Superfund Research Program

Development and Application of Integrated In Vitro and Cell-Based Bioassays

Project Leader: Michael S. Denison
Co-Investigator: Isaac N. Pessah
Grant Number: P42ES004699
Funding Period: 1995-2015

Project-Specific Links

Final Progress Reports

Year:   2014  2009  2004  1999 

The overall goal of this project is the development and validation of bioassay systems for the detection of toxicants and development of biomarkers of toxicant exposure and effect. Project investigators have made significant progress in several areas and these advances are highlighted below.

Taking advantage of the ligand-dependence of nuclear receptors to activate gene expression, we have continued development and validation of recombinant cell bioassay systems for toxicant detection. Researchers have optimized the previously described chemically-activated luciferase expression (CALUX) cell bioassay to now allow direct detection of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD, dioxin) and related chemicals in whole serum samples. The CALUX bioassay allows rapid (four hours) and direct analysis of small volumes (25 to 50 µl) of whole serum in a 96-well microtiter plate format, without the need for solvent extraction. This bioassay can detect as little as 10 parts per trillion of TCDD, and is also sensitive to other HAHs and PAHs. The use of simple procedures correct for inter- and intra-plate variability and the Ah receptor dependence of the induction response is accounted for by use of the antagonist 4-amino-3-methoxyflavone. This modification of the assay has application for large-scale epidemiological screening of blood samples for the presence of these chemicals. Researchers have also spent considerable time examining the false positives and false negatives which can arise with the CALUX assay and an in vitro bioassay (Gel Retardation Analysis) that has been developed. Results clearly indicate that because of the high false positive rates, in vitro bioassays that are currently available for the detection of dioxin-like chemicals are inaccurate and unsuitable for screening for these chemicals. Finally, using techniques developed for the generation of the CALUX bioassay, project investigators have generated an analogous recombinant cell bioassay for the detection of estrogenic and antiestrogenic chemicals. Human ovarian carcinoma cells have been stably transfected with the estrogen responsive luciferase reporter plasmid. The resulting recombinant cell line responds to 17b-estradiol at concentrations as low as 1 ppM. The utility of BG1Luc4E2 cells as a bioassay screening system for environmental estrogens was demonstrated both by their response to known xenoestrogens, and also by the putative identification of two polychlorinated biphenyls (2,3',4,4'-tetrachlorobiphenyl and 2,2',3,5',6-pentachlorobiphenyl) as novel estrogenic chemicals. This new cell bioassay system has applications for rapid screening, identification, and characterization of chemicals that disrupt normal estrogen signaling pathways.

Knowledge of the biochemistry, regulatory biology and structure of the soluble epoxide hydrolase has expanded greatly based in part on the first solution of a crystal structure of this enzyme. These studies demonstrated that a group of alkyl and aryl ureas were potent transition state mimics of the enzyme binding tightly to the catalytic side. Some of these compounds have Ki's as low as 5 nanomolar. Researchers have found that a group of widely used herbicides and insecticides are inhibitors of the recombinant murine and human enzyme, and the effects of these compounds on the developmental biology of several fish species have been examined because they are known aquatic contaminants. Epoxides of arachidonic acid are well known chemical mediators and are excellent substrates for the soluble epoxide hydrolase. Thus, urea inhibitors alter the levels of these regulatory lipids and thus alter the biology of rodents. Project investigators have found that a variety of epoxides and diols of fatty acids, including the most common dietary fat, linoleate, have biological activity in vitro. These compounds seem to act to induce apoptosis and necrosis at the mitochondrial level. Analytical methods have now been developed for the most active of these epoxylipids and are being applied to rodent models exposed to environmental chemicals and to human urine. Researchers have found that the epoxides are made in human lymphocytes by CYP2C9 and have found biological activity associated with these lipids in in vivo rodent assays and with the human Jurkatt and HL-60 lines. This enables the investigators to see what messages are induced by administration of bioactive lipid epoxides and diols and how these effects are modulated by induction and inhibition of the soluble epoxide hydrolase by xenobiotics.

It has been previously reported that arsenic treatment of human keratinocytes in culture suppresses expression of a battery of differentiation markers (including involucrin) at concentrations where the cells still grow well. To explore the hypothesis that arsenic acts, in the case of involucrin, by preventing transcription factor utilization of critical response elements in the promoter, regulatory regions in the promoter have been characterized. In the process, a novel response element important for high level expression has been identified whose activity appears restricted to keratinocytes. The responsiveness of this element to arsenic will be tested. Another gene whose expression we have now shown to be suppressed by arsenic is keratinocyte transglutaminase. Researchers have identified two critical response elements in the promoter (AP1 and CRE) that, in initial experiments, appear in part to mediate arsenic sensitivity. Meanwhile the ability of TCDD to prevent retinoid induction of tissue transglutaminase has been characterized. Contrary to expectation, TCDD did not act by enhancing the depletion of active retinoid from the culture medium or by preventing retinoid receptor action. Rather, it appeared to act indirectly by interfering with a late phase of retinoid induction of the gene. Finally, the sensitivity of keratinocytes to the carcinogenic food mutagen Trp-P-2 has been examined. In contrast to Trp-P-1, from which it differs by a single methyl group, this agent is nontoxic to human epidermal cells despite being more mutagenic than Trp-P-1. The mutagenicity of both compounds is greatly stimulated by TCDD. Rat epidermal cells are sensitive to toxicity from Trp-P-2, possibly because they express an order of magnitude lower glutathione S-transferase activity than the human cells. These experiments show that keratinocyte cultures are promising models to study the complex relationships among toxicity, mutagenicity and carcinogenicity.