Open Access

Martin Wagner, Jörg Oehlmann

• Human exposure to endocrine disruptors is well documented by biomonitoring data. However, this information is limited to few chemicals like bisphenol A or phthalate plasticizers. To account for so-far unidentified endocrine disruptors and potential mixture effects we employ bioassays to detect endocrine activity in foodstuff and consequently characterize the integrated exposure to endocrine active compounds. Recently, we reported a broad contamination of commercially available bottled water with estrogenic activity and presented evidence for the plastic packaging being a source of this contamination. In continuation of that work, we here compare different sample preparation methods to extract estrogen-like compounds from bottled water. These data demonstrate that inappropriate extraction methods and sample treatment may lead to false-negative results when testing water extracts in bioassays. Using an optimized sample preparation strategy, we furthermore present data on the estrogenic activity of bottled water from France, Germany, and Italy: eleven of the 18 analyzed water samples (61.1%) induced a significant estrogenic response in a bioassay employing a human carcinoma cell line (MCF7, E-Screen). The relative proliferative effects ranged from 19.8 to 50.2% corresponding to an estrogenic activity of 1.9-12.2 pg estradiol equivalents per liter bottled water. When comparing water of the same spring that is packed in glass or plastic bottles made of polyethylene terephthalate (PET), estrogenic activity is three times higher in water from plastic bottles. These data support the hypothesis that PET packaging materials are a source of estrogen-like compounds. Furthermore, the findings presented here conform to previous studies and indicate that the contamination of bottled water with endocrine disruptors is a transnational phenomenon.