A hierarchical testing strategy for micropollutants in drinking water regarding their potential endocrine-disrupting effects-towards health-related indicator values.

In Germany, micropollutants that (may) occur in drinking water are assessed by means of the health-related indicator value (HRIV concept), developed by the German Federal Environment Agency. This concept offers five threshold values (≤ 0.01 to ≤ 3 µg l-1) depending on availability and completeness of data regarding genotoxicity, neurotoxicity, and germ cell-damaging potential. However, the HRIV concept is yet lacking integration of endocrine disruptors as one of the most prominent toxicological concerns in water bodies, including drinking water. Thresholds and proposed bioassays hence urgently need to be defined. Since endocrine disruption of ubiquitary chemicals as pharmaceuticals, industrial by-products, or pesticides is a big issue in current ecotoxicology, the aim of this study was to explore endocrine effects, i.e., estrogenic and androgenic effects, as an important, additional toxicological mode of action for the HRIV concept using a hierarchical set of well-known but improved bioassays. Results indicate that all of the 13 tested substances, industrial chemicals and combustion products (5), pharmaceuticals and medical agents (4), and pesticides and metabolites (4), have no affinity to the estrogen and androgen receptor in human U2OS cells without metabolic activation, even when dosed at their water solubility limit, while in contrast some of these substances showed estrogenic effects in the RYES assay, as predicted in pre-test QSAR analysis. Using a specifically developed S9-mix with the U2OS cells, those micropollutants, i.e., Benzo[a]pyrene, 2,4-Dichlorophenol, 3,3-Dichlorbenzidin, 3,4-Dichloranilin, and diclofenac, they show estrogenic effects at the same concentration range as for the yeast cells. Three of the drinking water-relevant chemicals, i.e., atrazine, tributyltin oxide, and diclofenac, caused effects on hormone production in the H295R assay, which can be correlated with changes in the expression of steroidogenic genes. One chemical, 17α-Ethinylestradiol, caused an estrogenic or anti-androgenic effect in the reproduction test with Potamopyrgus antipodarum. Considering these results, a proposal for a test strategy for micropollutants in drinking water regarding potential endocrine effects (hormonal effects on reproduction and sexual development) will be presented to enhance the existing HRIV concept.

Influence of GSM signals on human peripheral lymphocytes: study of genotoxicity.

Exposure to radiofrequency (RF) electromagnetic fields (EMF) is continuously increasing worldwide. Yet, conflicting results of a possible genotoxic effect of RF EMF continue to be discussed. In the present study, a possible genotoxic effect of RF EMF (GSM, 1,800 MHz) in human lymphocytes was investigated by a collaboration of six independent institutes (institutes a, b, c, d, e, h). Peripheral blood of 20 healthy, nonsmoking volunteers of two age groups (10 volunteers 16-20 years old and 10 volunteers 50-65 years old) was taken, stimulated and intermittently exposed to three specific absorption rates (SARs) of RF EMF (0.2 W/kg, 2 W/kg, 10 W/kg) and sham for 28 h (institute a). The exposures were performed in a setup with strictly controlled conditions of temperature and dose, and randomly and automatically determined waveguide SARs, which were designed and periodically maintained by ITIS (institute h). Four genotoxicity tests with different end points were conducted (institute a): chromosome aberration test (five types of structural aberrations), micronucleus test, sister chromatid exchange test and the alkaline comet assay (Olive tail moment and % DNA). To demonstrate the validity of the study, positive controls were implemented. The genotoxicity end points were evaluated independently by three laboratories blind to SAR information (institute c = laboratory 1; institute d = laboratory 2; institute e = laboratory 3). Statistical analysis was carried out by institute b. Methods of primary statistical analysis and rules to adjust for multiple testing were specified in a statistical analysis plan based on a data review before unblinding. A linear trend test based on a linear mixed model was used for outcomes of comet assay and exact permutation test for linear trend for all other outcomes. It was ascertained that only outcomes with a significant SAR trend found by at least two of three analyzing laboratories indicated a substantiated suspicion of an exposure effect. On the basis of these specifications, none of the nine end points tested for SAR trend showed a significant and reproducible exposure effect. Highly significant differences between sham exposures and positive controls were detected by each analyzing laboratory, thus validating the study. In conclusion, the results show no evidence of a genotoxic effect induced by RF EMF (GSM, 1,800 MHz).

Swimming pool water--fractionation and genotoxicological characterization of organic constituents.

Swimming pool water treatment in general includes flocculation, sand filtration, and subsequent disinfection with chlorine. The continuous chlorination and input of organic material by bathers in combination with recirculation of the pool water leads to an accumulation of disinfection by-products (DBPs) in the water. Several DBPs have been identified as human carcinogens and are thought to cause allergic asthma. Therefore, the elimination of DBPs is one major aim of pool water treatment. Using membrane filtration as an alternative treatment technology, DBPs can be removed more efficiently than with conventional treatment. In this study membrane filtration and genotoxicity testing were applied for the characterization of pool water constituents and for the identification of the necessary molecular weight cut off of the membrane for an efficient elimination. Two-step membrane filtration revealed that most of the DBPs (as adsorbable organically bound halogen, AOX) were present in the molecular weight fraction below 1000 g/mol. The fraction below 200 g/mol contained more than 30% of the AOX. The distribution of the dissolved organic carbon (DOC) across the fractions was similar to that of the AOX. The genotoxicity was found to be strongest in the low-molecular weight fraction. Thus, considerable DBP removal by membrane treatment requires membranes with low-molecular weight cut offs down to 200 g/mol. The comprehensive elimination of the genotoxic compounds requires further treatment steps.

On the relevance of genotoxicity for fish populations II: genotoxic effects in zebrafish (Danio rerio) exposed to 4-nitroquinoline-1-oxide in a complete life-cycle test.

In order to characterize the impact of genotoxic potentials on populations of aquatic organisms in surface waters, zebrafish (Danio rerio) were exposed to the model genotoxicant 4-nitroquinoline-1-oxide (NQO) in a complete life-cycle test. Fish exposed to mean NQO concentrations of 0, 0.1, 0.3, 1.1, and 2.9 microg/l were examined by several genotoxicity assays with different endpoints. Assays included the unscheduled DNA synthesis (UDS) test, the comet assay, the alkaline filter elution, and the micronucleus test. The genotoxicity assays revealed an increasing genotoxicity, ranging from induction of DNA repair (even at the lowest concentration tested) to primary and secondary DNA alterations at higher concentrations of 1.1 and 2.9 microg/l NQO. Whether the lowered reproductivity observed in the life-cycle test is caused by genotoxic pathways of NQO, remains unclear. However, the results indicate a contradiction to an earlier assumption that genotoxicants as found in the environment are likely to not impact natural populations.