The Safety Assessment of Cosmetic Perfumes by Using In Chemico and In Vitro Methods in Combination with GC-MS/MS Analysis.

Animal testing has been prohibited for the safety assessment of cosmetic ingredients or finished products. Thus, alternative non-animal methods, followed by confirmatory clinical studies on human volunteers, should be used as the sole legally acceptable approach within the EU. The safety assessment of cosmetic products requires the involvement of multiple scientific disciplines, including analytical chemistry and biomedicine, as well as in chemico, in vitro and in silico toxicology. Recent data suggest that fragrance components may exert multiple adverse biological effects, e.g. cytotoxicity, skin sensitisation, (photo)genotoxicity, mutagenicity, reprotoxicity and endocrine disruption. Therefore, a pilot study was conducted with selected samples of fragrance-based products, such as deodorant, eau de toilette and eau de parfum, with the aim of integrating results from a number of alternative non-animal methods suitable for the detection of the following toxicological endpoints: cytotoxicity (with 3T3 Balb/c fibroblasts); skin sensitisation potential (in chemico method, DPRA); skin sensitisation potential (LuSens in vitro method, based on human keratinocytes); genotoxicity potential (in vitro Comet assay with 3T3 Balb/c cells); and endocrine disruption (in vitro YES/YAS assay). The presence of twenty-four specific known allergens in the products was determined by using GC-MS/MS. The strategies for estimation of the NOAEL of a mixture of allergens, which were proposed by the Scientific Committee on Consumer Products in their 'Opinion on Tea tree oil' document and by the Norwegian Food Safety Authority in their 'Risk Profile of Tea tree oil' report, were used as models for the NOAEL estimation of the mixtures of allergens that were identified in the individual samples tested in this study.

Toxicological testing of a photoactive phthalocyanine-based antimicrobial substance.

The aim of the study was toxicological testing of an innovative and efficient antimicrobial agent based on photoactive phthalocyanine (Pc) derivative. A promising Aluminium phthalocyanine (AlPc) with efficient and stable antimicrobial effects was subjected to a battery of toxicological tests to avoid local and systemic toxicity hazard. In compliance with the current European legislation restricting the use of experimental animals, the methods comprised exclusively in vitro procedures based on cellular and tissue models of human origin or mimicking human tissues. The battery of toxicological tests to identify local toxicity included skin corrosion/irritation, eye irritation, and phototoxicity. The basic systemic toxicity tests included acute toxicity, skin sensitization, genotoxicity, and endocrine disruption. The results showed that AlPc induced skin and eye irritation, exhibited borderline sensitization potential and mutagenic potential in one test strain of the Ames test, which was not confirmed in the chromosome aberration test. The AlPc was found to be phototoxic. The results from the cytotoxicity test designed for acute oral toxicity estimation were not conclusive, the acute toxicity potential has to be determined by conventional tests in vivo. Regarding endocrine disruption, no agonistic activity of the AlPc on human estrogen receptor α, nor human androgen receptor was observed. The skin penetration/absorption test revealed that the AlPc has not penetrated into the dermis and receptor fluid, confirming no risk of systemic exposure via the bloodstream.

Bio-mining of Lanthanides from Red Mud by Green Microalgae.

Red mud is a by-product of alumina production containing lanthanides. Growth of green microalgae on red mud and the intracellular accumulation of lanthanides was tested. The best growing species was Desmodesmus quadricauda (2.71 cell number doublings/day), which accumulated lanthanides to the highest level (27.3 mg/kg/day), if compared with Chlamydomonas reinhardtii and Parachlorella kessleri (2.50, 2.37 cell number doublings and 24.5, 12.5 mg/kg per day, respectively). With increasing concentrations of red mud, the growth rate decreased (2.71, 2.62, 2.43 cell number doublings/day) due to increased shadowing of cells by undissolved red mud particles. The accumulated lanthanide content, however, increased in the most efficient alga Desmodesmus quadricauda within 2 days from zero in red-mud free culture to 12.4, 39.0, 54.5 mg/kg of dry mass at red mud concentrations of 0.03, 0.05 and 0.1%, respectively. Red mud alleviated the metal starvation caused by cultivation in incomplete nutrient medium without added microelements. Moreover, the proportion of lanthanides in algae grown in red mud were about 250, 138, 117% higher than in culture grown in complete nutrient medium at red mud concentrations of 0.03, 0.05, 0.1%. Thus, green algae are prospective vehicles for bio-mining or bio-leaching of lanthanides from red mud.

Selected bisphenols and phthalates screened for estrogen and androgen disruption by in silico and in vitro methods.

OBJECTIVES: The aim of this study was to detect endocrine disruption potential of selected bisphenols and phthalates, compare in silico prediction with results from two in vitro methods and bring up-to-date information on development of EU legislation, available in vitro methods and biomechanisms involved in endocrine disruption. MATERIAL AND METHODS: In silico approach based on the OECD QSAR Toolbox was used for prediction of estrogen receptor α binding. OECD TG 455 assay and a yeast-based YES/YAS assay was used to determine the interactions with human estrogen (ERα) and androgen receptors. RESULTS: In silico results predicted the screened phthalates as non binders and bisphenols as very strong binders of the ERα. In vitro results differed from in silico prediction in several cases but exhibited concordance mainly for strong binders of ERα. Most of the substances exhibited parallel activity (agonist-antagonist) on both estrogen and androgen receptors. Agonistic studies showed the effective concentration of 10% activity (EC10) from 5.0E-07 for strong agonists (e.g. BPC, BPTMC). Cytotoxicity was observed after 48 h exposure of S. cerevisiae to BPFL, BPG, BPM, BPTMC in concentrations starting at 3.6E-05 mol/l. CONCLUSION: Our results suggest multiple parallel interactions of tested compounds and emphasize the importance of determination of an appropriate battery of in vitro methods that will include more receptors and will be appropriate to target specific molecular mechanisms involved in endocrine disruption. Results in agonistic studies indicate agonistic potential and are supported by results of antagonistic studies with consideration of possible multiple interactions.

Alternative methods in vitro for screening of endocrine disruptors.

OBJECTIVES: The aim of this study was to compare in silico data with results obtained in two alternative in vitro methods; and to investigate the potential endocrine activity of bisphenol A analogues. This article contributes to recent findings and brings up-to-date information on development of EU legislation and in vitro testing methods of endocrine disruption. METHODS: In silico approach based on the OECD QSAR Toolbox was used for prediction of potential ligands of human estrogen receptor α. Estrogen Receptor Transactivation in vitro Assay to Detect Estrogen Receptor Agonists and Antagonists (OECD TG 455/457) using the VM7Luc4E2 (formerly designated BG1Luc4E2) cell line was performed for measurement of transactivation activity of the tested substances. Commercially available yeast-based microplate assay (XenoScreen YES/YAS, Xenometrix, Switzerland) for detection of compounds with estrogenic and androgenic agonistic/antagonistic activity was used as a comparative test to estrogen receptor transactivation assay (OECD TG 455/457) and for screening of the agonistic/antagonistic potential of human estrogen receptor and agonistic/antagonistic activity of tested compounds on human androgen receptor. RESULTS: The study showed good correlation between the two in vitro assays and significant correlation with in silico data. All tested substances were identified as agonists for human estrogen receptor α by methods in silico and in vitro, four substances showed a potentially higher estrogenic activity comparing to bisphenol A, two substances were identified as very weak antagonists of human androgen receptor and one compound showed a potential of agonistic activity to human androgen receptor. CONCLUSIONS: The study contributes to recent findings and brings new in silico and in vitro data of bisphenol A analogues, revealing that these analogous substances should be further tested as they may show similar or higher activity in vivo comparing to bisphenol A, which has been recently legislatively regulated.

Estrogen Receptor Binding Affinity of Food Contact Material Components Estimated by QSAR.

AIM: The presented work characterized components of food contact materials (FCM) with potential to bind to estrogen receptor (ER) and cause adverse effects in the human organism. METHODS: The QSAR Toolbox, software application designed to identify and fill toxicological data gaps for chemical hazard assessment, was used. Estrogen receptors are much less of a lock-and-key interaction than highly specific ones. The ER is nonspecific enough to permit binding with a diverse array of chemical structures. There are three primary ER binding subpockets, each with different requirements for hydrogen bonding. RESULTS: More than 900 compounds approved as of FCM components were evaluated for their potential to bind on ER. All evaluated chemicals were subcategorized to five groups with respect to the binding potential to ER: very strong, strong, moderate, weak binder, and no binder to ER. In total 46 compounds were characterized as potential disturbers of estrogen receptor. CONCLUSION: Among the group of selected chemicals, compounds with high and even very high affinity to the ER binding subpockets were found. These compounds may act as gene activators and cause adverse effects in the organism, particularly during pregnancy and breast-feeding. It should be considered to carry out further in vitro or in vivo tests to confirm their potential to disturb the regulation of physiological processes in humans by abnormal ER signaling and subsequently remove these chemicals from the list of approved food contact materials.

Bio-removal of rare earth elements from hazardous industrial waste of CFL bulbs by the extremophile red alga Galdieria sulphuraria.

In recent decades, a shift has been seen in the use of light-emitting diodes over incandescent lights and compact fluorescent lamps (CFL), which eventually led to an increase in wastes of electrical equipment (WEE), especially fluorescent lamps (FLs) and CFL light bulbs. These widely used CFL lights, and their wastes are good sources of rare earth elements (REEs), which are desirable in almost every modern technology. Increased demand for REEs and their irregular supply have exerted pressure on us to seek alternative sources that may fulfill this demand in an eco-friendly manner. Bio-removal of wastes containing REEs, and their recycling may be a solution to this problem and could balance environmental and economic benefits. To address this problem, the current study focuses on the use of the extremophilic red alga, Galdieria sulphuraria, for bioaccumulation/removal of REEs from hazardous industrial wastes of CFL bulbs and the physiological response of a synchronized culture of G. sulphuraria. A CFL acid extract significantly affected growth, photosynthetic pigments, quantum yield, and cell cycle progression of this alga. A synchronous culture was able to efficiently accumulate REEs from a CFL acid extract and efficiency was increased by including two phytohormones, i.e., 6-Benzylaminopurine (BAP - Cytokinin family) and 1-Naphthaleneacetic acid (NAA - Auxin family).

Low-cost and prototype-friendly method for biocompatible encapsulation of implantable electronics with epoxy overmolding, hermetic feedthroughs and P3HT coating.

The research of novel implantable medical devices is one of the most attractive, yet complex areas in the biomedical field. The design and development of sufficiently small devices working in an in vivo environment is challenging but successful encapsulation of such devices is even more so. Industry-standard methods using glass and titanium are too expensive and tedious, and epoxy or silicone encapsulation is prone to water ingress with cable feedthroughs being the most frequent point of failure. This paper describes a universal and straightforward method for reliable encapsulation of circuit boards that achieves ISO10993 compliance. A two-part PVDF mold was machined using a conventional 3-axis machining center. Then, the circuit board with a hermetic feedthrough was placed in the mold and epoxy resin was injected into the mold under pressure to fill the cavity. Finally, the biocompatibility was further enhanced with an inert P3HT polymer coating which can be easily formulated into an ink. The biocompatibility of the encapsulants was assessed according to ISO10993. The endurance of the presented solution compared to silicone potting and epoxy potting was assessed by submersion in phosphate-buffered saline solution at 37 °C. The proposed method showed superior results to PDMS and simple epoxy potting.

New approach methods for assessing indoor air toxicity.

Indoor air is typically a mixture of many chemicals at low concentrations without any adverse health effects alone, but in mixtures they may cause toxicity and risks to human health. The aim of this study was by using new approach methods to assess the potential toxicity of indoor air condensates. In specific, different in vitro test methods including cyto-and immunotoxicity, skin sensitization and endocrine disruption were applied. In addition to biological effects, the indoor air samples were subjected to targeted analysis of 25 volatile organic compounds (VOCs) and Genapol X-80 (a nonionic emulsifier) suspected to be present in the samples, and to a non-targeted "total chemical scan" to find out whether the chemical composition of the samples is associated with the biological effects. The results confirm that assessing health risks of indoor air by analysing individual chemicals is not an adequate approach: We were not able to detect the VOCs and Genapol X-80 in the indoor air samples, yet, several types of toxicity, namely, cytotoxicity, immunotoxicity, skin sensitization and endocrine disruption were detected. In the non-targeted total chemical scan of the indoor air samples, a larger number of compounds were found in the cytotoxic samples than in the non-cytotoxic samples supporting the biological findings. If only one biological method would be selected for the screening of indoor air quality, THP-1 macrophage/WST-1 assay would best fit for the purpose as it is sensitive and serves as a good representative for different sub-toxic end points, including immunotoxicity, (skin) sensitization and endocrine disruption.

Sensitization potential of medical devices detected by in vitro and in vivo methods.

Medical devices must be tested before marketing in accordance with ISO EN 10993-10 in order to avoid skin sensi­tization. This standard predominantly refers to the in vivo test but does not exclude the use of in vitro methods that have been sufficiently technically and scientifically validated for medical device testing. It is foreseen that, due to the complexity of the sensitization endpoint, a combination of several methods will be needed to address all key events occurring in the sensitization process. The objective of this pilot study was to evaluate the sensitization potential of selected medical devices using a combination of in chemico (DPRA, OECD TG 442C) and in vitro (LuSens, OECD TG 442D) methods in comparison with the in vivo (LLNA DA, OECD TG 442A) method and to suggest a possible testing strategy for the safety assessment of medical device extracts. Overall, one of the 42 tested samples exhibited positive results in all employed test methods, while 33 samples were predicted as non-sensitizing in all three performed methods. This study demonstrated good agreement between in vitro and in vivo results regarding non-sensitizing samples; however, some discrepancies in positive classification were recorded. A testing strategy is suggested in which negative results are accepted and any positive results in the in chemico or in vitro tests are followed up with a third in vitro test and evaluated in accordance with the "2 out of 3 approach". This strategy may reduce and replace animal use for testing the sensitization potential of medical devices.

The kinetic direct peptide reactivity assay (kDPRA): Intra- and inter-laboratory reproducibility in a seven-laboratory ring trial

While the skin sensitization hazard of substances can be identified using non-animal methods, the classification of potency into UN GHS sub-categories 1A and 1B remains challenging. The kinetic direct peptide reactivity assay (kDPRA) is a modification of the DPRA wherein the reaction kinetics of a test substance towards a synthetic cysteine-containing peptide are evaluated. For this purpose, several concentrations of the test substance are incubated with the synthetic peptide for several incubation times. The reaction is stopped by addition of monobromobimane, which forms a fluorescent complex with the free cysteine of the model peptide. The relative remaining non-depleted amount of peptide is determined. Kinetic rate constants are derived from the depletion vs concentration and time matrix and used to distinguish between UN GHS sub-category 1A sensitizers and test substances in sub-category 1B/not classified test substances. In this study, we present a ring trial of the kDPRA with 24 blind-coded test substances in seven laboratories. The intra- and inter-laboratory reproducibility were 96% and 88%, respectively (both for differentiating GHS Cat 1A sensitizers from GHS Cat 1B/not classified). Following an independent peer review, the kDPRA was considered to be acceptable for the identification of GHS Cat 1A skin sensitizers. Besides GHS Cat 1A identification, the kDPRA can be used as part of a defined approach(es) with a quantitative data integration procedure for skin sensitization potency assessment. For this aim, next to reproducibility of classification, the quantitative reproducibility and variability of the rate constants were quantified in this study.

Acute toxicity estimation by calculation--Tubifex assay and quantitative structure-activity relationships.

A quantitative structure-activity relationship (QSAR) model dependent on log P(n - octanol/water), or log P(OW), was developed with acute toxicity index EC50, the median effective concentration measured as inhibition of movement of the oligochaeta Tubifex tubifex with 3 min exposure, EC50(Tt) (mol/L): log EC50(Tt) = -0.809 (+/-0.035) log P(OW) - 0.495 (+/-0.060), n=82, r=0.931, r2=0.867, residual standard deviation of the estimate 0.315. A learning series for the QSAR model with the oligochaete contained alkanols, alkenols, and alkynols; saturated and unsaturated aldehydes; aniline and chlorinated anilines; phenol and chlorinated phenols; and esters. Three cross-validation procedures proved the robustness and stability of QSAR models with respect to the chemical structure of compounds tested within a series of compounds used in the learning series. Predictive ability was described by q2 .801 (cross-validated r2; predicted variation estimated with cross-validation) in LSO (leave-a structurally series-out) cross-validation.

The Tubifex tubifex assay for the determination of acute toxicity.

An express (3-minute) test for acute toxicity determination by using the oligochaete annelid, Tubifex tubifex, is described. The EC50(Tubifex tubifex) [EC50(Tt)] for movement inhibition was calculated by using a concentration-response dependence. The reproducibility of the test was checked over several years and by several workers. Its applicability is limited to compounds which are soluble in water. The calculated EC50(Tt) indices correlate with LC50 values determined by using the fish, Pimephales promelas (96-hour assay), and with ICG50 values determined by using the ciliate, Tetrahymena pyriformis (48-hour assay) with high statistical significance (r = 0.822, n = 35, and r = 0.927, n = 80, respectively). The correlation between the EC50(Tt) indices and rat oral LD50 values (48-hour assay) was r = 0.519 (n = 67). The correlation within organic compounds was closer (r = 0.635, n = 60) than with the heterogeneous series of chemicals. A similar trend was noticed for the correlation with mouse oral LD50 values (r = 0.479, n = 56) with the heterogeneous series of chemicals, as compared that with the series without inorganic salts (r = 0.605, n = 42), and similarly with mouse intraperitoneal LD50 values, where r = 0.543 (n = 50) with the heterogeneous series of chemicals and r = 0.893 (n = 33) with the series of organic chemicals.

The glutathione-activated thiopurine prodrugs trans-6-(2-acetylvinylthio)guanine and cis-6-(2-acetylvinylthio)purine cause less in vivo toxicity than 6-thioguanine after single- and multiple-dose regimens.

trans-6-(2-Acetylvinylthio)guanine (trans-AVTG) and cis-6-(2-acetylvinylthio)purine (cis-AVTP) are glutathione-activated prodrugs of 6-thioguanine (6-TG) and 6-mercaptopurine, respectively. In tumor cell lines, these prodrugs exhibit similar IC50 values that are comparable to or lower than those of 6-TG and 6-mercaptopurine, respectively. In this study, the in vivo toxicity and metabolism of the prodrugs were assessed. Mice given multiple treatments of 6-TG and, to a lesser extent, trans-AVTG exhibited decreased peripheral WBC and RBC counts and increased myeloid:erythroid ratios in bone marrow; no change was observed in mice given cis-AVTP. Similarly, intestinal epithelial crypt cell apoptosis was more extensive in mice treated with 6-TG than in those treated with trans-AVTG, whereas mice given cis-AVTP had little apoptosis. Epithelial crypt cell apoptosis was more extensive in the small intestine than in the large intestine in all treatment groups. Histopathological examination detected no kidney or liver toxicity, whereas mild increases in the activities of hepatocellular leakage enzymes were observed in mice treated with trans-AVTG. Only metabolites of trans-AVTG and cis-AVTP were recovered in urine. A higher fraction of the dose was recovered in urine as the parent thiopurine and the metabolites thiopurine riboside, thioxanthine, and thiouric acid after 6-TG treatment than after trans-AVTG treatment; cis-AVTP recovery was slightly less than that of 6-TG. Thioxanthine and thiouric acid comprised a higher fraction of the recovered dose after cis-AVTP treatment than after trans-AVTG or 6-TG treatment. Overall, the results suggest that the prodrugs exhibit less in vivo toxicity than 6-TG. Thus, investigations into their antitumor efficacy are warranted.

Glutathione peroxidase activity in the selenium-treated alga Scenedesmus quadricauda.

The function of selenium in an organism is mediated mostly by selenoproteins including glutathione peroxidase. Glutathione peroxidase is a potent anti-oxidative enzyme, scavenging a variety of peroxides. The green alga Scenedesmus quadricauda was used to investigate the relationship between the toxicity of selenium and the glutathione peroxidase activity. Selenium resistant strains SeIV and SeVI were synchronized and grown in high concentrations of Se (selenite or selenate). As a measure of selenium toxicity the EC(50) values were determined. During growth of the untreated wild type, glutathione peroxidase activity increased slightly and then declined gradually until the end of the cell cycle. A similar pattern was observed in untreated resistant strains and when resistant strains were grown in the presence of selenium in the oxidation state to which they were resistant. In the wild type cultivated with 50 mg Se L(-1) (selenite or selenate), activity increased to a high level and slowly declined until the end of the cell cycle. Similarly, activity increased in strains SeIV and SeVI when grown in the presence of selenium in the oxidation state to which they were not resistant. We followed the effect of selenium on the ultrastructure of S. quadricauda. After exposure to selenite, the chloroplast membranes of wild type were reorganized into thick bundles of thylakoids and the stroma became granulose. When selenate was added, the chloroplast of wild type had a fingerprint-like appearance, the stroma became less dense and starch production increased. In selenium resistant strains, when treated with the selenium form to which they were resistant, the chloroplast was affected, but not to such an extent as in the wild type. The activity of glutathione peroxidase in Scenedesmus was affected by selenium in an oxidation state-dependent manner. The most apparent effects of selenium on the ultrastructure involved impairment of the chloroplast and the overproduction of starch.

Toxicity of perfluorinated carboxylic acids for aquatic organisms.

Toxicity of perfluorinated carboxylic acids with carbon chain C(8) to C(12) were tested with oligochaeta Tubifex tubifex. Toxicity was evaluated as the exposure time ET(50) from onset of damage of the oligochaeta in saturated aqueous solutions. The ET(50) fluctuated between 25 and 257 minutes. No statistically significant difference was found among the C(8), C(9) and C(12) acids (ET(50) between 143 and 257 minutes with large standard deviation). The acids with carbon chain C(10) and C(11) induced the effect significantly quicker (25 to 47 minutes). No acute toxicity measured in the three-minute test was observed in any case.

Extrapolation of toxic indices among test objects.

Oligochaeta Tubifex tubifex, fish fathead minnow (Pimephales promelas), hepatocytes isolated from rat liver and ciliated protozoan are absolutely different organisms and yet their acute toxicity indices correlate. Correlation equations for special effects were developed for a large heterogeneous series of compounds (QSAR, quantitative structure-activity relationships). Knowing those correlation equations and their statistic evaluation, one can extrapolate the toxic indices. The reason is that a common physicochemical property governs the biological effect, namely the partition coefficient between two unmissible phases, simulated generally by n-octanol and water. This may mean that the transport of chemicals towards a target is responsible for the magnitude of the effect, rather than reactivity, as one would assume suppose.

Validation of QSAR models for legislative purposes.

OECD principles of validation of Quantitative Structure - Activity Relationships (QSAR) models for legislative purposes are given and explained. Reasons of their origination and development, like system REACH, are described. A basic impulse has come from some OECD countries followed by all (almost) other countries of the world.

Novel glutathione-dependent thiopurine prodrugs: evidence for enhanced cytotoxicity in tumor cells and for decreased bone marrow toxicity in mice.

Elevated glutathione (GSH) levels have been detected in many tumors compared with the healthy, surrounding tissue. Often, this GSH up-regulation is associated with drug resistance. The prodrugs 6-(2-acetylvinylthio)guanine (AVTG) and 6-(2-acetylvinylthio)purine (AVTP) contain a novel butenone moiety that allows the prodrugs to react selectively with sulfhydryl nucleophiles to release the chemotherapeutic drug 6-thioguanine (6-TG) or 6-mercaptopurine (6-MP), respectively. The cellular uptake and metabolism of trans-AVTG in two human renal carcinoma cell lines that were used as models were rapid and associated with depletion of intracellular GSH. Formation of 6-TG from trans-AVTG correlated positively with intracellular GSH concentrations, and was significantly reduced by diethyl maleate pretreatment. Intracellular concentrations of 6-TG after incubations with trans-AVTG were significantly higher than the 6-TG concentrations obtained after incubations with equimolar concentrations of 6-TG; thus, the prodrug delivered more 6-TG to the cell than did 6-TG itself. Cytotoxicity studies demonstrated that AVTG and AVTP had similar IC(50) values that were comparable with those of 6-TG, but were significantly lower than those of 6-MP. Furthermore, after in vivo treatment of mice with the prodrugs, no reduction was observed in circulating white blood cell counts, whereas white blood cell counts of mice treated with equimolar or 60% lower doses of 6-TG were reduced by 50 to 60%. Collectively, the results show that AVTG and AVTP are novel potential chemotherapeutic agents that may target tumors with up-regulated levels of GSH, and may exhibit less systemic toxicity than the parent thiopurines.

Risk assessment of mixtures: possibility of prediction of interaction between chemicals.

OBJECTIVES: Potentiation, inhibition, antagonism, synergism or other interactions of chemicals in an organism may significantly affect the toxicity of chemical mixtures. Therefore, a suitable physicochemical parameter is sought as a predictor of these effects. METHODS: Acute toxicity EC50 (mol/l) was used as an index of toxic hazard. The EC50 was defined here as the concentration inhibiting movement of 50% of Tubifex tubifex worms in a standardised short-term test. The EC50 for the mixture benzene/ethanol was measured over the whole range of molar fractions between 0 and 1. The determined EC50 values were compared with those calculated by the effect-addition and concentration-addition models. The distribution of benzene and ethanol between the aqueous and gaseous phases was determined by gas chromatography. RESULTS: The determined EC50 of the benzene/ethanol mixtures exhibited deviations from the values predicted using the EC50 of the single compounds. The magnitude of this deviation depended on the composition of the mixture, and could be as high as 80% if the effect-addition model is used or 20% if the concentration-addition model is applied. The air/water distribution coefficients of benzene and ethanol measured in aqueous solutions of their mixtures exhibit unobservable to slight deviations from those obtained by simple summation. CONCLUSION: The determined values of the EC50 of benzene/ethanol mixtures calculated by either the concentration-addition or effect-addition model differ from the true values due to interactions among chemicals. The deviations depend on a quantitative composition of a mixture. The air/water distribution coefficient of chemicals in a mixture is not the most suitable predictor of the interactions.