Assessment Frameworks of Multiple Stressors

Assessment Frameworks of Multiple Stressors A review of environmental and human health risk assessment frameworks of multiple stressors: the case of endocrine disruptors Abstract This review is a compilation of the recommended approaches and methods for the development of a risk assessment framework of multiple stressors. Some of the driving forces contributing to address this subject are the current demands of stakeholders like the drinking water industry, the society and regulators of evaluating the risks of mixtures of contaminants that may harm human beings and the environment. Therefore, our work aims at exploring the possibility of integrating within the risk assessment, environmental, human and societal aspects, acknowledging new international regulations and policies for the safe use of chemicals which enforce the integrative study of the hazards of multiple chemicals on humans and the environment throughout their life-cycle. We chose the group of compounds called endocrine disruptors as an example of multiple stressors because of their emerging relevance and the fact that they usually integrate complex mixtures, originate from multiple sources and ex ist in several environmental compartments, generating adverse effects on receptors through different routes and pathways. Their actions may be severe chronic and long-term modifications of the normal development and reproduction patterns of the individual and/or its progeny, eventually representing systemic risks at the population level which may affect sustainability and biodiversity. Due to the ubiquity of these chemicals, it is necessary to address the inclusion of human beings as potential receptors and deal with risk assessment in an integrated manner. As endocrine disruptors may provoke differentiated responses relative to the developmental stage, acting through varied mechanisms and at very low concentrations, with the particularity that their toxicokinetics may present sometimes unusual dose-response curves, might challenge long-term predictions and hazard characterization, adding to risk assessment uncertainties. References to the current methodologies including the applica ble bioassays, chemical analysis, modeling, statistics tools and equations to calculate joint effects considering the interactions of toxicants within a mixture are also discussed in this review. KEY WORDS: risk assessment, endocrine disruptors, multiple stressors 1. Introduction The aim of this review is to analyze the risk assessment frameworks and current practices, the integration of environmental and human health methodologies, the effects evaluation and how to assess the risk of complex mixtures of chemicals. Assessing the risks of multiple stressors for human health and the environment arises from the realization that biological and physical stressors usually coexist in complex mixtures in the natural and constructed environment, sometimes generating impacts on living creatures. Historically, since the publication in 1962 of the book â€Å"Silent Spring†, by Rachel Carson, a warning on the deleterious actions on wildlife of persistent toxicants, such as DDT, which caused a decrease of birds in Pennsylvania, became presentwas recognised among the scientific world, the media and the society. More recently, the research by Dr. Theo Colborn on the reproductive disorders of predators in the Great Lakes of North America and her book, published in 1996 â€Å"Our Stolen Future† co-authored by herself with Dianne Dumanoski and John Peterson Myers, was able to generate public awareness on the threats that EDCs might pose to human beings and the environment. Some of the driving forces for writing this review are the demands of stakeholders represented by the community, the NGOs, the health and environmental regulators, the manufacture industrial sector and the drinking water companies. For instance, the European Environment and Health Strategy emphatically stresses the gaps in knowledge related to risk assessment methodologies that include foetuses, infants and children and calls for the precautionary principle within the strategy for environmental contaminants, for which there is a sufficient level of scientific evidence at the effect level (e.g. molecular, cellular, or tissue-related) to show the likelihood of health impacts. Not enough information exists on the link between emissions of dioxins and PCBs and other substances and their accumulation in ecosystems and foodstuffs. The need for research on the understanding of the links between environmental factors and certain diseases is recognized, but, due to the complexity of the issue , the immediate action is to gather evidence in order to concrete measures to protect human health and the environment. Many reports are able to demonstrate through laboratory experiments and field surveys that exposure of animals to chemicals released into the environment exert reproductive or developmental effects on the individual and its off-spring, affecting the viability of the species at a population level (Colborn and Smolen 1997). As a matter of fact, these kind of adverse effects have been observed in wildlife and many of them can be attributed to the exposure to man-made chemicals. The cause-effect relationship is still controversial in human beings, but still a matter of concern due to the severity of probable harm that could affect individuals and populations. Thus, regulatory decisions must be informed by risk assessment on this important topic (Fenner-Crisp 2000). Endocrine-mediated effects may be more relevant in populations rather than in individuals. As there is still not consensus about a cause-effect relationship, it is recommended a science-based precautionary approach to protect public health. Further research is needed to demonstrate effects and carry out birth defect registries and epidemiology studies designed to track delayed effects of environmental exposures (Solomon and Schettler 2000). The classical paradigm of human health risk assessment authored by the National Research Council (NRC, 1983) is composed of four steps: hazard identification, dose-response assessment, exposure assessment and risk characterization. This paradigm was modified in 1994 to include characterization of each component. One of the approaches considered by some authors as best suited for developing a risk assessment of multiple stressors is a cumulative risk assessment framework, which may include societal aspects with participatory involvement of stakeholders (Gentile and Harwell 2001; Callahan and Sexton 2007; USEPA 2003). The global trend towards a policy-driven integration applied to risk assessment, demanded by regulations on safety of chemicals and industrial operations should ideally include a multiplicity of stressors, compartments, geographical scales, and end-points (Assmuth and Hildà ©n 2008). For example, the new European Union Regulation on chemicals and its safe use (REACH, EC 1907/2006) enforces linking risks to human health and the environment for chemicals throughout their life cycle. In United States, the Environmental Protection Agency, still discusses both topics separately because of practical reasons, but acknowledging the need to integrate them. An overview of the most suitable risk assessment frameworks is described in this literature review, focusing on the case of mixtures of reproductive endocrine disruptors. Matters that differentiate this topic are also presented, regarding particularities in mechanistic and toxicokinetics aspects and some of the sources of uncertainties and confounding factors. Developing a novel approach to the classical risk assessment scheme is also a goal, with the intention of contributing to design a risk assessment framework comprising a choice of tests, models, computational and statistical tools. 2. Hazard identification In this stage the nature of the hazard is described taking into account toxicity data. The hazard can then be characterized deriving numerical values of acceptability of the risk, based on mechanisms of action, biological extrapolation, dose-response and dose-effect relationships, and related uncertainties (Fenner-Crisp, 2003). 2.1. Nature of the hazard Endocrine disruptors (EDCs) are substances that interfere with the endocrine system by changing homeostasis and producing developmental effects and/or diminishing the fertility of the organisms. EDCs include a broad range of substances which can be classified according to their effect. The best known are the environmental estrogens, alkylphenol and its ethoxylates, the monomer in polycarbonate manufacture bisphenol A, some pesticides and chlorinated organics. 2.2. Sources of EDCs Possible sources of watercourses pollution with EDCs are wastewater sewage discharge, industrial effluents, or point and non-point source contamination of watercourses with agrochemicals such as herbicides or pesticides. Sewage discharge from cities contains residues of domestic products such as personal care products, pharmaceuticals and detergents and excreta with natural and artificial steroidal hormones from contraceptive pill usage. Nonylphenol (NP) is a highly hydrophobic bioccumulating biodegradation by-product of nonylphenol ethoxylate non-ionic surfactants which persists in sewage sludge and river sediments. Its use and production have been banned in EU countries and strictly monitored in many other countries such as Canada and Japan (Soares et al. 2008). Pulp mills are some of the industries associated to studies on endocrine disruption. Bleached Kraft pulp mill effluents have been linked to both estrogenic and androgenic effects on biota, depending on the process characteristics and wastewater treatment. Several studies have associated the chlorination of organic matter to the estrogenicity of the effluent. Nowadays, the application of elemental-chlorine-free processes has diminished the risk of dioxins and furans formation, but not eliminated it, as other halogenated organics are formed by use of chlorine dioxide as bleaching agent. Various wood-extractive compounds produced in the pulping process, such as rosin acids and phytosterols and found in pulp mills effluents have also been considered potentially responsible of endocrine disruption (Hewitt et al. 2008). The main identified resin acids in pulp mill effluents are: pimaric, isopimaric, sandaracopimaric, palustric, dehydroabietic, abietic and neoabietic acid (Merilà ¤inen and Oikari, 2008). Other compounds found in this type of effluent are phenolic guaiacyl-based lignin degradation products, diterpenoids, and dimethoxy stilbene (Belknap et al, 2006). Modern analytical methods, like full-scan GC-MS have been used to identify wood related extractives in final effluent from a chlorine dioxide bleached pulp mill effluent, including monoterpenes, phenolics, fatty acids, resin acids, resin acid neutrals and sterols (Wartman et al. 2009). Receptor binding bioassays for androgens and estrogens indicated that androgens were most abundant in this effluent and the range of androgens for the various extraction protocols used was 189-283 ng/L as testosterone equivalent concentration. Some examples of common sources of EDCs and typical environmental concentrations are summarized on Table 1. Table 1.Sources of EDCs and typical environmental concentrations Origin, use and occurrence Source of environmental exposure EDC group Example molecule Typical concentrations Reference Industrial (pulp and paper mills) Contaminated fish Resin acids pimaric acid 4-140  µg g-1 Owens et al, 1994 Industrial (pulp and paper mills) Industrial wastewater treatment plant Chlorinated organics 2,4,6-trichlorophenol 1.5  µg l-1 Owens et al, 1994 Industrial (pulp and paper mills) Final stage secondary treatment Phytosterols ÃŽ ²-sitosterol 58.42  µg l-1 Landman et al, 2008 Domestic, (contraceptive pills) Sewage effluent Pharmaceuticals 17ÃŽ ±-ethynylestradiol 14-17 ng l-1 Liu et al, 2004 Human and animal excreta Sewage effluent Natural steroid hormones 17ÃŽ ²-estradiol 5.0 ng l-1 Koh YKK et al, 2007 Domestic and industrial (laundry detergents, wool scouring processes) Sewage sludge Non ionic surfactants 4-nonylphenol 829.3 mg/kg Gonzà ¡lez et al, 2010 Domestic and industrial (polycarbonate bottles) Leaching from solid waste, sewage effluent Polycarbonate bisphenol A 0.62  µg l-1 Sà ¡nchez-Avila et al, 2009 Agricultural (soil fertilization) Livestock waste Male steroid hormones testosterone 10-1830 ng l−1 Lange et al, 2002 Agricultural (dairy farming) Streams contaminated by dairy cow excreta Female steroid hormones 17ÃŽ ²-oestradiol 0.04-3.6 ng l−1 Matthiessen et al, 2006 Agricultural (weed and grass control in soybean crops) Run-off Herbicide glyphosate 0.1-0.7 mg l-1 Peruzzo et al, 2008 2.2. Dose-response assessment There are several methods to demonstrate dose-response relationships, either by in vivo or in vitro tests. Fish reproduction tests, like the ones using the model fish fathead minnow (Pimephales promelas) have shown a decrease in fecundity associated with depressed steroid and vitellogenin (Vtg) production in female specimens (Ankley et al. 2008). Many of the tests rely on the measurement of an increase of Vtg as biomarker of estrogenicity as seen in several publications (Schwaiger et al. 2002; An et al. 2008; Holbech et al. 2006 Panter et al, 1998; Sohoni et al. 2001; Kunz and Fent, 2009). Tests results on resin acids show different responses in the first generation of fish than in the second (Christianson-Heiska et al. 2007). In some cases there are not many examples of in vivo tests, like for glyphosate. A fish exposure tests with this compound showed Vtg induction in female fish, indicative of estrogenic activity (An et al, 2008). An investigation working the commercial formulation of the herbicide glyphosate and human placental cells demonstrated its toxicity at concentrations lower than the usual in agricultural practices. The aromatase activity disruption seems to be due not only to glyphosate but also to co-adjuvants (the surfactant nonylphenol or others), which enhance its bioavailability and/or bioaccumulation (Richard et al. 2005; Gasnier et al. 2009). Table 2 shows some examples of dose-response experiments working with fish, crustacea, molluscs and amphibia. Varied protocols exist to develop ecotoxicity tests, in flow-through, static or partly renewal conditions, and with different duration and end-points. Only chronic effects and particularly developmental and reproductive effects were conside red. Table 2. Dose-response for endocrine disruption effects in freshwater organisms exposed to single EDCs EDC chemical name Taxonomic group Species Dose to produce effect Effect Test conditions Reference 4-nonylphenol Fish Rivulus marmoratus 300  µg l-1 Testicular agenesis and oogenesis inhibition in 100 % fish Static system, daily renewal Tanaka and Grizzle, 2002 4-nonylphenol Fish Oncorhynchus mykiss, rainbow trout 1 -10  µg l-1 10  µg l-1 High Vtg in adult male fish plasma Low hatching rate Intermittent exposure of adult fish for 4 months until spawning Schwaiger et al, 2002 4-nonylphenol Crustacean Ceriodaphnia dubia NOEC for reproduction: 1  µg l-1 Low hatching rate 7 days chronic exposure, static Isidori et al, 2005 Glyphosate Fish Carassius carassius, crucian carp 100% effluent Vtg induction in female fish (38.6 +/- 9.8  µg l-1) 3 weeks, continuous exposure An et al, 2008 Glyphosate Mollusk Pseudosuccinea columella, aquatic snail 1 mg l-1 10 mg l-1 Faster development of F3 embryos Hatching inhibition 3 generation continuous Tate et al, 1997 Estrone Fish Danio rerio LOEC: 14 ng l-1 50 ng l-1 Significant Vtg increase Higher female ratio 40 days fish sexual development test Holbech et al, 2006 17ÃŽ ²-Estradiol Fish Danio rerio LOEC: 54 ng l-1 Significant Vtg increase Higher female ratio 40 days fish sexual development test Holbech et al, 2006 17ÃŽ ²-Estradiol Fish Pimephales promelas, fathead minnow 100 ng l-1 Significant Vtg increase Testicular growth inhibition 21 days male fish exposure Panter et al, 1998 Estriol Fish Danio rerio LOEC: 0.6  µg l-1 21.7  µg l-1 Significant Vtg increase Higher female ratio 40 days fish sexual development test Holbech et al, 2006 Dehydroabietic acid (DHAA), resin acid Fish Danio rerio, zebra fish 50  µg l-1 Low plasma Vtg in female in F0; high Vtg and affected spermatogenesis in F1 males 2 generations, continuous Christianson-Heiska et al 2008 ÃŽ ²-sitosterol Fish Danio rerio 10-20  µg l-1 F1: higher ratio of male fish; F2: higher ratio of female fish 2 generation fish exposure test Nakari and Erkomaa, 2003 Quercetin, phytoestrogen Amphibian Xenopus laevis, frog 200  µg l-1 Higher female ratio > 10% abnormal testes (some with ovotestes) Exposure up to 1 month post-metamorphosis Cong et al, 2006 Phenanthrene, PAH Fish Oryzias latipes, Medaka NOEL: 100  µg l-1 Developmental, hatching 18 days, renewal Rhodes et al, 2005 Bisphenol A Mollusk Marisa cornuaretis, aquatic snail NOEC: 640  µg l-1 Developmental 12 weeks, juvenile snails Forbes et al, 2007 Bisphenol A Fish Pimephales promellas 16  µg l-1 640 and 1280  µg l-1 640  µg l-1 1280  µg l-1 Altered spermatogenesis Growth inhibition and Vtg induction in male fish Reduced hatchability in F1 generation Egg production inhibition 3 generation reproduction exposure test Sohoni et al, 2001 Bisphenol A Fish Brachydanio rerio, zebrafish EC50: 2.90  µg l-1 Embryo malformation and low hatchability 72 h exposure Liu et al, 2007 Benzo-ÃŽ ±-pirene (BaP) (PAH) Fish Fundulus heteroclitus , common mummichog 10  µg l-1 CYP19A1 expression decreased by about 50% in immature stage I oocytes Exposure for 10 or 15 days by in situ hybridization, several developmental stages Dong et al, 2008 Polychlorinated biphenyl 126 Fish Danio rerio, zebrafish LC50: 3.270 mg l-1 Developmental effects through aryl hydrocarbon receptor activation Dilutions of PCB 126 for 12 weeks SiÃ…Å ¸man et al, 2007 Polychlorinated biphenyl 126 Fish Salvelinus namaycush, lake trout 3 ÃŽ ¼g kg−1body weight Retinol depletion Oral exposure for 12 weeks; confirmation with radiolabelled retinol Palacea et al, 1997 Benzophenone-1 Fish Pimephales promelas 4919  µg l-1 Vtg induction 14 days exposure, semi-static, renewal Kunz and Fent, 2009 3. Exposure assessment 3.1. Ecosystems and human sub-populations potentially at risk of endocrine disruption effects Increasing evidence generated by scientists turn endocrine disruption into a recognized risk to the environment. Due to the ubiquity of EDCs and the widespread routes of exposure, most ecosystems and human populations are potentially at risk of endocrine disruption. Notwithstanding this fact, under the scope of a risk assessment of EDCs the potentially most vulnerable risk subgroups are identified corresponding to maternal, fetal and early developmental stages. The concern that prenatal or childhood exposure to EDCs may be responsible for abnormalities in human sexual and reproductive health are still in the hypothetical ground. However, many reports on exposure to high concentrations of recognized EDCs such as DES, certain PCBs, and DDT demonstrate this fact. At low-doses the question remains unanswered whether there could be a critical window where they could harm the fetal development (Hood 2005). Several reports on human developmental anomalies and reproductive ailments have been raising international concern, such as a seven fold increase risk of testicular cancer since 1969 to 2002 in men from several countries of Europe, United States and New Zealand. Also, the sperm density halved, as rates of cryptorchidism (undescended testicles) and hypospadias (shortened urinary tracts) simultaneously rose. It is thought that human congenital malformation of sex organs, low sperm quality, endometriosis, reduced fertility and some types of cancers of breast and testis could be linked to exposure to EDCs. More than 80000 synthetic chemicals are produced in the world and have still not been fully evaluating with regards to endocrine disruption. In 1996, the U.S. Environmental Protection Agency initiated an Endocrine Disruption Screening Program to evaluate more than 15,000 chemicals calling for a policy based on the â€Å"precautionary approach† to be overcautious and protect huma n health and the environment. A historical example of policies which demanded the banning of a drug due to these after-effects is the case of diethylstilbestrol (DEADES), which used to be prescribed to pregnant women to prevent spontaneous abortions because it produced higher risk of genital deformities and cancer in the offspring, among other effects (Stair 2008). Internationally there is consensus that the most vulnerable group for EDCs exposure are children. For example, in European countries, the Strategy for Environment and Health known as â€Å"SCALE† for Science, Children, Awareness, Legislation and Evaluation, has set as a priority agenda for the evaluation diseases caused by endocrine disruptors in children. The exposure to insecticides and herbicides used in agricultural practices has been linked to developmental or reproductive effects in wild animals and also in human beings. The occupational exposure to pesticide has received much attention, as for example prolonged time-to-pregnancy was observed in male greenhouse workers exposed to pesticides before conception of their first pregnancy (Bretveld et al 2008). The domestic exposure of children to residues of pesticides in low-level long-term exposures are associated to chronic effects and include routes of exposure such as fruit or breast milk (Goodman and Laverda 2002). 3.2. Evidence of endocrine disruption effects in wildlife around the world There are reports on impacts on wildlife reproduction and development observed in invertebrates, fish, reptiles, birds and mammals, sometimes confirmed by laboratory tests. In laboratory experiments the impacts to fish populations by EDCs affect reproductive health and persistence of various fish species (Mills and Chichester 2005). Many examples of impacts due to exposure to endocrine disruptors exist in wildlife, such as the seals population decline in the Baltic and North Sea, the high levels of female egg yolk in male fish or snail imposex and intersex around the world. Intersexuality of fish has been demonstrated in several investigations carried out in rivers around the world. The findings of abnormal reproductive female-like ducts and oocytes in male fish were related to the treated sewage discharge from the cities in laboratory experiments measuring induction of plasma vitellogenin in exposed male fish (Jobling et al. 2002). Field studies were carried out using wild roach as a model fish to confirm the incidence and the severity of intersex which correlated with the predicted concentrations of the natural estrogens (E1 and E2) and the synthetic contraceptive pill estrogen (EE2) present (Jobling et al. 2006). Some case-studies have made clear that the estrogenic activity of municipal wastewater correlates to demographics. The number of inhabitants was found to correlate with changes in estrogenic activities in a research conducted at a university city in US, with seasonal fluctuations in population. The concentrations of synthetic and natural estrogens and other EDCs were measured and effects demonstrated through the application of in vivo and in vitro tests (fish exposure with Vtg induction measurement and the yeast estrogen screen) (Brooks et al. 2003). The demonstration of effects of pulp mill effluents has also been supported by fish surveys with a sampling design that includes upstream and downstream sites from the discharge pipe of the pulp mill. For instance, Munkittrick et al. (1994) have demonstrated that the absence of chlorine bleaching or the presence of secondary treatment does not eliminate estrogenic responses evidenced by decreased circulating levels of sex steroids, decreased gonadal size, which implies that there may be multiple causative agents. In other cases, androgenic effects have been noticed, such as a biased male to female ratio in fish in Sweden downstream from pulp mills (Larsson and Fà ¶rlin 2002). As seen on Table 3, several adverse endocrine effects are evidenced in various animals, from mollusks to amphibian but they also appear in higher animal species. Table 3. Effects of EDCs in wildlife evidenced through field studies Animal Effect EDCs Postulated mechanism or causative agent Reference Frog High incidence of deformed frogs in Minnesota, United States Multiple EDCs Retinoid signaling pathways activation Gardiner et al. 2003 Marine Gastropods Masculinization of female snails (imposex) occurs worldwide. Females grow accessory sex organs including sperm ducts, seminal vesicles, external sperm grooves, and penises. Exposure to low levels of tributyltin (TBT) (1ng/l) Aromatase inhibition, testosterone inhibition, or neuroendocrine disorder or interaction with retinoid receptors Novà ¡k et al. 2008 Wild roach (Rutilius rutilus) Intersex, and high plasma Vtg concentration Multiple EDCs Sewage effluent from wastewater treatment plant discharging into rivers Joblin et al. 2006 Mosquitofish (Gambusia affinis) Masculinization (90% affected in number of segments in the longest anal fin ray). Androgen-dependent gene expression by luciferase test Kraft pulp mill effluent Affinity for human androgen receptor (hAR) Parks et al. 2001 Eastern Mosquitofish, (Gambusia holbrooki) Androgenic activity measured by androgen receptor transcription assay with human receptor in sediment. Fish masculinization. Paper mill effluent, river Pine pulp-derived phytosteroids accumulate in river sediment where they are converted by microbes into progesterone and this into androstenedione and other bioactive steroids Jenkins et al. 2003 3.3. Conceptual model Deriving a conceptual model requires knowing the pathways and toxicokinetics of the EDCs identified in the hazard identification step. An effects-based assessment start by identifying the effects and the relevant stressors and geographically located (for example through the use of GIS software). On the other hand, the model used in stressor-based assessments, depicts how stressors affect receptors and it is commonly applied when evaluating risks of environmental pollution. If a river basin is evaluated, the sources of contamination are studied, identifying the pathways, receptors and effects. To develop the human health risk assessment component, the fish consumption of the population and the drinking water intake are two of the main factors to consider especially for the most vulnerable population, which are newborn and lactating infants. The food chain is the main source of exposure, and in particular, fish consumption and drinking water are possible sources for the nursing mother and the pathway of distribution through the milk to the baby, but the direct intake of drinking water is important in the case of formula preparation. The environmental risk assessment should consider fish, crustacean and sediment dwelling organisms within the framework. During pregnancy maternal fat is moved, releasing to the blood the bioaccumulated compounds, due to their liposolubility and persistence, through all the different exposure routes (foodstuffs, environmental, occupational) throughout her life. Acute exposure should also be considered if it happened previously to gestation or during this period. There are substances that traspass the placental barrier and chemicals reach the offspring. Also, through the breastmilk, explaining the extrangely high levels of some xenobiotics (Fernà ¡ndez et al. 2007). 3.4. Methodologies to determine dose-response in exposure assessment The analysis of exposure and effect determines the concentration of the EDC on the environment matrixes matrices (water courses, ground water, drinking water, soil, sediment, air, biota), and assesses the potential or actual effects. In order to do so, many tools are recommended and in general a tiered approach is the most suited for this task as it helps to work in a logical order and increasing the specificity of the tests. One of the main sources of exposure to most chemicals is through the food chain. The bioconcentration of organics in beef, cow milk and vegetation correlates to the octanol-water partition coefficient (Kow) to predict the bioaccumulation in the aquatic and terrestrial food chains (Travis and Arms 1988). There are many models based on the characteristics of the chemicals, such as the fugacity model, which allows to predict the expected concentrations in six environmental compartments (water, air, soil, bottom and suspended sediment and fish) (MacKay et al. 1985). 3.4.1. The use of a tiered methodology to demonstrate endocrine disruptive effects This type of approach is carried out including different tests, such as bioassays, in vitro tests and field studies as part of the experimental design. The methodologies generally employed are in vivo fish reproduction exposure tests and in vitro receptor binding bioassays, for androgens and estrogens (Wartman et al., 2009). Even though there is an international trend towards diminishing the use of live organisms for experimentation for safety testing, in vivo tests are still of key importance for the confirmation of the findings of in vitro screens. Some of the most utilized tests relay on the use of fish as model experimental organism in various life-stages, as for example the 21 days reproduction fish test with fathead minnow (EPA/600/R-01/067). 3.4.2. In vitro screens and tests Some of the in vitro assays that can be used as screening tools of estrogenic activity are the following: yeast based assays, cell proliferation assays, bindin