Contribution of new technologies to characterization and prediction of adverse effects.

Identification of the potential hazards of chemicals has traditionally relied on studies in laboratory animals where changes in clinical pathology and histopathology compared to untreated controls defined an adverse effect. In the past decades, increased consistency in the definition of adversity with chemically-induced effects in laboratory animals, as well as in the assessment of human relevance has been reached. More recently, a paradigm shift in toxicity testing has been proposed, mainly driven by concerns over animal welfare but also thanks to the development of new methods. Currently, in vitro approaches, toxicogenomic technologies and computational tools, are available to provide mechanistic insight in toxicological Mode of Action (MOA) of the adverse effects observed in laboratory animals. The vision described as Tox21c (Toxicity Testing in the 21st century) aims at predicting in vivo toxicity using a bottom-up-approach, starting with understanding of MOA based on in vitro data to ultimately predict adverse effects in humans. At present, a practical application of the Tox21c vision is still far away. While moving towards toxicity prediction based on in vitro data, a stepwise reduction of in vivo testing is foreseen by combining in vitro with in vivo tests. Furthermore, newly developed methods will also be increasingly applied, in conjunction with established methods in order to gain trust in these new methods. This confidence is based on a critical scientific prerequisite: the establishment of a causal link between data obtained with new technologies and adverse effects manifested in repeated-dose in vivo toxicity studies. It is proposed to apply the principles described in the WHO/IPCS framework of MOA to obtain this link. Finally, an international database of known MOAs obtained in laboratory animals using data-rich chemicals will facilitate regulatory acceptance and could further help in the validation of the toxicity pathway and adverse outcome pathway concepts.

Toxicogenomics in vitro as an alternative tool for safety evaluation of petroleum substances and PAHs with regard to prenatal developmental toxicity.

The REACH legislation requires chemicals - including petroleum substances - that are put on the EU market in quantities greater than 1000 tonnes/year, to be tested for prenatal developmental toxicity. This will require large numbers of animals since prenatal development toxicity testing is animal-intensive. The application of toxicogenomic technologies might reduce the number of animals to study prenatal developmental toxicity of petroleum substances by allowing their grouping into categories with the same toxicological properties. This substance categorization may be supported by similarities in molecular fingerprints. The developmental toxicity effects observed in some oil products are most likely related to polycyclic aromatic hydrocarbons (PAHs) with high-molecular weight. However, the current review indicates that even though the available studies provide clues regarding the HOX, FOX, SHH and PAX family genes, which regulate functions in skeleton development, single individual genes cannot be used as biomarkers of PAHs exposure and subsequent prenatal developmental toxicity. Furthermore, it should be considered that toxicogenomic technologies applied to specific tissues/organs testing might lead to unreliable results regarding developmental toxicity due to induction of tissue-specific pathways. Thus, an approach which applies a battery of in vitro tests including the zebrafish embryo test, embryonic stem cells, and the whole embryo culture is suggested as it would be more relevant for studying developmental effects in the terms of substances categorization.

The use of biomarkers of toxicity for integrating in vitro hazard estimates into risk assessment for humans.

The role that in vitro systems can play in toxicological risk assessment is determined by the appropriateness of the chosen methods, with respect to the way in which in vitro data can be extrapolated to the in vivo situation. This report presents the results of a workshop aimed at better defining the use of in vitro-derived biomarkers of toxicity (BoT) and determining the place these data can have in human risk assessment. As a result, a conceptual framework is presented for the incorporation of in vitro-derived toxicity data into the risk assessment process. The selection of BoT takes into account that they need to distinguish adverse and adaptive changes in cells. The framework defines the place of in vitro systems in the context of data on exposure, structural and physico-chemical properties, and toxicodynamic and biokinetic modeling. It outlines the determination of a proper point-of-departure (PoD) for in vitro-in vivo extrapolation, allowing implementation in risk assessment procedures. A BoT will need to take into account both the dynamics and the kinetics of the compound in the in vitro systems. For the implementation of the proposed framework it will be necessary to collect and collate data from existing literature and new in vitro test systems, as well as to categorize biomarkers of toxicity and their relation to pathways-of-toxicity. Moreover, data selection and integration need to be driven by their usefulness in a quantitative in vitro-in vivo extrapolation (QIVIVE).

Dermal penetration of propylene glycols: measured absorption across human abdominal skin in vitro and comparison with a QSAR model.

The dermal penetration of undiluted monopropylene glycol (MPG) and dipropylene glycol (DPG) has been measured in vitro using human abdominal skin under conditions of infinite dose application, and the results compared with predictions from the SKINPERM QSAR model (ten Berge, 2009). The measured steady-state penetration rates (Jss) for MPG and DPG were 97.6 and 39.3 µg/cm2/h, respectively, and the permeability coefficients (Kp) were 9.48×10(-5) cm/h for MPG and 3.85×10(-5) cm/h for DPG. In comparison, the SKINPERM model slightly over-predicted Jss and Kp for MPG and DPG by between 2.6- and 5.1-fold, respectively. The model predictions of 254 µg/cm2/h and 24.6×10(-5) cm/h for MPG, and 202 µg/cm2/h and 19.8×10(-5) cm/h for DPG were in fairly good agreement with the measured values. Further, the model predicted a Jss of 101 µg/cm2/h and a Kp of 9.9×10(-5) cm/h for the homologue tripropylene glycol. Assuming that the measured Jss was the same under conditions of finite dose application (taken to be 10 µL/ cm2) and was maintained over a 24-h period (both conservative assumptions), the relative dermal absorption of the applied dose was estimated to be 23% (0.96%/h) for MPG and 9% (0.39%/h) for DPG. However, the extrapolation for MPG may be further overestimated due to possible residence in the stratum corneum under infinite conditions of exposure that would not be applicable to a finite loading dose.

Subacute effects of hexabromocyclododecane (HBCD) on hepatic gene expression profiles in rats.

Hexabromoyclododecane (HBCD), used as flame retardant (FR) mainly in textile industry and in polystyrene foam manufacture, has been identified as a contaminant at levels comparable to other brominated FRs (BFRs). HBCD levels in biota are increasing slowly and seem to reflect the local market demand. The toxicological database of HBCD is too limited to perform at present a solid risk assessment, combining data from exposure and effect studies. In order to fill in some gaps, a 28-day HBCD repeated dose study (OECD407) was done in Wistar rats. In the present work liver tissues from these animals were used for gene expression profile analysis. Results show clear gender specificity with females having a higher number of regulated genes and therefore being more sensitive to HBCD than males. Several specific pathways were found to be affected by HBCD exposure, like PPAR-mediated regulation of lipid metabolism, triacylglycerol metabolism, cholesterol biosynthesis, and phase I and II pathways. These results were corroborated with quantitative RT-PCR analysis. Cholesterol biosynthesis and lipid metabolism were especially down-regulated in females. Genes involved in phase I and II metabolism were up-regulated predominantly in males, which could explain the observed lower HBCD hepatic disposition in male rats in this 28-day study. These sex-specific differences in gene expression profiles could also underlie sex-specific differences in toxicity (e.g. decreased thyroid hormone or increased serum cholesterol levels). To our knowledge, this is the fist study that describes the changes in rat hepatic gene profiles caused by this commonly used flame retardant.

Estrogenic effects in the immature rat uterus after dietary exposure to ethinylestradiol and zearalenone using a systems biology approach.

Residues of illegally used hormones are regularly detected in animal products, feed, or cocktails recovered at farms. In order to better understand the effects of dietary exposure to ethinyl estradiol (EE2, 0.03-1 microg/kg body weight [bw]) and zearalenone (ZEA, 0.03-1 mg/kg bw), an immature rat uterotrophic assay was performed and effects were studied at morphological, histological, and gene expression levels. Ligand-mediated coregulator recruitment by estrogen receptor alpha (ERalpha) was studied in vitro. Uterine weight and epithelial cell height were increased dose dependently after a 3-day oral exposure of rats to the highest tested doses of EE2 or ZEA, respectively. At low doses 0.03 microg/kg EE2 and 0.1 mg/kg ZEA, edema, and vacuolization could already be observed in some animals. Exposure to 1 mg/kg ZEA resulted in severe damage of the uterine epithelial layer. Our study suggests similar coregulator recruitment and gene expression patterns for the two estrogenic compounds. Main regulated pathways were remodeling of extracellular matrix, alternative complement activation, cell proliferation, and estrogen-mediated calcium signaling. The level of regulation differed between EE2 and ZEA, attributing a much lower estrogenic potency to ZEA than to EE2. A major difference was their ability to recruit coregulator inhibitor of kappa B beta and induce expression of the matrix metalloproteinase 7 gene (381.4- and 6.9-fold upregulation by EE2 and ZEA, respectively), which plays an important role in the maintenance of the integrity of the epithelial layer of the uterus during proliferation and growth. This observation may explain the observed differences at the histological level.

Additive estrogenic effects of mixtures of frequently used UV filters on pS2-gene transcription in MCF-7 cells.

In order to protect consumers from ultraviolet (UV) radiation and enhance light stability of the product, three to eight UV filters are usually added to consumer sunscreen products. High lipophilicity of the UV filters has been shown to cause bioaccumulation in fish and humans, leading to environmental levels of UV filters that are similar to those of PCBs and DDT. In this paper, estrogen-regulated pS2 gene transcription in the human mammary tumor cell line MCF-7 was used as a measure of estrogenicity of four individual UV filters. Since humans are exposed to more than one UV filter at a time, an equipotent binary mixture of 2-hydroxy-4-methoxy-benzophenone (BP-3) and its metabolite 2,4-dihydroxy benzophenone (BP-1), as well as an equipotent multi-component mixture of BP-1, BP-3, octyl methoxy cinnamate (OMC) and 3-(4-methylbenzylidene) camphor (4-MBC), were also evaluated for their ability to induce pS2 gene transcription in order to examine additivity. An estrogen receptor-mediated mechanism of action was expected for all UV filters. Therefore, our null-hypothesis was that combined estrogenic responses, measured as increased pS2 gene transcription in MCF-7 cells after exposure to mixtures of UV filters, are additive, according to a concentration-addition model. Not all UV filters produced a full concentration-response curve within the concentration range tested (100 nM-1 microM). Therefore, instead of using EC50 values for comparison, the concentration at which each compound caused a 50% increase of basal pS2 gene transcription was defined as the C50 value for that compound and used to calculate relative potencies. For comparison, the EC50 value of a compound is the concentration at which the compound elicits an effect that is 50% of its maximal effect. Individual UV filters increased pS2 gene transcription concentration-dependently with C50 values of 0.12 microM, 0.5 microM, 1.9 microM, and 1.0 microM for BP-1, BP-3, 4-MBC and OMC, respectively. Estradiol (E2) had a C50 value of 4.8 pM. Experiments with equipotent mixtures all supported our null hypothesis that mixtures of UV filters act additively to activate the estrogen receptor (ER). In view of our results and observed plasma levels it cannot be excluded that daily exposure to sunscreen formulations may have estrogenic effects in humans.

Inhibition of aromatase activity by methyl sulfonyl PCB metabolites in primary culture of human mammary fibroblasts.

In this study, the effects on catalytic activity and mRNA levels of aromatase in primary human mammary fibroblasts were evaluated after exposure to promoter-specific modulators of aromatase expression and methyl sulfonyl polychlorinated biphenyl metabolites (MeSO(2)-PCBs). A method for fibroblast isolation from primary breast tissue was developed and optimized, and aromatase activity and promoter-specific mRNA levels were assessed in these cells after exposure to test compounds. A 24-h exposure of fibroblasts to dexamethasone (DEX) (1-100 nM) increased aromatase activity to a maximum of 313-fold. DEX also elevated promoter I.4-specific RNA levels. A 24-h exposure of fibroblasts to 3-MeSO(2)-PCB-132, 4-MeSO(2)-PCB-132, 4-MeSO(2)-PCB-91, or 4-MeSO(2)-PCB-149 (0.1-10 microM) resulted in a concentration-dependent decrease of aromatase activity. Exposure of fibroblasts to MeSO(2)-PCBs just for the limited duration (6 h) of the catalytic assay caused a concentration-dependent inhibition of aromatase enzyme activity. mRNA levels were not altered by a 24-h MeSO(2)-PCB exposure nor was cytotoxicity observed. In aromatase-expressing human adrenocortical carcinoma H295R cells, a 24-h exposure to 3-MeSO(2)-PCB-132, 4-MeSO(2)-PCB-132, 4-MeSO(2)-PCB-91, or 4-MeSO(2)-PCB-149 (0.1-10 microM) also resulted in a concentration-dependent decrease of aromatase activity. Additionally, there were no changes in aromatase mRNA levels after 24-h exposure of H295R cells to MeSO(2)-PCBs. We conclude that in primary human mammary fibroblasts as well as in H295R cells, aromatase inhibition by MeSO(2)-PCBs is likely to be due to catalytic inhibition.

Co-culture of primary human mammary fibroblasts and MCF-7 cells as an in vitro breast cancer model.

Approximately 60% of all breast tumors are estrogen-responsive and chemicals that show estrogenic or anti-estrogenic properties are able to interact with breast tumor growth. In a breast tumor, adipose stromal cells (fibroblasts) surrounding the epithelial tumor contain the aromatase enzyme, which converts androgens into estrogens. Exposure to aromatase inducers can therefore lead to increased estrogen levels and possibly to accelerated breast tumor growth. Subsequently, breast tumor cells synthesize and secrete elevated levels of factors such as prostaglandin E2 (PGE2), interleukin-6 (IL-6), and IL-6 soluble receptor (IL-6sR), which in turn have the ability to stimulate aromatase gene transcription in fibroblasts, establishing a positive feedback loop. In this study, a technique that allows for culturing MCF-7 epithelial breast tumor cells and healthy primary human mammary fibroblasts together in one compartment was developed. To establish the positive feedback loop, the co-culture was exposed to estrogenic compounds. RNA was isolated and reverse-transcriptase polymerase chain reaction (RT-PCR) was performed on the aromatase and pS2 genes. Exposure of the co-culture to estradiol (E2), diethylstilbestrol (DES), and bisphenol-A (BPA), resulted in a three- to seven-fold increase of pS2 transcription levels. Furthermore, pS2 transcription levels increased even more when the aromatase substrate testosterone (20 nM) was present in the co-culture medium. Exposure of the co-culture to the aromatase inducer dexamethasone (DEX) resulted in increased pS2 transcription levels, as well as increased aromatase transcription levels. Simultaneous exposure to DEX and the synthetic anti-estrogen ICI 182,780 almost completely blocked the pS2 response. The aromatase induction response was not altered by ICI 182,780 treatment. Simultaneous exposure to DEX and the non-steroidal aromatase inhibitor fadrozole, abolished the effect of the presence of testosterone in the co-culture medium, but did not result in pS2 gene transcription levels as low as seen after exposure to ICI 182,780. These observations indicate the presence of a positive feedback loop in our co-culture system. This co-culture provides a more sophisticated and sensitive system to detect direct and indirect estrogenic effects of compounds and their possible effects on breast tumor promotion.

A comparison of human H295R and rat R2C cell lines as in vitro screening tools for effects on aromatase.

In this study we evaluated the rat Leydig cell carcinoma cell line R2C and the human adrenocorticocarcinoma cell line H295R for their suitability as in vitro screening tools for potential interference of xenobiotics with aromatase activity. A 24h exposure to prochloraz (PRO), fadrozole (FAD) and epoxyconazole (EPO) resulted in complete catalytic inhibition in H295R and R2C cells. In H295R cells, PRO and FAD were mixed-type inhibitors with apparent K(i) values of 0.04 and 0.03 microM, and apparent K(i)' values of 0.33 and 0.06 microM, respectively. EPO was a competitive inhibitor with an apparent K(i) value of 0.51microM. In R2C, all three compounds showed mixed type inhibition kinetics, with apparent K(i) values (microM) of 0.004, 0.003 and 0.07, and apparent K(i)' values (microM) of 0.41, 0.01 and 2.42, respectively. Exposure for 24h of H295R cells to 8-bromo-cyclic adenosine monophosphate (8-Br-cAMP), prostaglandin E2 (PGE2), phorbol 12-myristate 13-acetate (PMA), or dexamethasone (DEX) resulted in 4.0, 2.8, 3.6 or 3.6-fold induction of aromatase activity, respectively, as well as in an increase of several human aromatase transcripts with promoter-specific 5'-ends (pII and I.3). In R2C cells, only PMA slightly induced aromatase activity. A 24h exposure of H295R cells to atrazine (ATR), resulted in a three-fold induction of aromatase activity and a slight increase in pII and I.3 aromatase transcripts. However, ATR did not induce aromatase activity in R2C cells. We conclude that the H295R cell line contains aromatase promoter regions, which are responsive to the respective stimulants. They play a role in aromatase regulation in various tissues such as brain, placenta, healthy and diseased gonadal and breast tissue and therefore, they may play an important role in tumor genesis, development, behavior and reproduction. The H295R cell line may therefore be a relevant and useful tool in risk assessment of xenobiotics. The R2C cell line, although not suitable for studying induction, appears to be a more sensitive cell line for studying inhibitory effects of xenobiotics on aromatase activity.