Health safety of parabens evaluated by selected in vitro methods.

Seven selected parabens (4 allowed, 3 banned in cosmetics) were tested in order to confirm and expand historical data on their toxicological properties and safety. The aim was to apply novel in vitro methods, which have been sufficiently technically and scientifically validated for the purposes of toxicological testing of chemicals. The study included several toxicological endpoints such as skin/eye irritation, skin sensitization, endocrine disruption and genotoxicity. The battery of selected methods comprised regulatory accepted EpiDerm™ skin model (OECD TG 439); EpiOcular™ corneal model (OECD TG 492) and scientifically valid test method HET-CAM (DB-ALM Protocol No. 47); in chemico test DPRA (OECD TG 442C); in vitro test LuSens (OECD TG 442D) and in vitro test h-CLAT (OECD TG 442E); Ames MPF™ (Xenometrix) and XenoScreen YES/YAS (Xenometrix). Overall, none of the 4 allowed parabens exhibited skin/eye irritation or genotoxicity. However, all allowed parabens in cosmetics were predicted as samples with potentially sensitizing properties in the LuSens and h-CLAT test methods, but not confirmed by DPRA. Endocrine disruption was recorded only at high concentrations, whereas methyl paraben and ethyl paraben exhibited the lowest activity. This study confirmed the safety of use of the allowed parabens in the highest recommended concentrations in cosmetics or pharmaceuticals.

Safety testing of adult novelties using in vitro methods.

Despite widespread and prolonged use of adult novelties, their health safety is not regularly tested or legally regulated. In the EU, adult novelties are subjected to the General Product Safety Directive, placing the burden of proof regarding safe products onto the manufacturers. The aim of our pilot study was to expand knowledge on potential application of in vitro methods for hazard prediction of extracts from final products. We subjected extracts of 20 adult novelties, purchased on the Czech market to toxicological tests including NRU cytotoxicity assay, sensitization tests DPRA and LuSens and the YES/YAS endocrine assay. Four samples produced cytotoxicity. Sensitization potential was recorded by DPRA (three samples) while the LuSens reported ten samples. Regarding endocrine disruption, three samples produced antiestrogen and antiandrogen effects. Six samples exhibited androgenic potential and one sample showed estrogenic potential. Positive results with possible health effects were recorded repeatedly for samples made of ABS, PVC and latex. The study has confirmed promising usefulness of our test methods combination with regard to safety testing of this type of consumer products. The results should be evaluated with care, however, the data bring added-value to the limited knowledge of mixture toxicology and are indicative for further testing.

Prediction of acute toxicity of chemicals in mixtures: worms Tubifex tubifex and gas/liquid distribution.

The aim of this contribution is to support our proposal of the procedure for predicting acute toxicity of binary mixtures by QSAR analysis techniques. The changes of a mixture composition are described by molar ratio R and visualized in the R-plot (QCAR--quantitative composition-activity relationships). The approach was inspired by Rault and Dalton's laws, their positive and negative deviations in the behavior of a mixture of real gases, by Loewe and Muischnek isoboles and by the Finney test of additivity. Acute toxicity was determined by the laboratory test with woms Tubifex tubifex. The additivity of the acute toxicity in the binary mixture benzene + nitrobenzene was confirmed and a new interaction is described: "mixed interaction" with the binary mixture aniline + ethanol. The "mixed interaction" means that depending on mixture composition, both potentiation and inhibition can occur. As the first physicochemical descriptor of the changes caused by the changing composition of binary mixtures, the gas/liquid equilibrium was studied and a composition of the gaseous phase was determined by a gas chromatographic method. The method for determination of concentrations in the gaseous phase was described. The gaseous phase composition of benzene + nitrobenzene. benzene + ethanol, benzene + aniline and ethanol + aniline mixtures was analyzed. It was found that if the concentrations of the mixture's components in the gaseous phase behave nonideally (they are not additive), the acute toxicity of the same mixture is not additive as well. Another descriptor to distinguish between potentiation and inhibition will be, however, necessary. The properties, both gaseous phase composition and the acute toxicity, of the benzene + nitrobenzene mixture are additive. In mixtures with the mixed interaction, the R-plot of the composition of the gaseous phase is complex with a large variation of results.

Acute toxicity of binary mixture benzene-ethanol and partition coefficient K(ow) of benzene and ethanol.

The study related to partition coefficients between n-octanol and water of compounds in binary mixture benzene-ethanol was carried out. Partition coefficients of benzene and ethanol for different values of molar ratio of benzene in the mixture were determined. Collected results show statistically significant deviations the K(ow) of benzene for some molar ratios (0.2 to 0.6) from values for pure compound. For ethanol, there are no statistically significant deviations from values for pure compound, however there are some trends of changes of K(ow).

Acute toxicity of alcohols: prediction by QSAR analysis and by molecular similarity.

The purpose of the paper is to demonstrate the usefulness of techniques of molecular similarity as an alternative method for estimation of both acute toxicity of chemicals and their partition coefficients. The technique of atom pairs was chosen. A series of aliphatic alcohols was treated. Acute toxicity was determined by the express laboratory method as the EC50 (inhibition of movement of the worms Tubifex tubifex). Partition coefficients were determined between oil and gas. The calculated results were sufficiently close to both experimental results and the estimates calculated by a technique of QSAR analysis.