concentration of the chemicals that give the positive signal. Meat samples showed lower CALUX-TEQ values per gram lipid compared to fish samples. The fish samples showed a wider range of CALUX-TEQ values than the meat samples.     

Ecotoxic Effect of Photocatalytic Active Nanoparticles (TiO2) on Algae and Daphnids (8 pp)

Background, Aim and Scope Due to their large potential for manifold applications, the use of nanoparticles is of increasing importance. As large amounts of nanoparticles may reach the environment voluntarily or by accident, attention should be paid on the potential impacts on the environment. First studies on potential environmental effects of photocatalytic TiO2 nanoparticles have been performed on the basis of widely accepted, standardized test systems which originally had been developed for the characterization of chemicals. The methods were adapted to the special requirements of testing photocatalytic nanoparticles. Materials and Methods: Suspensions of two different nanoparticles were illuminated to induce their photocatalytic activity. For testing, the growth inhibition test with the green alga Desmodesmus subspicatus and the immobilization test with the daphnid Daphnia magna were selected and performed following the relevant guidelines (algae: ISO 8692, OECD 201, DIN 38412-33; daphnids: ISO 6341, OECD 202, DIN 38412-30). The guidelines were adapted to meet the special requirements for testing photocatalytic nanoparticles. Results: The results indicate that it is principally possible to determine the ecotoxicity of nanoparticles. It was shown that nanoparticles may have ecotoxicological effects which depend on the nature of the particles. Both products tested differ in their toxicity. Product 1 shows a clear concentration-effect curve in the test with algae (EC50: 44 mg/L). It could be proven that the observed toxicity was not caused by accompanying contaminants, since the toxic effect was comparable for the cleaned and the commercially available product. For product 2, no toxic effects were determined (maximum concentration: 50 mg/L). In the tests with daphnids, toxicity was observed for both products, although the concentration effect-curves were less pronounced. The two products differed in their toxicity; moreover, there was a difference in the toxicity of illuminated and non-illuminated products. Discussion: Both products differ in size and crystalline form, so that these parameters are assumed to contribute to the different toxicities. The concentration-effect curves for daphnids, which are less-pronounced than the curves obtained for algae, may be due to the different test organisms and/or the differing test designs. The increased toxicity of pre-illuminated particles in the tests with daphnids demonstrates that the photocatalytic activity of nanoparticles lasts for a period of time. Conclusions: The following conclusions can be drawn from the test results: (I) It is principally possible to determine the ecotoxicity of (photocatalytic) nanoparticles. Therefore, they can be assessed using methods comparable to the procedures applied for assessing soluble chemicals. - (II) Nanoparticles may exert ecotoxicological effects, which depend on the specific nanoparticle. - (III) Comparable to traditional chemicals, the ecotoxicity depends on the test organisms and their physiology. - (IV) The photocatalytic activity of nanoparticles lasts for a relevant period of time. Therefore, pre-illumination may be sufficient to detect a photocatalytic activity even by using test organisms which are not suitable for application in the pre-illumination-phase. Recommendations and Perspectives: First results are presented which indicate that the topic 'ecotoxicity and environmental effects of nanoparticles' should not be neglected. In testing photocatalytic nanoparticles, there are still many topics that need clarification or improvement, such as the cause for an observed toxicity, the improvement of the test design, the elaboration of a test battery and an assessment strategy. On the basis of optimized test systems, it will be possible to test nanoparticles systematically. If a potential risk by specific photocatalytic particles is known, a risk-benefit analysis can be performed and, if required, risk reducing measures can be taken.     

Recombinant human AhR-mediated GUS reporter gene assays for PCB congeners in transgenic tobacco plants in comparison with recombinant mouse and guinea pig AhRs

Four expression plasmids for recombinant human aryl hydrocarbon receptor (hAhR) consisting of a ligand binding domain of hAhR, a DNA-binding domain of LexA and a transactivation domain of VP16 as well as β-glucuronidase (GUS) reporter genes were constructed. All the expression plasmids were transformed into tobacco plants. The selected transgenic tobacco plants were used to assay. PCB congeners showed GUS activity in a TEF-dependent manner. The selected transgenic tobacco plant XhD4V17 was compared with the transgenic tobacco plants XmD4V26 and XgD2V23 containing recombinant mouse (m) AhR-mediated GUS reporter gene expression cassette and recombinant guinea pig (g) AhR-mediated GUS reporter gene expression cassette for PCB congener-inducible GUS activity. The data revealed that the tobacco plant XgD2V23 was the most active in PCB congener-inducible GUS activity. In a 1:1 mixture of PCB126 and PCB80 a reduced PCB126-induced GUS activity was observed in plant XgD2V23, which could possibly be due to interaction between PCB126 and PCB80.     

Effect-Directed Analysis of Key Toxicants in European River Basins. A Review (9 pp)

Background, Aim and Scope Extensive monitoring programs on chemical contamination are run in many European river basins. With respect to the implementation of the European Union (EU) Water Framework Directive (WFD), these programs are increasingly accompanied by monitoring the ecological status of the river basins. Assuming an impact of chemical contamination on the ecological status, the assignment of effects in aquatic ecosystems to those stressors that cause the effects is a prerequisite for taking political or technical measures to achieve the goals of the WFD. Thus, one focus of present European research is on toxicant identification in European river basins in order to allow for a reduction of toxic pressure on aquatic ecosystems according to the WFD. Main Features: An overview is presented on studies that were performed to link chemical pollution in European river basins to measurable ecotoxic effects. This includes correlation-based approaches as well as investigations that apply effect-directed analysis (EDA) integrating toxicity testing, fractionation and non-target chemical analysis. Effect-based key toxicants that were identified in European surface waters are compiled and compared to EU priority pollutants. Further needs for research are identified. Results: Studies on the identification of effect-based key toxicants focused on mutagenicity, aryl hydrocarbon receptor-mediated effects, endocrine disruption, green algae, and invertebrates. The identified pollutants include priority pollutants and other well-known environmental pollutants such as polycyclic aromatic hydrocarbons, polychlorinated dibenzo-p-dioxins, furans, and biphenyls, nonylphenol, some pesticides and tributyltin, but also other compounds that were neither considered as environmental pollutants before nor regulated such as substituted phenols, natural or synthetic estrogens and androgens, dinaphthofurans, 2-(2-naphthalenyl)benzothiophene, and N-phenyl-2-naphthylamine. Discussion: Individual studies at specific sites in a European river basin demonstrated the power of combined biological and chemical analytical approaches and, particularly, of effect-directed analysis. However, the available information on effect-based key toxicants is very limited with respect to the entirety of rivers possibly at risk due to chemical contamination and with respect to toxicological endpoints considered at a specific site. A relatively broad basis of information exists only for estrogenicity and aryl hydrocarbon, receptor-mediated effects. Conclusions: The development of tools and strategies for an identification of key toxicants on a broader scale are a challenging task for the next years. Since investigations dealing with toxicant identification are too labor and cost-intensive for monitoring purposes, they have to be focused on the key sites in a river basin. These should include hot spots of contamination, particularly if there is evidence that they might pose a risk for downstream areas, but may also involve accumulation zones in the lower reach of a river in order to get an integrated picture on the contamination of the basin. Perspectives: While EDA is almost exclusively based on measurable effects in in vitro and in vivo biotests to date, an increasing focus in the future should be on the integration of EDA into Ecological Risk Assessment and on the development of tools to confirm EDA-determined key toxicants as stressors in populations, communities and ecosystems. Considering these requirements and applied in a focused way, toxicant identification may significantly help to implement the Water Framework Directive by providing evidence on the main stressors and possible mitigation measures in order to improve the ecological status of a river ecosystem.