Recent Advances in Environmental Risk Assessment of

CRITICAL REVIEW pubs.acs.org/est

Recent Advances in Environmental Risk Assessment of Transformation Products Beate I. Escher*,† and Kathrin Fenner*,‡,§ †

The University of Queensland, National Research Centre for Environmental Toxicology (Entox), 39 Kessels Road, Brisbane, Qld 4108, Australia ‡ Eawag, Swiss Federal Institute of Aquatic Science and Technology, 8600 D€ubendorf, Switzerland § ETH Zurich, Institute of Biogeochemistry and Pollutant Dynamics, 8092 Zurich, Switzerland ABSTRACT: 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.

’ INTRODUCTION Chemical pollution poses an increasing threat to our environment, amplified by population growth and climate change. Thousands of chemicals, including pesticides, biocides, pharmaceuticals, industrial chemicals, and chemicals from consumer products are present in our wastewater and water from other sources, such as agricultural runoff or stormwater,1 as well as in the terrestrial environment. Compounding this issue, chemicals are also transformed in the environment, and while some of the resulting transformation products (TPs) are known to be more abundant in the aquatic environment than their parent compounds,2,3 the majority of TPs present most likely have r 2011 American Chemical Society

not even been identified yet. Thus, humans and aquatic ecosystems are exposed to a highly variable and unknown cocktail of chemicals. Although individual chemicals are typically present at low concentrations, they can interact with each other resulting in additive or potentially even synergistic mixture effects. The formation and environmental presence of TPs thus adds further complexity to chemical risk assessment. TPs may contribute Received: September 8, 2010 Accepted: March 18, 2011 Revised: March 12, 2011 Published: April 07, 2011 3835

dx.doi.org/10.1021/es1030799 | Environ. Sci. Technol. 2011, 45, 3835–3847

Environmental Science & Technology significantly to the risk posed by the parent compound (a) if they are formed with a high yield; (b) if they are more persistent or more mobile than the parent compounds; or (c) if they have a high toxicity. TPs cannot be regarded in isolation from their parent compounds as they often exhibit the same mode of toxic action and act concentration-additive in mixtures,4,5 meaning that the effects from TPs and parent compounds must conservatively be considered additive, but synergistic effects could even enhance overall toxicity. The need to include stable and/or toxic TPs in risk assessment is mentioned in all relevant regulatory assessment schemes. However, these regulations differ widely with respect to how much concrete guidance and tools they provide to achieve this goal. The Pesticide Directive,6 for instance, gives concise definitions of TPs that are relevant in the context of registration of the active substance: “Relevant metabolites, defined as metabolites for which there is reason to assume that they pose a risk similar to or higher than the parent pesticide, are treated like the parent active substance in the assessment”.6 It further gives guidance on environmental fate studies and modeling tools to be used to predict exposure to TPs,7 and defines the procedure for generating (eco)toxicological data for selected TPs to evaluate the risk associated with their formation.8,9 In contrast, regulatory assessment schemes for other compound classes such as industrial chemicals or pharmaceuticals do not provide guidance on any of these elements. For example, according to the European chemicals regulation REACH10 “consideration should be given to whether the substance being assessed can be degraded to give stable and/or toxic degradation products. Where such degradation products can occur, the assessment should give due consideration to the properties (including toxic effects and bioaccumulation potential) of the products that might arise”.11 The VICH guidelines for veterinary medicines12 and the EMEA guideline for human medicines13 address the issue of TPs only insofar as simulation-type degradation studies are to be carried out for the relevant environmental compartments at higher tiers of the assessment, which typically include identification of major TPs. Here, we review different approaches to identify TPs that contribute significantly to the overall environmental risk posed by the use and release of a chemical. We distinguish between exposure- and effect-driven approaches for the identification of relevant TPs at the level of simulation studies, and introduce a model-based framework to prioritize TPs for more detailed investigations in the context of chemical risk assessment and water quality assessment. The review focuses on environmental risk assessment and specifically on the aquatic environment due to a wider availability of literature references, but the general principles should also be applicable to other environmental compartments and to human health end points.

’ TRANSFORMATION PRODUCTS IN THE ENVIRONMENT Transformation products occurring in the environment can be classified into three categories:14 Metabolites of organic compounds formed during phase I and phase II metabolism.

This group includes mammalian and human metabolites of pharmaceuticals15 and metabolites of pro-pesticides, which are converted to the active substance through initial metabolic

CRITICAL REVIEW

reactions in the target organisms.16 Whereas phase I metabolites are typically formed by oxidation processes, phase II metabolism encompasses conjugation reactions. The metabolites of pharmaceuticals are introduced into the environment through excretion into wastewater, surface waters or onto soils, or indirectly through fertilization of soils with manure or digested sludge. Some phase I metabolites are not only more pharmacologically active than the administered pharmaceutical itself, but recently some have also been shown to exhibit ecotoxicological effects similar to the parent pharmaceutical (e.g., propranolol,17 fluoxetine,4 oseltamivir18). Metabolites might be stable but can also be further transformed in the environment to yield stable TPs. Phase II metabolites, especially glucoronide- and sulfate-conjugates, have been observed to be readily deconjugated during biological wastewater treatment to reform the active substance or its corresponding phase I metabolite.19,20 Similar metabolic processes are relevant for pro-pesticides.16 Prominent examples are organophosphate insecticides containing a phosphorothioate or phosphorodithioate moiety. These are oxidized by cytochrome P450 monoxygenases to yield the oxon metabolites, which are more potent inhibitors of acetylcholinesterases than their parent.21 Related examples were recently reviewed in ref 16. Transformation products and process byproducts formed during advanced water treatment processes. Oxidation processes such as chlorination, chloramination, ozonation, and advanced oxidation by UV/H2O2 treatment are the major processes used in advanced water treatment for disinfection and removal of micropollutants. The disappearance of parent compounds due to these chemical reactions is often considered as indication of risk reduction.22 On the other hand, these reactions are known to form disinfection byproducts from harmless inorganic precursors and natural organic matter (e.g., formation of bromate through ozonation23 and formation of halogenated disinfection byproduct through chlorination24). There is further evidence that under the relatively mild oxidizing conditions often used in water treatment, numerous micropollutant TPs are formed that may have adverse effects on aquatic life and human health, or may be persisting even after the parent compound has been fully degraded.25 In addition, so-called “non-relevant metabolites” of pesticides, i.e., those exhibiting no pesticidal activity or genotoxicity, could reach drinking water due their mobility and persistence and also act as precursors for disinfection byproducts.26 Transformation products from transformation reactions occurring in the environment and engineered systems such as microbial degradation, redox reactions, hydrolysis, or photolysis. Whether and what kind of TPs are formed in the environment from a given chemical depends on the precursor’s chemical structure, but also its distribution among different environmental compartments and the prevailing environmental conditions (for an overview see ref 27). Oxidative processes typically lead to TPs that are more polar and consequently more mobile and less toxic than the parent compounds.2 Conversely, plenty of evidence exists for transformation of molecules through different types of chemical reactions that yield toxicologically relevant TPs. Examples include the propesticides discussed above,16 the photochemical condensation 3836

dx.doi.org/10.1021/es1030799 |Environ. Sci. Technol. 2011, 45, 3835–3847

Environmental Science & Technology

CRITICAL REVIEW

Figure 1. (I) Exposure- and (II) effect-driven approaches for the identification of relevant transformation products in process-based simulation studies. TP: transformation product, P: parent compound.

of triclosan and its chlorinated derivatives to dioxin-like compounds28 and its microbial transformation to the more bioaccumulative methyl-triclosan,29 or the microbially mediated oxidative cleavage of phenoxy herbicides to yield substituted phenols that act as uncouplers of energytransduction.30 More recently, evidence is growing that microbial transformation to form stable TPs is not only relevant in environmental compartments such as soils or sediments, but also in engineered systems such as activated sludge.31,32

’ IDENTIFICATION OF RELEVANT TRANSFORMATION PRODUCTS IN PROCESS-BASED SIMULATION STUDIES The usual approach to single out TPs formed in simulation studies is exposure-driven. It is based on the identification of TPs followed by their synthesis and effect assessment (Figure 1, I). An alternative approach is to screen the reaction mixture of a chemical undergoing a transformation process for mixture toxicity, and to relate the change in toxicity to the decrease in parent compound concentration (Figure 1, II). Exposure-Driven Approach. In the exposure-driven approach (Figure 1, I), TPs formed during simulation studies are isolated and identified, followed by environmental fate assessment and/or (eco)toxicity testing. This approach is typically recommended as the default approach in regulatory risk assessment frameworks that require the identification of risk-relevant TPs. The European Pesticide and Biocide Directives,6,33 for instance, refer to the OECD testing guidelines34 for simulation-type degradation studies in soil, sediment-water, surface water, or activated sludge systems to identify major TPs. These are defined as TPs formed in g10% of the initial amount of parent compound, or TPs whose concentrations are continuously increasing during the study. For an initial assessment of whether major TPs are formed at all, these guidelines propose the use of radiolabeled parent compounds and separation of the reaction mixture, or extracts from different phases of the reaction mixture, by HPLC followed by radioactivity detection. Subsequent structural identification of

Figure 2. Details of the effect-driven approach to assess TPs.

TPs requires further analyses. If likely transformation pathways are known, direct chemical or biological synthesis of plausible TPs can be used in combination with analytical techniques such as GC-MS or LC-MS for an unequivocal structural identification of the TPs. Knowledge on transformation pathways can be obtained through read-across from structurally related compounds, e.g., within one class of active substances, or from 3837



CRITICAL REVIEW

Figure 3. (A) Doseresponse curves for ozone-treated reaction mixtures of the parent roxithromycin are shifted toward higher doses (expressed as dilution of the initial sample) with increasing ozone exposure. (B) If reaction mixture toxicity is dominated by the parent compound, measured potency equivalents are expected to decrease in direct proportion to residual parent compound concentration, consistent with the experimental data. Figure adapted and reprinted with permission from ref 42. Copyright 2009 American Chemical Society.

knowledge databases such as the University of Minnesota Biodegradation/Biocatalysis Database (UM-BBD).35 Alternatively, more recent developments in chemical analytical techniques increasingly facilitate structural identification of TPs even in the absence of chemical reference standards.3638 As reviewed by Celiz et al.39 and Kraus et al.,40 such techniques include highresolution mass spectrometry combined with MSn experiments and/or H-1 and C-13 nuclear magnetic resonance. To circumvent the need for laborious sample enrichment or the use of high concentrations that are not of environmental relevance, new LCNMR techniques coupled to MS detection are currently being developed.41 Once unequivocally identified, and hence in most cases also synthesized, TPs might be directly subjected to effect testing. For instance, standard testing for effects on aquatic organisms is required for all major TPs identified in sediment-water studies in the EU Pesticide Directive. Effect-Driven Approach. In the effect-driven approach (Figure 1, II), samples from simulation studies taken over time are not only subjected to chemical analysis to determine the transformation kinetics of the parent compound, but are also analyzed with one or more bioassays to follow the development of toxicity over the course of the experiment. In Figure 2, a flowchart describing the effect-driven approach is given. Note that the selection of the appropriate bioassays is crucial for the success of this approach, which is discussed in more detail further below. At the first tier, the effect-driven approach provides information on whether the toxicity is decreasing or increasing during a transformation process, and hence whether or not toxic TPs are likely to have been formed. This question can be answered by determining if the change in toxicity can be fully explained by the decrease of parent compound concentration. This first step of the approach was illustrated by Dodd et al.42 who followed the oxidation of antibiotics with a bacterial bioassay based on the growth inhibition of Escherichia coli and Bacillus subtilis (Figure 3). A single antibiotic compound (here roxithromycin) was oxidized with ozone, and doseresponse curves of the reaction mixture at different time points were recorded (Figure 3A). Effect concentrations for 50% growth inhibition, EC50, were derived

from the doseresponse curves, and the ratios of the EC50 of the initial sample divided by the EC50 of the reaction mixture at given time points were defined as “potency equivalents”. The potency equivalents were then plotted against the decreasing parent compound concentration (Figure 3B). Since the experimental data points in the plot in Figure 3B fell on a 1:1 line, they concluded that all antimicrobial activity was due to the parent compound and that the TPs did not contribute appreciably to mixture toxicity. All antibiotics (macrolides, β-lactams, fluoroquinolones, and other antibiotics) that were oxidized by ozone and hydroxyl radicals42 gave similar plots, clearly indicating that TPs are either not formed in significant amounts and/or their toxicity is much smaller than that of the parent compound. Similar studies were conducted with ciprofloxacin, which lost all activity during photocatalytic processes,43 triclosan and three sulfa antibiotics,44 which lost antibacterial activity during photolysis, ethinylestradiol, whose estrogenic activity was completely lost or strongly decreased upon chlorination, bromination, hydroxylation, or other oxidative processes typically used during water treatment,45 and with diuron, which did not form any toxicologically relevant TPs during (photo)oxidation.46 If, at the first tier, no measurable difference in transformation kinetics of the parent and toxicity kinetics is observed, as was the case in all examples reviewed in the preceding, it can be concluded that no toxic TPs have been formed, and hence no further experiments are necessary. If, in contrast, there are significant differences in the kinetics of parent compound transformation and toxicity development, further steps are needed to identify and characterize the TPs that contribute to the mixture toxicity of the reaction mixture. An example of such a case is the biodegradation of propranolol, where no appreciable reduction in toxicity was observed even when the parent compound was degraded by 70%.17 In this case, at a second tier, preparative fractionation of the reaction mixture according to hydrophobicity followed by toxicity testing of the individual fractions with relevant bioassays will elucidate if the increase in toxicity can be explained by an increase in hydrophobicity or if specifically acting/inherently more toxic TPs have been formed (Figure 2). This approach is also called “effectdirected fractionation” and was initially developed to identify toxic components of complex mixtures.4749 If the increase in 3838


type of study/degradation process

3839

various advanced oxidation processes photodegradation with simulated sunlight catalytic (TiO2) and noncatalytic ozonation photodegradation with simulated s unlight ozonation (lab scale) in spiked wastewater treatment plant effluent

simulated solar irradiation of synthetic seawater and freshwater

photodegradation with simulated sunlight

disappearance of mutagenicity of frazolidone, low mutagenicity of nitrofurazone TPs isomerization and polymerization of isoamylmethoxycinnamate and ethylhexylmethoxycinnamate led to decrease in toxicity; initial degradation product of 2-ethylhexyl-4-(dimethylamino)benzoate exhibited same toxicity as P, but further degradation decreased toxicity increased toxicity of reaction mixture after photolysis, but only % inhibition at the end of the experiment determined

initial increase in toxicity during ozonation 2-[2-(chlorophenyl)amino]benzaldehyde, more toxic than P due to higher bioconcentrationg first generation TPs (various sulfoxides) retained antibacterial activity, second generation TPs lost antibacterial activity

Vibrio fischeria and Daphnia magnab Scenedesmus vacuolatuse Bacillus subtilisf

more toxic TPs for imipramine; all others decrease in toxicity with decrease of P higher toxicity of TPs for D. magna

Vibrio fischeria Vibrio fischeria and Daphnia magnab

Daphnia magnab

53

50

114

Compilation in 104 113

93

112

111

110

109

Salmonella typhimorium (Ames test for mutagenicity) Scenedesmus vacuolatusc

in vitro gap junctional intercellular communication and cytotoxicity with a rat epithelial cell line Chlorella vulgarisd

47

several TPs identified: anthracene-1,4-dione more toxic to V. fischeri than P; 10hydroxyanthrone and anthracene-9,10-dione more toxic to S. vacuolatus than P; 1-hydroxyanthracene-9,10-dione and 1,4-dihydroxyanthracene-9,10-dione more genotoxic than Pg initial increase in toxicity during ozonation and three toxic TPs identified: 2-(20 formyl)phenyl-1-naphthaldehyde, 2-(20 -formyl)phenyl-1-naphthoic acid, 2-2carboxyphenyl-1-naphthoic acid. decrease in toxicity with increasing illumination; no toxicologically relevant TPs formed

Vibrio fischeria, Scenedesmus vacuolatusc and umuC test for genotoxicity

ref 108

identified TPs or changes in toxicity of reaction mixture toxicity retained during treatment, even after methomyl has completely disappeared

bioassay/test organism Vibrio fischeria and Daphnia magnab

a 30 min-EC50 for bioluminescence inhibition of Vibrio fischeri (Microtox assay). b 48 h-EC50 for immobilization. c 24 h-EC50 for the inhibition of reproduction of synchronized algae. d 72 h-EC50 for inhibition of growth rate. e 24 h-EC50 for the inhibition of reproduction of synchronized algae and various physiological end points. f EC50 for growth inhibition. g Effect-directed fractionation with RP-HPLC.

β-lactam antibiotics (penicillin and cephalexin)

diclofenac

clofibric acid

dipyrone, 4-methylaminoantipyrine, 4-formylaminoantipyrine, 4-acetylaminoantipyrine various pharmaceuticals sulfamethoxazole

photocatalytic degradation with UV radiation and TiO2 catalyst

sulfacetamide, sulfathiazole, sulfamethoxazole, and sulfadiazine nitrofurazone and frazolidone antibiotics various sunscreen agents

hypochlorite oxidation

ozonation (lab scale)

heterogeneous photocatalysis with titanium dioxide and homogeneous photocatalysis by photo-Fenton photodegradation with simulated sunlight

chrysene

anthracene

methomyl

parent compound

Table 1. Examples of the Use of Bioassays or Effect-Directed Fractionation to Evaluate if Toxicologically Relevant TPs are Formed in Aqueous Reactions Mixtures (P = Parent Compound)

Environmental Science & Technology CRITICAL REVIEW


Environmental Science & Technology toxicity is proportional to the hydrophobicity of the tested fraction, then the toxicity of the TP(s) can be explained by an increased bioaccumulation potential and the toxicity of the individual compounds isolated in the fractions can be estimated with existing methods from the octanolwater partition coefficient.14 In this way, the increased phytotoxicity of a phototransformation product of diclofenac could be explained by its increased hydrophobicity.50 Similarly, during the electrochemical degradation of the β-blocker metoprolol spiked to reverse osmosis concentrate produced in an advanced water treatment plant, the specific algal toxicity remained unchanged but the nonspecific toxicity of the reaction mixture increased 50-fold due to formation of more bioaccumulative, halogenated byproducts, which were identified with LC-MS.51 If, in contrast, the analysis at the second tier indicates that specifically acting or inherently more toxic TPs are formed, i.e., if toxicity is higher in a fraction of low hydrophobicity, a more detailed analysis has to follow to identify the structure and properties of the TPs employing the analytical techniques outlined for the exposure-driven approach. Unequivocal assignment of a single, most toxic TP might be possible in some cases by comparing the change in concentration of single TPs over time with the changes in toxicity. Examples of Combined Exposure and Effect Assessment in Process-Based Simulation Studies. Many of the earlier studies have combined both approaches discussed above and simultaneously identified TPs using chemical analysis while evaluating the (eco)toxicity of the reaction mixture (for a list of relevant studies see Table 1). An interesting example is the phototransformation of anthracene, where the algal toxicity was hardly changed during degradation but toxicity toward Vibrio fischeri and genotoxicity increased with irradiation. Two TPs (Table 1) could clearly be identified as genotoxic, while the parent was not genotoxic.52 A more recent example of a TP that was shown to be more toxic than the parent compound is the phototransformation product of diclofenac (Table 1), which was identified by effect-directed analysis,50 as discussed above. A comprehensive study by Dodd et al.53 identified TPs produced by ozonation of penicillin G and cephalexin with chemical analysis and measured their individual antibacterial activity using Bacillus subtilis. From the toxicity data of the parent compound and the TPs, the potency equivalents were calculated and compared to experimentally determined potency equivalents during the course of the experiment. Thus, a full mass and toxicity balance during the degradation reaction assured that no unidentified toxic TPs escaped attention.

’ PRIORITIZATION OF TPS WITH RESPECT TO THEIR ENVIRONMENTAL RISK As outlined above, most of the current work on TPs deals with their formation in studies representing single chemical or biological transformation processes or specific environmental compartments. However, most studies and also regulatory assessment schemes, with the exception of the EU Pesticide Directive, fall short of providing a more comprehensive assessment of which TP(s) might significantly contribute to the overall environmental risk posed by a given chemical or which, for that reason, might need to be included in environmental monitoring programs. To answer these questions, one would need to combine the information from simulation studies and effect assessments, be they exposure- or effect-driven, with a consideration of the

CRITICAL REVIEW

environmental fate and behavior of the parent chemical and all of its TPs. Given the large number of TPs that may form through various environmental transformation processes, an overall assessment is a daunting task, especially given the sobering reality that worldwide regulatory authorities do not have enough resources to adequately assess the risk posed by parent chemicals. The task is made even more difficult by the fact that, due to the current regulatory situation, only few experimental fate and (eco)toxicological data are available for TPs. Therefore, regulators as well as industry are seeking rapid and cost-effective approaches to deal with TPs in regulatory risk assessment.54 Given these challenges, a tiered strategy that, at the first tier, uses mostly models or simple screening tests for an initial prioritization of TPs seems most promising. This initial prioritization should highlight TPs needing further experimental investigation as estimates of their exposure and effects indicate that they are likely to significantly contribute to the overall risk of a chemical. Different prioritization methods for human metabolites and environmental TPs, evaluating their risk in surface or drinking water, have already been suggested.5557 However, they do not fully exploit all the opportunities in assessing TPs because (a) they are mostly scoring-based and do not use existing exposure modeling tools to estimate environmental concentrations of TPs; and (b) they rely only on experimental (eco)toxicity and environmental fate data and consequently neglect TPs for which no data are available. The prioritization methods could also be further improved by “read-across” from what is known about the fate and effects of the parent compound, because typically large parts of the parent compound’s molecular structure are conserved in the TPs. Since usually some experimental fate and effect data as well as monitoring data are available for the parent compound, this knowledge can be exploited to focus on the incremental changes in fate and effects due to transformation reactions. In this way, a full, independent risk assessment for the TPs becomes unnecessary, as only their risk relative to the parent compound needs to be evaluated. In Figure 4, a framework that integrates relative exposure and effect prediction methods into a comprehensive prioritization strategy for TPs is proposed. It starts with TP candidate(s) having been identified as discussed above (A). Exposure and effects of one or several TPs identified are estimated relative to exposure and effects of the parent compound (B and C). The product of relative exposure and relative effects is then compared against a threshold (D). If the threshold is set to 1, for instance, TPs that pose risk about equal to the parent compound are considered relevant and submitted to higher-tier risk assessment (E). If the threshold is set to 0.1, a 10% increase in risk is considered sufficient for a TP to be deemed relevant. More details on the individual steps follow. Predicting Exposure to Transformation Products. Environmental fate models that allow for a quantitative description of the formation and further fate of TPs have been developed over the last ten years, with the most prominent examples being multispecies, multimedia models5863 and the FOCUS groundwater and surface water models used in pesticide regulatory risk assessment in Europe.7 For compounds other than pesticides, one major obstacle preventing the use of such models to include TPs in the risk assessment seems to be the lack of experimental fate data for TPs required to parametrize these models. Fate data for TPs, such as partition coefficients, compartmental half-lives, and fractions of formation, can be estimated with (quantitative) 3840



CRITICAL REVIEW

Figure 4. Scheme for model-based prioritization of TPs for higher-tier assessment.

structureproperty relationships (QSPRs), but there remain considerable uncertainties with these prediction methods.6466 Particularly, environmental half-lives of TPs have in several instances been identified as the most influential, but also the most uncertain model input parameters in TP exposure assessment.54,59,67 At the same time, innovative approaches to read-across from the parent compound’s degradation half-life are rare, especially for biodegradation. This is mostly due to a lack of fundamental understanding of how a wide range of xenobiotic chemicals are transformed by environmental microbial communities, which hinders the development of appropriate quantitative structurebiodegradation relationships (QSBR).68 For the same reasons, the prediction of transformation pathways and hence TP structures and fractions of formation remains very challenging. Two such models to predict biodegradation products are currently available and under continued development, e.g., the commercial product CATABOL69 and the freely available University of Minnesota Pathway Prediction System (UM-PPS).38,70,71 Much of the work on these models focuses on improving the selectivity of the predictions, i.e., reducing the number of predicted TPs to only the most likely ones, which is an important prerequisite to making these tools useful for risk assessment.70 Given the lack of accurate input data, the use of highly parametrized and site-specific models to assess exposure to

TPs seems futile in most cases. Rather, the use of more generic models to evaluate the behavior of a TP relative to its parent compound in combination with more detailed model predictions or monitoring data for the parent compound seems most advisible. For the prediction of partition coefficients of TPs, prediction accuracy might be further improved by basing the prediction on experimental data for the parent compound (log Kexp(P) in Figure 4B) and only using QSPR tools to predict the difference in the partition coefficient between parent compound and TP (Δpred log Kexp(TP-P) in Figure 4B). To validate existing models for prediction of exposure to TPs, we compared two environmental fate models that predict loadings into surface waters of TPs relative to their parent chemicals against consistently measured field data; one for parent chemicals applied to soils,67 e.g., pesticides, and one for wastewater-relevant substances, e.g., pharmaceuticals.31 The study results indicated order-ofmagnitude agreement between measured and predicted concentrations for the pesticide TPs, and deviations of less than a factor of 2 for TPs of pharmaceuticals, suggesting that the models used were suitable to predict aquatic exposure to TPs to a level of accuracy that is sufficient to prioritize TPs with a high exposure potential. However, findings for pesticide TPs also reinforced earlier findings that for TPs, which are often more polar and hence more mobile than their precursors, the groundwater compartment becomes more important, 3841



CRITICAL REVIEW

Figure 5. (A) Toxicokinetics and toxicodynamics as two dimensions of toxicity; (B) possible positions of the TPs in the two-dimensional toxicity plot. ECx: Effect concentration causing x% effect; TP1: TP has lower bioconcentration potential; TP2: TP has higher bioconcentration potential.

both as a receiving compartment but also, through exfiltration, as a source of TPs to surface water. Although there is ample evidence that the latter pathway leads to high and constant levels of certain TPs in surface waters,7274 it is so far not explicitly included in any of the existing models for exposure assessment of TPs. Predicting (Eco)Toxicological Effects of Transformation Products. In contrast to exposure modeling, modeling of (eco)toxicological effects is less advanced, less accepted, and most likely also more challenging. Effect assessment as proposed in Figure 4C relies on information available for the parent compound and aims to estimate a relative effect in comparison to the parent, not absolute effect data. We have developed an approach that systematically exploits effect data for the parent compound to predict ranges of toxicity for the TPs. The approach is based on analyzing toxicity data in terms of the two inherent dimensions of toxicity instead of simply assessing toxicity (Figure 5A). One dimension is related to the toxicokinetics of a chemical. The other dimension focuses on the toxicodynamics or specificity of effects (e.g., if a specific biological receptor is affected). This type of analysis has been applied qualitatively to rationalize the increased toxicity of some TPs16 and quantitatively for the prediction of TP toxicity of various pharmaceuticals and pesticides.14 Subsequent studies have validated the approach experimentally for the TPs of diuron5 and fluoxetine.4 Toxicokinetics encompasses all processes from uptake of a chemical into a cell/organism to reaching its target site, i.e., uptake, distribution, metabolism, and excretion. As a first approximation, one can use equilibrium bioconcentration as a surrogate for toxicokinetics. Generally, the more hydrophobic a chemical, the more easily it can cross biological barriers and be taken up into a cell, and consequently the greater the bioconcentration potential and thus toxicity of that chemical. For this reason, the toxicokinetics dimension of toxicity is assumed to scale with the bioconcentration potential of a chemical (x-axis in Figure 5A). Typically, the octanolwater partition coefficient is used as an indicator of bioconcentration potential, but since many water-relevant chemicals and TPs are acids or bases, we recommend using a biomembranewater partition coefficient which better accounts for speciation, e.g., the liposomewater partition coefficient.75 Toxicodynamics refers to the interaction between a toxic chemical and its target site(s), which ultimately leads to observable toxic effects. Toxicodynamics are closely interlinked

with the mode of toxic action of a chemical. In our twodimensional analysis of toxicity, a measure for the specificity of the toxic effect is the toxic ratio (TR). TR is defined as the ratio between the predicted minimum (“baseline”) effect concentration, ECbaseline toxicity, of a compound and its actually measured effect concentration, ECexperimental (Figure 5A). Chemicals with a TR e 10 are classified as baseline toxicants and those with a TR > 10 as exhibiting a specific mode of toxic action.76 For each bioassay or biological end point, the QSAR equation (quantitative structureactivity relationship) for baseline toxicity can be experimentally established using a set of reference chemicals that are known to exhibit baseline toxicity only.75,77,78 As toxicity data of parent compounds are usually required for their registration and authorization, the TRs of the parent for different biological end points can be derived from experimental data. Note that a compound that is a baseline toxicant for one species might be specifically acting (TR > 10) in another species. TPs can now differ from the parent compounds in two ways: in their bioconcentration (toxicokinetics) and/or their mode of toxic action (toxicodynamics). Transformation reactions often lead to smaller, more polar, and thus less hydrophobic molecules, which are in turn less toxic (Figure 5B, TP1). In some instances, however, e.g., when polar or charged parts of a molecule are cleaved off, bioconcentration can increase (Figure 5B, TP2). It is straightforward to predict bioconcentration potential from biomembranewater partition coefficients, which in turn can be predicted from first principles14 or readily measured.79 In previous work, we predicted plausible ranges of toxicity for a large number of TPs by assuming that the toxicity of a TP lies between baseline toxicity (TR = 1) and the same specific mode of toxic action as the parent compound (TRTP = TRparent).14,15,80 Sinclair et al.16 compared several models for the prediction of TP toxicity and concluded that this toxicity range prediction approach performed best. However, none of the methods compared could be used to confidently predict a single best estimate TR for TPs. The key question when it comes to predicting the TR of TPs is whether toxicophores are being formed or destroyed during transformation processes. A toxicophore, also called structural alert, is a substructure in a molecule that causes a specific mode of toxic action. An example for a toxicophore is the organophosphate structure of insecticides that allows them to bind to acetylcholinesterase and inhibits this enzyme’s action. If, during a transformation process, the toxicophore of a chemical is lost, 3842


Environmental Science & Technology the resulting TP will most likely exhibit baseline toxicity (empty dots in Figure 5B). If the toxicophore is retained, the TP will stay on a line parallel to the baseline QSAR (black dots in Figure 5B), unless its intrinsic toxicity is modified. For example, for chemicals that react directly with DNA, a change in TR would most likely be correlated to a change in the reactivity of the TP toward DNA relative to the parent compound.81 In rare cases, transformation will form new toxicophores, resulting in TPs exhibiting higher intrinsic toxicity than the parent, which may, depending on the change in the toxicokinetics, also lead to higher toxicity. If a new toxicophore is formed, the TR may be higher than that of the parent (gray dots in Figure 5B). Inventories of toxicophores have been published previously76,82,83 and have been proposed for use as repositories of structural alerts for identification of compounds with specific modes of toxic action. While such structural alerts cannot be used to quantitatively predict toxicity and suffer from being fairly nonspecific, they could still be useful in the context of TP assessment to learn whether toxicophores are retained or destroyed, or if new ones are likely to have formed during transformation. The toxicokinetic/toxicodynamic approach to estimating TP toxicity has the additional advantage that it facilitates addressing mixture toxicity between parent compound and TPs by use of predictive mixture models. If compounds have the same toxicophore, they are likely to act according to the same mode of toxic action and accordingly to act concentration-additive in mixtures.84 The same holds for compounds with TR < 10, which are baseline toxicants and therefore are also expected to act concentration-additive. If, in contrast, the mixture contains TPs with significantly different TR values than the parent compound and/or toxicophores are lost or formed during transformation, the mixture hypothesis to be tested is called independent action.84 Neuwoehner et al. have demonstrated the application of these mixture principles on the example of mixtures of fluoxetine and its human metabolites4 and diuron and its environmental TPs.5 For relative effect assessment as it is described in Figure 4C, parent compound and TP structures are analyzed for the presence of toxicophores as a first step. If the TP is suspected to contain a new toxicophore, no prediction can be made and one must proceed directly to a higher-tier assessment. If the parent is already a baseline toxicant or it is specifically acting but the TP loses the known toxicophore, then the relative effect can be calculated as the ratio of experimental EC of the parent and the predicted baseline toxicity EC of the TP. If the TP retains the toxicophore of the parent, then the relative effect would be equivalent to the ratio of the specific ECs. If it is assumed that the TR remains constant, the relative effect is mathematically equivalent to the ratio of the predicted baseline toxicity EC values. Once a TP is estimated to contribute significantly to the risk of the parent compound, based on the assessment of relative risk (Figure 4D), dedicated environmental fate and (eco)toxicological investigations need to be conducted. These topics are reviewed in the next section.

’ HIGHER-TIER ASSESSMENT OF TRANSFORMATION PRODUCTS Environmental Fate Studies with Transformation Products. Whereas in laboratory and mesocosm studies data on

TP appearance are often collected alongside data on parent

CRITICAL REVIEW

compound fate (e.g., refs 8588), we have found that these studies are often not sufficient to confidently deduce environmental fate parameters for TPs due to problems with parameter identification. To accurately derive the fractions of formation and transformation rate constants for TPs from such studies, they should be run over long enough time periods and data should be collected to densely cover formation and further transformation of the targeted TPs. Alternatively, separate testing of TPs allows deriving more accurate fate parameters, but requires their chemical synthesis or their preparative isolation from simulation studies in which they are formed. Finally, formation and fate of TPs can also be studied in situ using appropriate analytical techniques and monitoring designs, a thorough discussion of which lies beyond the scope of this review. Most data on sorption and degradation of TPs have been generated in the context of pesticide registration and are made public by the authorities in pesticide reregistration dossiers. Reports on dedicated TP fate studies in the scientific literature rather address detailed investigations of nonstandard end points for prominent TPs that are either already known to be widespread in the environment or are known toxicants. Examples include studies on the sorption of three toxicologically relevant TPs of DDT to activated carbon used for remediation purposes,89 or soil degradation studies with three TPs of atrazine in adapted and nonadapted soils.90 More recently, with the increasing attention to risks associated with pharmaceuticals excreted into wastewater, dedicated environmental fate studies have been reported for major, active metabolites of human and veterinary pharmaceuticals. Examples include studies on the photolytic and microbial degradation of the active metabolite of Tamiflu, oseltamivir carboxylate,91,92 on the photodegradation of 4-methylamino-antipyrine, a major metabolite of the analgesic dipyrone,93 and on the sorption of major metabolites of veterinary pharmaceuticals to various types of organic matter.94,95 Ultimately, the environmental presence of TPs can only be confirmed through monitoring. Kolpin et al.96 recently reviewed the occurrence of TPs in the environment. Measurements of TPs in the field, so far, have mostly been restricted to major known TPs due to the limited availability of chemical reference standards for TPs. More recently, high-resolution mass spectrometric techniques have been shown to enable fast, sensitive, and reliable detection of TPs in the absence of chemical reference standards.93,97100 This has been exploited by Kern et al.100 to screen for the presence of about 2000 known and predicted TPs of 50 pesticides, biocides, and pharmaceuticals in Swiss surface waters in order to work toward obtaining a more comprehensive picture of the presence of TPs in the aquatic environment. For about half of the investigated parent compounds one to two TPs were found to be present in the seven surface water samples investigated. (Eco)Toxicological Evaluation of Transformation Products. A crucial step in the higher-tier effect assessment of TPs is the choice of the appropriate bioassay. Similar principles govern the choice of bioassays for identification of relevant TPs in process-based simulation studies and in environmental risk assessment. Ideally, one bioassay should be selected to target the specific mode of action of the parent compound. Such a bioassay should always be complemented by a nonspecific bioassay to capture cases where the TP has lost the specific mode of action of the parent compound but develops high nonspecific toxicity (if it is more hydrophobic than the parent) or a new mode of toxic 3843


Environmental Science & Technology action that is not picked up by the specific bioassay that targets the parent. If structural alerts point to a suspected new toxicophore in the TP, an appropriate specific bioassay should be applied. An illustrative example is a study we conducted on the ecotoxicological effects of the TPs of the herbicide diuron.101 Diuron and its primary TPs that retain the phenylurea toxicophore (1-(3,4-dichlorophenyl)-3-methlyurea (DCPMU), 3-(3chlorophenyl)-1,1-dimethylurea (MCPDMU), and 1-(3,4-dichlorophenyl)urea (DCPU)) are direct inhibitors of photosynthesis and thus their specific toxicity was well reflected in algae bioassays.5 Once both methyl groups were lost from the phenylurea (DCPU), the TR was greatly reduced and further TPs were identified as baseline toxicants in algae. Structural alerts indicated that the third generation TP, 3,4-dichloroaniline, might exhibit a specific mode of toxic action in invertebrates.83 Bioassays conducted with Daphnia magna confirmed a specific mode of action for 3,4-dichloroaniline, but indicated baseline toxicity of the parent and the first and second generation TPs.5 However, in many previous studies, bioassays used to characterize toxicity of TPs did not reflect the specific mode of toxic action of the parent compound.102,103 For instance, many studies used the Microtox test with the bioluminescent bacterium Vibrio fischeri (reviewed in ref 104). Microtox is a rapid (30 min) acute toxicity screening test, which is very useful for obtaining a first idea of overall mixture toxicity, but is not suitable for identifying specifically acting TPs. Since loss of activity in the Microtox assay only reflects loss in bioconcentration potential (Figure 5), its use for the assessment of TPs of diuron102 or TPs of an organophosphate insecticide103 yielded only limited information. Finally, very few studies on the effects of TPs at higher levels of biological organization such as populations or ecosystems have been performed so far. These studies took the approach of either adding up exposure concentrations of TPs and parent compounds and relating the sum to quality criteria for the parent compound, effectively assuming identical toxicity of parent and TP,105 or of performing an independent risk assessment for parent compound and TP.106

’ WRAPPING IT ALL UP: A PLEA FOR A MORE SYSTEMATIC ASSESSMENT OF TPS In this review, we distilled existing literature on risk assessment of transformation products into a general framework that will hopefully stimulate future, more systematic studies. The overall picture emerging from this literature review indicates that the presence of TPs in the aquatic environment is not negligible and that TPs clearly contribute to the environmental and human health risk of organic micropollutants. Given the overwhelming number of transformation pathways and products, a comprehensive risk assessment for all TPs is out of reach. The future lies in better exploiting existing knowledge from parent compounds’ risk assessments and in using a tiered approach, which employs computational tools and simple screening tests to identify and prioritize relevant TPs for further testing. In this review, examples are given of how empirical parent compound exposure, fate, and effect data can be profited from to predict the corresponding TP data. Since TPs differ from their parent compounds typically by only incremental modifications of the molecular structure, it is suggested that this kind of readacross between parent and TPs reduces uncertainty in predictions for TPs. However, read-across methods are not well developed yet and

CRITICAL REVIEW

must be applied with caution. For instance, even minor structural changes can in some cases alter properties dramatically. Such effects might be easily overlooked if read-across approaches are used without appropriate expert knowledge. Examples include the oxidative activation of double bonds leading to reactive TPs, structural modifications enabling internal H-bond formation and thus changing sorption behavior, or the formation of structural moieties that fit into specific receptors, as is the case with the formation of dioxin TPs from chlorinated diphenyl ether precursors. In the future, attention should therefore be given to systematically validating read-across approaches between parent compounds and TPs, and to potentially also implementing them into prediction software and databases. A higher risk originating from TPs than identified for their parent compound appears to be the exception rather than the rule, but it is of utmost importance to identify these exceptions. TPs can pose a higher hazard than their parent with respect to any hazard indicator, including persistence, bioaccumulation, toxicity, and long-range transport potential.11 The high uncertainty in presently available methods for predicting persistence is one obstacle to be overcome. Research is ongoing to increase the accuracy of predicting rates of microbial transformations by developing improved, transformation mechanism-based QSBRs. While the prediction of bioaccumulation is relatively robust, future work on the effect side should focus on quantitative models to predict the toxic ratio as relevant indicator of the intrinsic potency of a chemical.107 To reach this goal, the identification of toxicophores as indicators of specific modes of action needs further scrutiny. As modes of action are speciesdependent, future developments in this field should account for species-specific toxicophores. Finally, toxicophore analysis currently is purely qualitative and should be developed into a quantitative tool in the future.

’ AUTHOR INFORMATION Corresponding Author

*Phone: þ61 7 3274 9180; e-mail: [email protected] (B.I.E.); phone: þ41 44 8235085; e-mail: [email protected] (K.F.).

’ ACKNOWLEDGMENT We thank Anita Poulsen for compiling the table on bioassay applications. We thank Silvio Canonica, Kristin Schirmer, Hana Mestankova, and Urs von Gunten for helpful discussions and Michael Dodd, Caroline Gaus, Damian Helbling, Juliane Hollender, Carla Ng, and Martin Scheringer for reviewing the manuscript. Funding by the Swiss Federal Office for the Environment (FOEN) within the project KoMet is gratefully acknowledged. ’ REFERENCES (1) Schwarzenbach, R. P.; Escher, B. I.; Fenner, K.; Hofstetter, T. B.; Johnson, C. A.; von Gunten, U.; Wehrli, B. The challenge of micropollutants in aquatic systems. Science 2006, 313, 1072–1077. (2) Boxall, A. B. A.; Sinclair, C. J.; Fenner, K.; Kolpin, D.; Maud, S. J. When synthetic chemicals degrade in the environment. Environ. Sci. Technol. 2004, 38, 368A–375A. (3) Boxall, A. B. A., Ed. Transformation Products of Synthetic Chemicals in the Environment; Springer: Berlin/Heidelberg, 2009; Vol. Part P. (4) Neuwoehner, J.; Fenner, K.; Escher, B. I. Physiological modes of action of fluoxetine and its human metabolites in algae. Environ. Sci. Technol. 2009, 43, 6830–6837. 3844


Environmental Science & Technology (5) Neuwoehner, J.; Zilberman, T.; Fenner, K.; Escher, B. I. Mixture toxicity and QSAR analysis of diuron and its transformation products to assess their mode of toxic action in algae and daphnids. Aquat. Toxicol. 2010, 97, 58–67. (6) EU Regulation (EC) No 1107/2009 of the European Parliament and of the Council of 21 October 2009 concerning the placing of plant protection products on the market and repealing Council Directives 79/ 117/EEC and 91/414/EEC; 2009; L 309, pp 150. (7) FOCUS Forum for the Co-ordination of Pesticide Fate Models and their Use. European Communities, 19952006; http://focus.jrc.ec. europa.eu/ (accessed 3 September 2010). (8) EC (European Commission). Guidance document on aquatic ecotoxicology in the context of the Directive 91/414/EEC, 2002. (9) EC (European Commission). Guidance document on the assessment of the relevance of metabolites in groundwater of substances regulated under Council Directive 91/414/EEC, 2003. (10) EU Regulation (EC) No 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC and repealing Council Regulation (EEC) No 793/93 and Commission Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/ EEC, 93/67/EEC, 93/105/EC and 2000/21/EC. Off. J. Eur. Communities: Legis. 2006, L 396, 1849. (11) ECHA. Guidance on Information Requirements and Chemical Safety Assessment, Part B: Hazard Assessment, Version 1.1; European Chemicals Agency: Helsinki, Finland, 2008. (12) VICH. VICH-GL38. Environmental impact assessment (EIAS) for veterinary medicinal products - Phase II Guidance; VICH International Cooperation on Harmonisation of Technical Requirements for Registration of Veterinary Medicinal Products: Brussels, Belgium, 2004. (13) EMEA. Guideline on the Environmental Risk Assessment of Medicinal Products for Human Use; Committee for Medicinal Products for Human Use (CHMP); European Medicines Agency (EMEA): London, 2006. (14) Escher, B. I.; Baumgartner, R.; Lienert, J.; Fenner, K. Predicting the Ecotoxicological Effects of Transformation Products. In The Handbook of Environmental Chemistry, Vol. 2, Reaction and Processes, Part P Transformation Products of Synthetic Chemicals in the Environment; Boxall, A. B. A., Ed.; Springer: Berlin/Heidelberg, 2009; pp 205244; DOI: 210.1007/1098_1002_1015. (15) Lienert, J.; G€udel, K.; Escher, B. I. Screening method for ecotoxicological hazard assessment of 42 pharmaceuticals considering human metabolism and excretory routes. Environ. Sci. Technol. 2007, 41, 4471–4478. (16) Sinclair, C. J.; Boxall, A. B. A. Ecotoxicity of transformation products. In The Handbook of Environmental Chemistry, Vol. 2, Reaction and Processes, Part P Transformation Products of Synthetic Chemicals in the Environment; Boxall, A. B. A., Ed.; Springer: Berlin/Heidelberg, 2009; pp 177204; DOI: 110.1007/1698-1002-1019. (17) Nalecz-Jawecki, G.; Wojcik, T.; Sawicki, J. Evaluation of in vitro biotransformation of propranolol with HPLC, MS/MS, and two bioassays. Environ. Toxicol. 2008, 23, 52–58. (18) Escher, B. I.; Bramaz, N.; Lienert, J.; Neuwoehner, J.; Straub, J. O. Mixture toxicity of the antiviral drug tamifluÒ (oseltamivir ethylester) and its active metabolite oseltamivir acid. Aquat. Toxicol. 2010, 96, 194–202. (19) D’Ascenzo, G.; Di Corcia, A.; Gentili, A.; Mancini, R.; Mastropasqua, R.; Nazzari, M.; Samperi, R. Fate of natural estrogen conjugates in municipal sewage transport and treatment facilities. Sci. Total Environ. 2003, 302, 19–209. (20) Gomes, R. L.; Scrimshaw, M. D.; Lester, J. N. Fate of conjugated natural and synthetic steroid estrogens in crude sewage and activated sludge batch studies. Environ. Sci. Technol. 2009, 43, 3612–3618. (21) Ternan, N. G.; Mc Grath, J. W.; Mc Mullan, G.; Quinn, J. P. Organophosphonates: occurrence, synthesis and biodegradation by microorganisms. World J. Microbiol. Biotechnol. 1998, 14, 635–647.

CRITICAL REVIEW

(22) Joss, A.; Siegrist, H.; Ternes, T. A. Are we about to upgrade wastewater treatment for removing organic micropollutants?. Water Sci. Technol. 2008, 57, 251–255. (23) von Gunten, U. Ozonation of drinking water: Part II. Disinfection and by-product formation in presence of bromide, iodide or chlorine. Water Res. 2003, 37, 1469–1487. (24) Richardson, S. D. Water Analysis: Emerging contaminants and current issues. Anal. Chem. 2009, 81, 4645–4677. (25) Gagnon, C.; Lajeunesse, A.; Cejka, P.; Gagne, F.; Hausler, R. Degradation of selected acidic and neutral pharmaceutical products in a primary-treated wastewater by disinfection processes. Ozone-Sci. Eng. 2008, 30, 387–392. (26) Dieter, H. H. The relevance of “non-relevant metabolites” from plant protection products (PPPs) for drinking water: The German view. Regul. Toxicol. Pharmacol. 2010, 56, 121–125. (27) Larson, R. A.; Weber, E. J. Reaction Mechanisms in Environmental Organic Chemistry; Lewis Publishers, CRC Press Inc.: Boca Raton, FL, 1994. (28) Buth, J. M.; Steen, P. O.; Sueper, C.; Blumentritt, D.; Vikesland, P. J.; Arnold, W. A.; McNeill, K. Dioxin photoproducts of triclosan and its chlorinated derivatives in sediment cores. Environ. Toxicol. Chem./ SETAC 2010, 44, 4545–4551. (29) Balmer, M. E.; Poiger, T.; Droz, C.; Romanin, K.; Bergqvist, P. A.; Muller, M. D.; Buser, H. R. Occurrence of methyl triclosan, a transformation product of the bactericide triclosan, in fish from various lakes in Switzerland. Environ. Sci. Technol. 2004, 38, 390–395. (30) Spycher, S.; Pellegrini, E.; Gasteiger, J. Use of structural descriptor to discriminate between modes of toxic action of phenols. J. Chem. Inf. Model. 2005, 45, 200–208. (31) Kern, S.; Baumgartner, R.; Helbling, D. E.; Hollender, J.; Singer, H.; Schwarzenbach, R. P.; Fenner, K. Qualitative and quantitative analysis of the formation of biotransformation products of selected pharmaceuticals and biocides during activated sludge treatment. J. Environ. Monit. 2010, 12, 2100–2111. (32) Schulz, M.; Loffler, D.; Wagner, M.; Ternes, T. A. Transformation of the X-ray contrast medium iopromide in soil and biological wastewater treatment. Environ. Sci. Technol. 2008, 42, 7207–7217. (33) EC (European Commission). Directive 98/8/EC of the European Parliament and of the council of 16 February 1998 concerning the placing of biocidal products on the market. Off. J. Eur. Communities: Legis. 1998, L 123, 163. (34) OECD Guidelines for the Testing of Chemicals; Organisation for Economic Co-operation and Development (OECD): Paris, 2010. (35) Ellis, L. B. M.; Roe, D.; Wackett, L. P. The University of Minnesota Biocatalysis/Biodegradation Database: the first decade. Nucl. Acid. Res. 2006, 34, D517–D521. (36) Kormos, J. L.; Schulz, M.; Wagner, M.; Ternes, T. A. Multistep approach for the structural identification of biotransformation products of iodinated X-ray contrast media by liquid chromatography/hybrid triple quadrupole linear ion trap mass spectrometry and H-1 and C-13 nuclear magnetic resonance. Anal. Chem. 2009, 81, 9216–9224. (37) Radjenovic, J.; Pereza, S.; Petrovic, M.; Barcelo, D. Identification and structural characterization of biodegradation products of atenolol and glibenclamide by liquid chromatography coupled to hybrid quadrupole time-of-flight and quadrupole ion trap mass spectrometry. J. Chromatogr. A 2008, 1210, 142–153. (38) Helbling, D. E.; Hollender, J.; Kohler, H.-P. E.; Fenner, K. High-throughput identification of microbial transformation products of organic micropollutants. Environ. Sci. Technol. 2010, 44, 6621–6627. (39) Celiz, M. D.; Tso, J.; Aga, D. S. Pharmaceutical metabolites in the environment: analytical challenges and ecological risks. Environ. Toxicol. Chem./SETAC 2009, 28, 2473–2484. (40) Krauss, M.; Singer, H.; Hollender, J. LC-high resolution MS in environmental analysis: from target screening to the identification of unknowns. Anal. Bioanal. Chem. 2010, 397, 934–951. (41) Yang, Z. Online hyphenated liquid chromatography-nuclear magnetic resonance spectroscopy-mass spectrometry for drug metabolite and nature product analysis. J. Pharm. Biomed. Anal. 2006, 40, 516–527. 3845


Environmental Science & Technology (42) Dodd, M. C.; Kohler, H. P. E.; Von Gunten, U. Oxidation of antibacterial compounds by ozone and hydroxyl radical: elimination of biological activity during aqueous ozonation Processes. Environ. Sci. Technol. 2009, 43, 2498–2504. (43) Paul, T.; Dodd, M. C.; Strathmann, T. J. Photolytic and photocatalytic decomposition of aqueous ciprofloxacin: Transformation products and residual antibacterial activity. Water Res. 2010, 44, 3121–3132. (44) Wammer, K. H.; Lapara, T. M.; McNeill, K.; Arnold, W. A.; Swackhamer, D. L. Changes in antibacterial activity of triclosan and sulfa drugs due to photochemical transformations. Environ. Toxicol. Chem./ SETAC 2006, 25, 1480–1486. (45) Lee, Y.; Escher, B. I.; von Gunten, U. Efficient removal of estrogenic activity during oxidative treatment of waters containing steroid estrogens. Environ. Sci. Technol. 2008, 42, 6333–6339. (46) Mestankova, H.; Escher, B. I.; Schirmer, K.; von Gunten, U.; Canonica, S. Evolution of algal toxicity during (photo)oxidative degradation of diuron. Aquat. Toxicol. 2011, 101, 466–473. (47) Brack, W. Effect-directed analysis: a promising tool for the identification of organic toxicants in complex mixtures?. Anal. Bioanal. Chem. 2003, 377, 397–407. (48) Brack, W.; Schmitt-Jansen, M.; Machala, M.; Brix, R.; Barcelo, D.; Schymanski, E.; Streck, G.; Schulze, T. How to confirm identified toxicants in effect-directed analysis. Anal. Bioanal. Chem. 2008, 390, 1959–1973. (49) Blasco, C.; Pico, Y. Prospects for combining chemical and biological methods for integrated environmental assessment. TRACTrends Anal. Chem. 2009, 28, 745–757. (50) Schulze, T.; Weiss, S.; Schymanski, E.; von der Ohe, P. C.; Schmitt-Jansen, M.; Altenburger, R.; Streck, G.; Brack, W. Identification of a phytotoxic photo-transformation product of diclofenac using effectdirected analysis. Environ. Pollut. 2010, 158, 1461–1466. (51) Radjenovic, J.; Escher, B.; Rabaey, K. Electrochemical degradation of the β-blocker metoprolol by Ti/Ru0.7Ir0.3O2 and Ti/SnO2-Sb electrodes. Water Res. 2011, in press (10.1016/j.watres.2011.03.040). (52) Brack, W.; Altenburger, R.; K€uster, E.; Meissner, B.; Wenzel, K. D.; Sch€u€urman, G. Identification of toxic products of anthracene photomodification in simulated sunlight. Environ. Toxicol. Chem./SETAC 2003, 22, 2228–2237. (53) Dodd, M. C.; Rentsch, D.; Singer, H. P.; Kohler, H. P. E.; Von Gunten, U. Transformation of β-Lactam antibacterial agents during aqueous ozonation: reaction pathways and quantitative bioassay of biologically-active oxidation products. Environ. Sci. Technol. 2010, 44, 5940–5948. (54) Ng, C. A.; Fenner, K.; Hungerb€uhler, K.; Scheringer, M. A framework for evaluating the contribution of degradation products to chemical persistence in the environment. Environ. Sci. Technol. 2010, 45, 111–117. (55) Belfroid, A. C.; van Drunen, M.; Beek, M. A.; Schrap, S. M.; van Gestel, C. A. M.; van Hattum, B. Relative risks of transformation products of pesticides for aquatic ecosystems. Sci. Total Environ. 1998, 222, 167–183. (56) Sinclair, C. J.; Boxall, A. B. A.; Parsons, S. A.; Thomas, M. R. Prioritization of pesticide environmental transformation products in drinking water supplies. Environ. Sci. Technol. 2006, 40, 7283–7289. (57) Hebert, A.; Forestier, D.; Lenes, D.; Benanou, D.; Jacob, S.; Arfi, C.; Lambolez, L.; Levi, Y. Innovative method for prioritizing emerging disinfection by-products (DBPs) in drinking water on the basis of their potential impact on public health. Water Res. 2010, 44, 3147–3165. (58) Fenner, K.; Kooijman, C.; Scheringer, M.; Hungerbuhler, K. Including transformation products into the risk assessment for chemicals: The case of nonylphenol ethoxylate usage in Switzerland. Environ. Sci. Technol. 2002, 36, 1147–1154. (59) Fenner, K.; Scheringer, M.; Hungerbuhler, K. Joint persistence of transformation products in chemicals assessment: Case studies and uncertainty analysis. Risk Anal. 2003, 23, 35–53. (60) Cahill, T. M.; Mackay, D. A high-resolution model for estimating the environmental fate of multi-species chemicals: application to malathion and pentachlorophenol. Chemosphere 2003, 53, 571–581. (61) Gandhi, N.; Bhavsar, S. P.; Gewurtz, S. B.; Diamond, M. L.; Evenset, A.; Christensen, G. N.; Gregor, D. Development of a

CRITICAL REVIEW

multichemical food web model: Application to PBDEs in Lake Ellasjoen, Bear Island, Norway. Environ. Sci. Technol. 2006, 40, 4714–4721. (62) Gasser, L.; Fenner, K.; Scheringer, M. Indicators for the exposure assessment of transformation products of organic micropollutants. Environ. Sci. Technol. 2007, 41, 2445–2451. (63) Schenker, U.; Scheringer, M.; Hungerbuhler, K. Including degradation products of persistent organic pollutants in a global multimedia box model. Environ. Sci. Pollut. Res. 2007, 14, 145–152. (64) Fenner, K.; Canonica, S.; Escher, B. I.; Gasser, L.; Spycher, S.; Tulp, H. C. Developing methods to predict chemical fate and effect endpoints for use within REACH. Chimia 2006, 60, 683–690. (65) Nguyen, T. H.; Goss, K. U.; Ball, W. P. Polyparameter linear free energy relationships for estimating the equilibrium partition of organic compounds between water and the natural organic matter in soils and sediments. Environ. Sci. Technol. 2005, 39, 913–924. (66) Aronson, D.; Boethling, R. S.; Howard, P. H.; Stiteler, W. Estimating biodegradation half-lives for use in chemical screening. Chemosphere 2006, 63, 1953–1960. (67) Kern, S.; Singer, H.; Hollender, J.; Schwarzenbach, R. P.; Fenner, K. Assessing exposure to transformation products of soil-applied organic contaminants in surface water: Comparison of model predictions and field data. Environ. Sci. Technol. 2011, 45, 2833–2841. (68) Jaworska, J. S.; Boethling, R. S.; Howard, P. H. Recent developments in broadly applicable structure-biodegradability relationships. Environ. Toxicol. Chem./SETAC 2003, 22, 1710–1723. (69) Dimitrov, S.; Pavlov, T.; Nedelcheva, D.; Reuschenbach, P.; Silvani, M.; Bias, R.; Comber, M.; Low, L.; Lee, C.; Parkerton, T.; Mekenyan, O. A kinetic model for predicting biodegradation. SAR QSAR Environ. Res. 2007, 18, 443–457. (70) Fenner, K.; Gao, J. F.; Kramer, S.; Ellis, L.; Wackett, L. Datadriven extraction of relative reasoning rules to limit combinatorial explosion in biodegradation pathway prediction. Bioinformatics 2008, 24, 2079–2085. (71) Ellis, L. B. M.; Gao, J.; Fenner, K.; Wackett, L. P. The University of Minnesota pathway prediction system: predicting metabolic logic. Nucleic Acids Res. 2008, 36, W427–W432. (72) Battaglin, W. A.; Thurman, E. M.; Kalkhoff, S. J.; Porter, S. D. Herbicides and transformation products in surface waters of the Midwestern United States. J. Am. Water Resour. Assoc. 2003, 39, 743–756. (73) Huntscha, S.; Singer, H.; Canonica, S.; Schwarzenbach, R. P.; Fenner, K. Input dynamics and fate in surface water of the herbicide metolachlor and of its highly mobile transformation product metolachlor ESA. Environ. Sci. Technol. 2008, 42, 5507–5513. (74) Kalkhoff, S. J.; Lee, K. E.; Porter, S. D.; Terrio, P. J.; Thurman, E. M. Herbicides and herbicide degradation products in upper Midwest agricultural streams during August base-flow conditions. J. Environ. Qual. 2003, 32, 1025–1035. (75) Escher, B.; Schwarzenbach, R. P. Mechanistic studies on baseline toxicity and uncoupling as a basis for modeling internal lethal concentrations in aquatic organisms. Aquat. Sci. 2002, 64, 20–35. (76) Verhaar, H. J. M.; Van Leeuwen, C. J.; Hermens, J. L. M. Classifying environmental-pollutants 1. Structure-activity-relationships for prediction of aquatic toxicity. Chemosphere 1992, 25, 471–491. (77) K€ onemann, H. Quantitative structure-activity relationships in fish toxicity studies. Part 1: relationship for 50 industrial pollutants. Toxicology 1981, 19, 209–221. (78) Vaes, W. H. M.; Urrestarazu-Ramos, E.; Verhaar, H.; Hermens, J. L. M. Acute toxicity of nonpolar versus polar narcosis: is there a difference?. Environ. Toxicol. Chem./SETAC 1998, 17, 1380–1384. (79) Escher, B. I.; Schwarzenbach, R. P.; Westall, J. C. Evaluation of liposome-water partitioning of organic acids and bases. 1. Development of a sorption model. Environ. Sci. Technol. 2000, 34, 3954–3961. (80) Escher, B. I.; Bramaz, N.; Richter, M.; Lienert, J. Comparative ecotoxicological hazard assessment of beta-blockers and their human metabolites using a mode-of-action-based test battery and a QSAR approach. Environ. Sci. Technol. 2006, 40, 7402–7408. (81) Harder, A.; Escher, B. I.; Landini, P.; Tobler, N. B.; Schwarzenbach, R. P. Evaluation of bioanalytical tools for toxicity assessment and mode of 3846


Environmental Science & Technology toxic action classification of reactive chemicals. Environ. Sci. Technol. 2003, 37, 4962–4970. (82) Verhaar, H. J. M.; Ramos, E. U.; Hermens, J. L. M. Classifying environmental pollutants. 2. Separation of class 1 (baseline toxicity) and class 2 ('polar narcosis’) type compounds based on chemical descriptors. J. Chemometr. 1996, 10, 149–162. (83) von der Ohe, P. C.; Kuhne, R.; Ebert, R. U.; Altenburger, R.; Liess, M.; Schuurmann, G. Structural alerts - A new classification model to discriminate excess toxicity from narcotic effect levels of organic compounds in the acute daphnid assay. Chem. Res. Toxicol. 2005, 18, 536–555. (84) Altenburger, R.; Nendza, M.; Schuurmann, G. Mixture toxicity and its modeling by quantitative structure-activity relationships. Environ. Toxicol. Chem. 2003, 22, 1900–1915. (85) Sapozhnikova, Y.; Pennington, P.; Wirth, E.; Fulton, M. Fate and transport of Irgarol 1051 in a modular estuarine mesocosm. J. Environ. Monitor. 2009, 11, 808–814. (86) Liu, X.-C.; Shen, M.; Li, S.-Q.; Chen, H. Residue dynamics of prochloraz and its metabolite 2,4,6-trichlorophenol in mushroom and soil. Nongyaoxue Xuebao 2009, 11, 362–366. (87) Doran, G.; Eberbach, P.; Helliwell, S. Sorption and degradation of fipronil in flooded anaerobic rice soils. J. Agric. Food Chem. 2009, 57, 10296–10301. (88) Unold, M.; Kasteel, R.; Groeneweg, J.; Vereecken, H. Transport and transformation of sulfadiazine in soil columns packed with a silty loam and a loamy sand. J. Contam. Hydrol. 2009, 103, 38–47. (89) Hale, S. E.; Tomaszewski, J. E.; Luthy, R. G.; Werner, D. Sorption of dichlorodiphenyltrichloroethane (DDT) and its metabolites by activated carbon in clean water and sediment slurries. Water Res. 2009, 43, 4336–4346. (90) Krutz, L. J.; Shaner, D. L.; Zablotowicz, R. M. Enhanced degradation and soil depth effects on the fate of atrazine and major metabolites in Colorado and Mississippi soils. J. Environ. Qual. 2010, 39, 1369–1377. (91) Sacca, M. L.; Accinelli, C.; Fick, J.; Lindberg, R.; Olsen, B. Environmental fate of the antiviral drug Tamiflu in two aquatic ecosystems. Chemosphere 2009, 75, 28–33. (92) Bartels, P.; von Tumpling, W. The environmental fate of the antiviral drug oseltamivir carboxylate in different waters. Sci. Total Environ. 2008, 405, 215–225. (93) Gomez, M. J.; Sirtori, C.; Mezcua, M.; Fernandez-Alba, A. R.; Aguera, A. Photodegradation study of three dipyrone metabolites in various water systems: Identification and toxicity of their photodegradation products. Water Res. 2008, 42, 2698–2706. (94) Kim, H. J.; Lee, D. S.; Kwon, J. H. Sorption of benzimidazole anthelmintics to dissolved organic matter surrogates and sewage sludge. Chemosphere 2010, 80, 256–262. (95) Khan, B.; Qiao, X. L.; Lee, L. S. Stereoselective sorption by agricultural soils and liquid-liquid partitioning of trenbolone (17 alpha and 17 beta) and trendione. Environ. Sci. Technol. 2009, 43, 8827–8833. (96) Kolpin, D. W.; Battaglin, W. A.; Conn, K. E.; Furlong, E. T.; Glassmeyer, S. T.; Kalkhoff, S. J.; Meyer, M. T.; Schnoebelen, D. J. Occurence of Transformation Products in the Environment. In The Handbook of Environmental Chemistry, Vol. 2, Reaction and Processes, Part P Transformation Products of Synthetic Chemicals in the Environment; Boxall, A. B. A., Ed.; Springer: Berlin/Heidelberg, 2009; pp 205244; DOI: 210.1007/1698_1002_1011. (97) Ibanez, M.; Sancho, J. V.; Pozo, O. J.; Hernandez, F. Use of quadrupole time-of-flight mass spectrometry in environmental analysis: Elucidation of transformation products of triazine herbicides in water after UV exposure. Anal. Chem. 2004, 76, 1328–1335. (98) Perez, S.; Eichhorn, P.; Celiz, M. D.; Aga, D. S. Structural characterization of metabolites of the X-ray contrast agent iopromide in activated sludge using ion trap mass spectrometry. Anal. Chem. 2006, 78, 1866–1874. (99) Durand, S.; Legeret, B.; Martin, A. S.; Sancelme, M.; Delort, A. M.; Besse-Hoggan, P.; Combourieu, B. Biotransformation of the triketone herbicide mesotrione by a Bacillus strain. Metabolite profiling

CRITICAL REVIEW

using liquid chromatography/electrospray ionization quadrupole timeof-flight mass spectrometry. Rapid Commun. Mass Spectrom. 2006, 20, 2603–2613. (100) Kern, S.; Fenner, K.; Singer, H. P.; Schwarzenbach, R. P.; Hollender, J. Identification of transformation products of organic contaminants in natural waters by computer-aided prediction and high-resolution mass spectrometry. Environ. Sci. Technol. 2009, 43, 7039–7046. (101) Giacomazzi, S.; Cochet, N. Environmental impact of diuron transformation: a review. Chemosphere 2004, 56, 1021–1032. (102) Tixier, C.; Sancelme, M.; Bonnemoy, F.; Cuer, A.; Veschambre, H. Degradation products of a phenylurea herbicide, diuron: Synthesis, ecotoxicity, and biotransformation. Environ. Toxicol. Chem./SETAC 2001, 20, 1381–1389. (103) Caceres, T.; Megharaj, M.; Naidu, R. Toxicity and transformation of fenamiphos and its metabolites by two micro algae Pseudokirchneriella subcapitata and Chlorococcum sp. Sci. Total Environ. 2008, 398, 53–59. (104) Radjenovic, J.; Petrovic, M.; Barcelo, D. Complementary mass spectrometry and bioassays for evaluating pharmaceutical-transformation products in treatment of drinking water and wastewater. TRACTrends Anal. Chem. 2009, 28, 562–580. (105) Rand, G. M.; Carriger, J. F.; Gardinali, P. R.; Castro, J. Endosulfan and its metabolite, endosulfan sulfate, in freshwater ecosystems of South Florida: a probabilistic aquatic ecological risk assessment. Ecotoxicology 2010, 19, 879–900. (106) Hall, L. W.; Anderson, R. D.; Killen, W. D.; Balcomb, R.; Gardinali, P. The relationship of irgarol and its major metabolite to resident phytoplankton communities in a Maryland marina, river and reference area. Mar. Pollut. Bull. 2009, 58, 803–811. (107) Maeder, V.; Escher, B. I.; Scheringer, M.; Hungerb€uhler, K. Toxic ratio as an indicator of the intrinsic toxicity in the assessment of persistent, bioaccumulative, and toxic chemicals. Environ. Sci. Technol. 2004, 38, 3659–3666. (108) Fernandez-Alba, A. R.; Hernando, D.; Aguera, A.; Caceres, J.; Malato, S. Toxicity assays: a way for evaluating AOPs efficiency. Water Res. 2002, 36, 4255–4262. (109) Luster-Teasley, S. L.; Yao, J. J.; Herner, H. H.; Trosko, J. E.; Masten, S. J. Ozonation of chrysene: Evaluation of byproduct mixtures and identification of toxic constituent. Environ. Sci. Technol. 2002, 36, 869–876. (110) Baran, W.; Sochacka, J.; Wardas, W. Toxicity and biodegradability of sulfonamides and products of their photocatalytic degradation in aqueous solutions. Chemosphere 2006, 65, 1295–1299. (111) Nakamura, H.; Kawakami, T.; Niino, T.; Takahashi, Y.; Onodera, S. Chemical fate and changes in mutagenic activity of antibiotics nitrofurazone and furazolidone during aqueous chlorination. J. Toxicol. Sci. 2008, 33, 621–629. (112) Rodil, R.; Moeder, M.; Altenburger, R.; Schmitt-Jansen, M. Photostability and phytotoxicity of selected sunscreen agents and their degradation mixtures in water. Anal. Bioanal. Chem. 2009, 395, 1513–1524. (113) Trovo, A. G.; Nogueira, R. F. P.; Aguera, A.; Sirtori, C.; Fernandez-Alba, A. R. Photodegradation of sulfamethoxazole in various aqueous media: Persistence, toxicity and photoproducts assessment. Chemosphere 2009, 77, 1292–1298. (114) Rosal, R.; Gonzalo, M. S.; Boltes, K.; Leton, P.; Vaquero, J. J.; Garcia-Calvo, E. Identification of intermediates and assessment of ecotoxicity in the oxidation products generated during the ozonation of clofibric acid. J. Hazard. Mater. 2009, 172, 1061–1068.

3847


Recent Advances in Environmental Risk Assessment of

Recommend Documents