Development and Modification of a Recombinant Cell Bioassay to Directly Detect Halogenated and Polycyclic Aromatic Hydrocarbons in Serum

Michael H. Ziccardi*,{dagger},1, Ian A. Gardner{dagger} and Michael S. Denison{ddagger},2

* Wildlife Health Center and {dagger} Department of Medicine and Epidemiology, School of Veterinary Medicine, University of California, Davis, California 95616; and {ddagger} Department of Environmental Toxicology, University of California, Davis, California 95616

Received June 22, 1999; accepted November 11, 1999


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Polycyclic and halogenated aromatic hydrocarbons (PAHs/HAHs) are a diverse group of widespread and persistent environmental contaminants that can cause a variety of detrimental effects in vertebrates. As most available methods to detect these contaminants are expensive, labor and time intensive, and require large amounts of tissue for extraction and analysis, several rapid mechanistically based bioassay systems have been developed to detect these chemicals. Here we describe application and optimization of a recently developed recombinant mouse cell bioassay system that responds to both PAHs and HAHs with the rapid induction of firefly luciferase for the detection of these chemicals in whole serum samples. This chemically activated luciferase expression (CALUX) bioassay has been modified to allow rapid (4-h) and direct analysis of small volumes (25–50 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 the most potent HAH, 2,3,7,8-TCDD, and is also sensitive to other HAHs and PAHs. The use of simple procedures corrects for interplate and intraplate variability and the Ah receptor dependence of the induction response is accounted for by use of the antagonist 4-amino-3-methoxyflavone.

Key Words: HAH; PAH; Ah receptor; bioassay; blood; CALUX; TCDD.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Halogenated and nonhalogenated polycyclic aromatic hydrocarbons (HAHs/PAHs) such as polychlorinated dibenzo-p-dioxins and biphenyls, benzo(a)pyrene, and related chemicals have long been recognized as significant and widespread pollutants in the environment. In mammalian systems, exposure to such xenobiotics has produced a variety of species- and tissue-specific toxic and biologic effects, the majority of which are mediated by the Ah receptor (AhR) (Denison et al., 1998aGo; Nelson et al., 1993Go; Whitlock, 1990Go). Because of the potential for high morbidity and mortality associated with exposure to these compounds, many analytical techniques have been developed to detect their presence both in biologic and environmental samples. These procedures often rely on very accurate methods, such as gas chromatography coupled with mass spectroscopy (GC/MS), high pressure liquid chromatography, or other methods; however, their use also has many limitations, such as high analysis costs, lack of rapidity, and the need for large sample volumes. All of these factors negatively affect their utility for rapid screening and analysis of large numbers of samples. Although the application of toxic equivalency factors (TEFs) has been used to estimate the relative biologic and toxicologic potency of a given mixture of HAHs (specifically that of polychlorinated dibenzo-p-dioxins, dibenzofurans and biphenyls) (Murk et al., 1996aGo,bGo; Safe et al., 1989Go; Sanderson et al., 1996Go; Tillitt et al., 1991Go), the usefulness of these estimates is adversely affected by the presence of other chemicals when other complex mixtures are present. Therefore, although often these are the most accurate and precise methods available, they have limitations for assessing both the presence and effects of HAHs/PAHs in biologic systems.

In order to assess biologic responses of such compounds in complex systems better, much research effort has been directed toward the development of bioassays. Two classes of bioassays based on the AhR-dependent induction of CYP1A1 and its monoxygenase activity have been developed to detect animal exposure to PAHs and/or HAHs: those relying on in vivo metabolism in animals, and those measuring the induction of CYP systems in vitro. The first class, which includes techniques that examine in vivo responses in tissues [such as measurement of CYP1A1 induction by the quantification of ethoxyresorufin-o-deethylase (EROD) activity, aryl hydrocarbon hydroxylase (AHH) activity, or CYP1A1 mRNA levels], have been used widely to determine exposure of individuals (Burke and Mayer, 1974Go; Nebert and Gelboin, 1968Go; Vanden Heuval et al., 1993Go). However, this approach has several disadvantages that limit their widespread applicability as screening procedures, including species differences in response to HAHs/PAHs (Aarts et al., 1996Go; Denison et al., 1995Go; Denison and Wilkinson, 1985Go). Additionally, the methods used to assess this type of activation often require either the euthanasia of the animal or the use of invasive surgical techniques for sample collection. More important, many CYP1A1-inducing chemicals (AhR ligands) are also substrates for this activity, thus resulting in competitive inhibition at high concentrations (Hahn et al., 1993Go; Kennedy et al., 1993Go), which would result in an underestimate of xenobiotic exposure.

Another class of bioassay, one that uses in vitro methods (through detection of enzymatic activity, Ah receptor binding, or the induction of reporter genes), has recently become more frequently used to evaluate exposure of animals or environments to HAHs/PAHs. The H4IIE rat hepatoma cell bioassay (Bradlaw and Casterline, 1979Go) has been used to detect HAHs/PAHs and provides an estimate of the biologic potency (expressed as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) equivalents or TEQs) of compounds or complex mixtures (Tillitt et al., 1991Go; Willett et al., 1997Go). However, because this method also measures increases in EROD activity, it has many of the same limitations described above. The application of recombinant reporter plasmids such as the firefly luciferase gene (Denison et al., 1993Go; Garrison et al., 1996Go; Postlind et al., 1993Go; Richter et al., 1997Go) has proven to be a very effective method to detect these chemicals and to provide an approach to determine TEQs in a variety of matrices (Aarts et al., 1993Go; Anderson et al., 1995Go; Murk et al., 1996). The chemically activated luciferase expression (CALUX) bioassay system is extremely sensitive in directly detecting AhR-dependent potential of a variety of pure compounds and extracts of environmental and biologic matrices (Aarts et al., 1996Go; Denison et al., 1996Go, 1998bGo; Heath-Pagliuso et al., 1998Go; Murk et al., 1996aGo,bGo; Phelan et al., 1998Go; Sanderson et al., 1996Go; Ziccardi et al., 1997Go).

Epidemiologic studies of the effects of HAHs and PAHs on human and wildlife populations require large-scale assessment of the level of exposure to these compounds and current analytical procedures are not amenable to large-scale screening procedures. Given the advantages of the CALUX assay, its modification to a high throughput screening system would have utility in these types of analyses. In order to maximize the utility of this method for use in these situations, however, analysis of readily accessible samples must be validated. In experimental studies, it has been noted that both PAHs and HAHs can be found in serum fractions at ng/g (or parts per billion) levels following acute exposure in laboratory animals (Abraham et al., 1988Go; Bartosek, 1984; Patterson et al., 1989Go), thereby being a potentially valuable biologic matrix to analyze. Here we describe studies in which the CALUX bioassay system has been optimized for use in a 96-well microtiter plate format for the direct detection of HAHs and PAHs in small volumes of whole serum.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Chemicals and samples.
TCDD, 1,2,4,7,8-pentachlorodibenzo-p-dioxin (PeCDD), 2,3,7,8-tetrachlorodibenzofuran (TCDF), 2,3,4,7,8-pentachlorodibenzofuran (PeCDF), 6-methyl-1,3,8-trichlorodibenzofuran (MCDF), 4-amino-3-methoxyflavone (AMF), and 3-methoxy-4-nitroflavone (MNF) were obtained from Dr. S. Safe (Texas A&M University). 3,3',4,4'-Tetrachlorobiphenyl (TCB), 2,3,4,4',5-pentachlorobiphenyl (2-PeCB), 3,3',4,4',5-pentachlorobiphenyl (3-PeCB), 2,3,3',4,4',5-hexachlorobiphenyl (HCB), Aroclor 1242, Aroclor 1254, Aroclor 1260, benzo(k)fluoranthene (BKF), benzo(a)pyrene (BAP), and benz(a)anthracene (BAA) were purchased from Accustandards Co. (St. Louis, MO). Alpha-naphthoflavone (ANF) was obtained from Sigma Chemical Co. (St. Louis, MO) and fluorescamine from Molecular Probes (Eugene, OR). All HAH/PAH compounds were considered extremely hazardous and appropriate personal protective methods and materials were used throughout all experiments.

Recombinant cell line.
Mouse hepatoma (H1L1.1c2) cells, which were stably transfected with the PAH/HAH-inducible luciferase expression vector pGudLuc1.1, were described previously (Garrison et al., 1996Go). This vector contains the firefly luciferase gene under PAH/HAH-inducible control of four dioxin responsive elements (DREs) and induction of luciferase occurs in a time-, dose-, and AhR-dependent manner. These cells were grown in 100-mm cell culture plates (Corning Glass Works; Corning, NY) using sterile technique, maintained in alpha-minimum essential media ({alpha}-MEM) supplemented with 10% (v/v) fetal bovine serum (FBS) and 1% (v/v) penicillin/streptomycin solution (Life Technologies, Gaithersburg, MD), and incubated at >80% humidity and 37°C.

General assay protocol.
Plates of stable cell clones (approximately 80–100% confluent) were trypsinized and resuspended in 20 ml {alpha}-MEM. An aliquot (200 µl) of the indicated cell suspension was added into rows two through eight of a sterile 96-well white CulturPlateTM (Packard Instruments) and plates were incubated for 24 h prior to ligand exposure, allowing cells to reach confluence. Prior to ligand addition, wells were washed with 100 µl of sterile 1X Dulbecco's phosphate-buffered saline (PBS), then exposed by the addition of either control serum [containing DMSO (1%) or TCDD (1 nM in DMSO)] or sample serum at the indicated concentration (in {alpha}-MEM supplanted with FBS). Plates were incubated for the indicated time at 37°C, after which cells were washed twice with PBS, 25 µl of 1X Lysis Buffer (Promega; Madison, WI) was added to each well and the plate placed on a plate shaker until cells were lysed (approximately 20 min). Luciferase activity was measured using an automated microplate luminometer (Dynatech ML3000; Chantilly, VA) in enhanced flash mode with the automatic injection of 50 µl of Promega stabilized luciferase reagent.

Correction for cell numbers (through protein quantification) was accomplished through the modification of an assay previously described (Kennedy et al., 1993Go). Bovine serum albumin (BSA) in lysis buffer was added to row one in differing concentrations ranging from 0 to 75 µg/well, then 100 µl of fluorescamine (Molecular Probes; Eugene, OR) in acetonitrile (at 500 µg/ml) was added to all wells. The subsequent fluorescence was quantified using a multiplate fluorometer (Fluostar, SLT) at 400 nm excitation and 460 nm emission wavelengths and results (in milligrams protein/well) were calculated by comparison to the standard curve.

Statistical methods.
Descriptive statistics and, where applicable, comparative analyses (Student t-test between samples and controls) were calculated using Excel (Microsoft, Redmond, WA). For TCDD, the median effective concentration (EC50) was calculated by fitting dose-response data to a logistic curve using Sigma Plot (SPSS, Chicago, IL) as previously described (Tillitt et al., 1991Go). Briefly, the regression equation used is as follows:

where y(d) is luciferase activity at ligand concentration d, Yb is the basal luciferase activity observed, and Ym is the maximal luciferase values measured for the ligand. For all other chemicals showing statistically significant induction above background, concentrations predicted to produce a luciferase response equal to 50% and 20% of the maximal TCDD response [designated EC50 (TCDD) and EC20 (TCDD) respectively] were calculated by logistic modeling using Sigma Plot. Corresponding induction equivalency factors (I-EFs) for the compounds of interest were calculated as the ratio of the EC value of TCDD to the EC value of the measured compound, using both the EC20 (TCDD) and EC50 (TCDD) values. Measures of correlation, ANOVAs, and MANOVA analyses were accomplished using either BMDP (BMDP, Los Angeles, CA) or Statistica (Statsoft, Tulsa, OK) statistical software.


    RESULTS AND DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Studies in our laboratory and others have employed reporter gene-based approaches to develop cell bioassay systems to detect xenobiotic exposure. Although we have developed a series of responsive CALUX bioassay systems using cell lines from different species (mouse, rat, guinea pig, and human (Garrison et al., 1996Go)) we have emphasized the use of the mouse CALUX cell bioassay system in our studies (Gebremichael et al., 1996Go; Garrison et al., 1996Go; Denison et al., 1998bGo; Phelan et al., 1998Go; Heath-Pagliuso et al., 1998Go). The mouse hepatoma (Hepa1c1c7) CALUX cell bioassay has numerous advantages over the other CALUX lines we have developed to date. They have the greatest concentration of AhR yet identified in a cell line, they are highly responsive to TCDD and other AhR ligands and are the most responsive CALUX line yet developed, and they are extremely easy to grow and work with. It has been argued that the human CALUX cell line might be a more relevant bioassay system to use because of its direct relevance to humans. However, because the CALUX bioassay is only useful as a screening system to detect the presence of AhR agonists in sample extracts and it does not provide any information as to the toxic potential of a given sample, the optimal system to use for screening purposes is obviously the one that is the most sensitive and responsive.

The CALUX bioassay has been used to determine the ability of natural and synthetic chemicals to activate the AhR system (Denison et al., 1998bGo; Heath-Pagliuso et al., 1998Go; Phelan et al., 1998Go) and to identify TCDD-like activity (i.e., AhR ligands/agonists) present in commercial and consumer products (Denison et al., 1996Go) and in environmental and biologic matrices (Denison et al., 1996Go; Ziccardi et al., 1997Go). Studies in our lab, as well as those in our collaborator's, have indicated the feasibility of using the CALUX bioassay to detect HAHs/PAHs and other AhR agonists in serum extracts (Aarts et al., 1996Go; Murk et al., 1996aGo,bGo, 1998Go) and whole serum samples (Ziccardi et al., 1997Go). Given these data, we sought to optimize this use of the bioassay as a screening tool for measurement of HAHs/PAHs and related chemicals in small volumes of whole serum.

Assay Optimization
In order to determine the minimal and optimal volume of serum needed for analysis, varying amounts of sterile goat serum (Gibco/BRL, Grand Island, NY) containing TCDD or DMSO were mixed with {alpha}-MEM (with 10% FBS) to a final volume of 50 µl, and the induction of luciferase activity in H1L1.1c2 cells contained within 96-well microtiter plates was determined. The final TCDD and DMSO concentrations in all initial optimization experiments were maintained at 1 nM and 1%, respectively. Sterile goat serum was used in these experiments because we expected that this would better mimic those samples to be collected during population surveys, rather than FBS, which would lack a variety of protein factors (such as IgG) present in adult blood. Other advantages of using goat serum included the ready availability of large stocks of samples from a single lot, the ability to acquire samples from specific pathogen-free (SPF) herds, and fewer health and regulatory concerns in its use (vs the use of human serum).

Although the highest absolute luciferase activity was observed with 25% serum (Fig. 1Go), the greatest overall fold-induction was observed at 50% level and was due to a decreased basal activity at this concentration. Because of this finding, as well as the minimization of variation of estimates at this concentration, subsequent experiments were carried out at 50% serum. It should be noted, however, that because the standard culture media already contained 10% FBS, the final serum proportion at this concentration actually contained 55% total serum (50% sample serum plus 5% FBS).



View larger version (18K):
[in this window]
[in a new window]
 
FIG. 1. Dose-response curves for induction of luciferase activity by TCDD or DMSO by different percentages of goat serum in {alpha}-MEM. A 96-well microtiter plate containing H1L1.1c2 cells approximately 80–100% confluent was incubated with DMSO or 1 nM TCDD (in DMSO) for 4 h, after which luciferase activity (corrected for protein concentration) was determined as described in Materials and Methods. Values represent the means ± SD of triplicate determinations. Values next to data points indicate fold-induction above control values at that concentration.

 
In order to determine the optimal volume of 50% goat serum necessary for maximal induction, the luciferase activity of cells exposed to increasing volumes of 50% serum (containing 1% DMSO or 1 nM TCDD) was examined. These results (Fig. 2Go) show that maximal induction was obtained between 25 and 75 µl of 50% goat serum per well in standard culture media. This translates to between 75 and 225 µl of serum being necessary for triplicate analysis. Although reduction in sample volume may not be as important in humans and in larger animals, in small animals (such as avian and mammalian species whose body weights are often less than 1 kilogram) and in endangered species, where samples are extremely precious, determination of exposures to potentially harmful levels of pollutants can be done where previously accomplishing it was extremely difficult or impossible. All subsequent trials were done using 50 µl of a 50% sample concentration.



View larger version (18K):
[in this window]
[in a new window]
 
FIG. 2. Dose-response curves for induction of luciferase activity TCDD or DMSO by different volumes of 50% goat serum (in {alpha}-MEM). A 96-well microtiter plate containing H1L1.1c2 cells approximately 80–100% confluent was incubated with DMSO or 1 nM TCDD (in DMSO) for 4 h, after which luciferase activity (corrected for protein concentration) was determined as described in Materials and Methods. Values represent the means ± SD of triplicate determinations. Values next to data points indicate fold-induction above control values at that concentration.

 
Next, we optimized the assay with emphasis on reducing analysis and preparation time. Incubation times required for maximal fold-increase between negative and positive controls were therefore determined for the following: 1) after ligand-containing serum was added to cells, 2) lysis time necessary for cells post-exposure, 3) delay time after substrate was added to lysed cells but before reading in the luminometer, and 4) integration time in the luminometer. In these experiments, cells were incubated with 50 µl of a 50% serum solution (1% DMSO or 1 nM TCDD final concentration) and luciferase activity was determined at various times. Comparable to our previous results (Garrison et al., 1996Go), maximal induction and fold-induction occurred at 4 h post-exposure (Fig. 3Go). Additional studies examining other PAH and HAH compounds in our laboratory have found that this maximum induction routinely occurs from 3 to 5 h post-exposure (data not shown). Using these optimal incubation conditions revealed that maximal luciferase activity was measured when cells were lysed for at least 15 min and the luminometer was set to read data with a delay time of greater than five s and an integration time of greater than 10 s (data not shown). Therefore, as 24 triplicate samples (and appropriate controls) can be run per 96-well plate and it takes approximately 5 h to complete an assay (from initial cell treatment to final analysis), more than 100 samples could be run in a typical day on a single luminometer (assuming parallel incubation and analysis of plates).



View larger version (19K):
[in this window]
[in a new window]
 
FIG. 3. Time dependent induction of luciferase activity by DMSO or TCDD. Several 96-well microtiter plates containing H1L1.1c2 cells approximately 80–100% confluent were incubated with DMSO or 1 nM TCDD (in DMSO) for the indicated time, after which luciferase activity (corrected for protein concentration) was determined as described in Materials and Methods. Values represent the mean ± SD of triplicate determinations. Values next to data points indicate fold-induction above control values at that concentration.

 
Correction for Interplate and Intraplate Variability
Factors that might affect cellular proliferation and viability between plates (interplate error) and within a single plate (intraplate error) include differing growth conditions, cell numbers, and time from plate splitting" to analysis. Given this potential for variation, a method to standardize the assay is an extremely important factor to incorporate into the protocol. To address this, we evaluated several procedures to maximize reliability of the assay. First, the number of cells added to each well was varied in order to optimize the bioassay relative to the highest fold- and overall induction while attempting to minimize the numbers of cells necessary to add to the plate. Cells were induced by the addition of 1% DMSO, 10, 1, or 0.1 nM TCDD in triplicate; overall induction and fold-induction were determined as described above. As increasing cell numbers were added to each well, overall induction increased in a relatively linear manner, with maximal activity evident above 60,000 cells seeded per well 24 h prior to serum (chemical) exposure (Fig. 4Go). A subsequent comparison of fold-induction between the different cell concentrations showed that levels greater than 60,000 cells per well were statistically similar and showed the highest comparative induction (data not shown).



View larger version (26K):
[in this window]
[in a new window]
 
FIG. 4. Dose-response curves for induction of luciferase activity of DMSO or TCDD with different concentrations of cells added to the microtiter plate. A 100-mm confluent plate of H1L1.1c2 cells was trypsinized and resuspended in {alpha}-MEM to obtain a dilution of 9.2 x 105 cells per ml. Five additional 2-fold dilutions (approximate) were made using {alpha}-MEM; 200 µl of these concentrations were added to a 96-well microtiter plate. After 24 h, wells were incubated with DMSO or 1 nM TCDD (in DMSO) for 4 h, after which luciferase activity (corrected for protein concentration) was determined as described in Materials and Methods. Values represent the means ± SD of triplicate determinations.

 
A second method to correct for possible interplate variation is to adjust all data using a normalization correction factor. The factor was estimated using the method of Kramer et al. (1983). This approach involves multiplying all values on a specific plate by a value calculated from the repeated analysis of a single known positive sample, reducing between-plate variability and resulting in the generation of an appropriate correction factor. For this assay system, this factor was generated from the measurement of induction of 1 nM TCDD in 30 separate analyses using the standardized conditions and was set at 14.95 for all subsequent trials (data not shown).

In order to correct for slight differences in cell numbers between wells, which would result in intraplate variability, a method to assess cell number postanalysis indirectly (through the assessment of protein concentrations within each well using fluorescamine) was optimized for use in this protocol. The fluorescamine protein assay has been previously used for rapid fluorescence-based estimation of protein concentrations in microplate assays (Heath-Pagliuso et al., 1998Go; Kennedy et al., 1993Go; Phelan et al., 1998Go; Sanderson et al., 1996Go). In the CALUX bioassay, microplates are used directly after luminescent analysis, with 100 µl of a 500 µg/ml fluorescamine concentration in acetonitrile as the optimal volume and concentration (data not shown). Acetonitrile was selected as the solvent vehicle because it does not affect the polystyrene microtiter plate (unlike solvents such as acetone), and it also quenches the luminescence present within wells; therefore, the developing fluorescence is not confounded by reporter activity. After incubation for 15 min on a plate shaker, fluorescence is quantified using a multiplate fluorometer (Fluostar, SLT) at 400 nm excitation and 460 nm emission wavelengths. Protein concentrations (milligrams protein per well) were then calculated by comparison to a standard curve of 0–75 µg/well BSA assayed concurrently in rows of the same plate that are void of cells. Further characterization of this assay has shown that the results are linear up to 300 µg protein per well (data not shown), which is sufficient to detect the protein concentration of lysed cells present in this system. Overall, the application of these normalization/correction methods allows direct comparison of results within and among plates with a greater degree of confidence. The optimized protocol flow used in all subsequent studies is shown in Fig. 5Go.



View larger version (31K):
[in this window]
[in a new window]
 
FIG. 5. Flow chart showing the optimized CALUX bioassay method for detection of Ah receptor ligands in whole serum.

 
Analytic Sensitivity of the Direct Serum Bioassay for TCDD
The utility of this application of the CALUX bioassay is highly dependent on its analytic sensitivity as well as its rapidity and ease of analysis. The dose dependence of the serum bioassay was determined by induction of cells with serum containing increasing concentrations of TCDD and measurement of luciferase activity after 4-h incubation. Induction of luciferase activity was dose dependent (Fig. 6Go), with maximal induction evident with concentrations greater than or equal to 2.5 nM TCDD, an EC50 of approximately 0.68 nM (or 10.90 pg per well), and a minimal detection limit (MDL) of approximately 0.1 nM (or 1.61 pg per well). Although these data seem to show that the serum-based CALUX bioassay is analytically less sensitive than previously-reported AhR-based luciferase cell bioassays, when these data are converted to an absolute amount of ligand present (in either grams or moles per well), we note that the results show a sensitivity only slightly less than that previously reported (Table 1Go). In addition, when compared to results in media alone, the resultant dose-response curve shifted only slightly to the right. Therefore, this reduction in analytic sensitivity is most likely not due to the direct analysis of serum, but from the incorporation of small sample volumes (50 µl vs up to 2 ml in some systems), the significant reduction in time required for analysis (4-h incubation vs several days), and the direct use of the 96-well microtiter plate format (vs methods employed in previous studies using larger format plates, well scraping, centrifugation, and aliquoting) in the methodology. Therefore, these data support the utility of this bioassay for detection of TCDD in unextracted serum.



View larger version (19K):
[in this window]
[in a new window]
 
FIG. 6. Dose-response curves for induction of luciferase activity of TCDD in 100% {alpha}-MEM or 50% goat serum (in {alpha}-MEM containing 5% FBS). A 96-well microtiter plate containing H1L1.1c2 approximately 80–100% confluent was incubated with 50 µl of TCDD-containing media at the indicated concentrations for 4 h. Luciferase activity (corrected for protein concentration) was determined as described in Materials and Methods. Values represent the means ± SD of triplicate determinations. *Value statistically significantly different from DMSO control (Student t-test, p < 0.05).

 

View this table:
[in this window]
[in a new window]
 
TABLE 1 Comparison of the Effective Doses for Half-Maximal Luciferase Induction (EC50s) and Minimum Detection Levels (MDLs) for TCDD in Several Studies Using the Luciferase Reporter Gene As a Bioassay for the Detection of Ah Receptor Ligands
 
Induction of Luciferase by HAHs and PAHs in Serum
In order to assess the analytic sensitivity and effectiveness of this diagnostic tool to detect different xenobiotic exposures in human or animals, sterile goat serum was spiked with increasing concentrations of several HAHs and PAHs and luciferase induction was determined using the optimized assay protocol and correction procedures (Figs. 7A–CGo). Significant dose-dependent induction of luciferase was observed with eight chemicals (TCDD, TCDF, PeCDF, TCB, 3-PeCB, BKF, BAP, and BAA). Minimum detection limits (MDLs), EC50 (TCDD) values, EC20 (TCDD) values, and induction equivalency factors (I-EFs) were next determined for these compounds and are presented in Table 2Go. Such I-EF values are akin to TEFs (or toxic equivalency factors) reported in the toxicologic literature. However, their calculation does not imply the addition of different xenobiotic fractions, but only their relative induction potency. In this study, the calculated I-EF values correlate extremely well with other published data on both HAH and PAH compounds (correlation coefficients from 0.85–0.99; data not shown) with EC50 values falling in the range of those reported (Anderson et al., 1995Go; Richter et al., 1997Go; Sanderson et al., 1996Go; Willett et al., 1997Go).



View larger version (20K):
[in this window]
[in a new window]
 
FIG. 7. Dose-response curves for induction of luciferase activity by selected polycyclic or halogenated aromatic hydrocarbons in 50% goat serum (in {alpha}-MEM) that showed significant increases above the negative control. Several 96-well microtiter plates containing H1L1.1c2 cells approximately 80–100% confluent were incubated with (A) TCDD, 2,3,7,8-tetrachlorodibenzofuran (TCDF) and 2,3,4,7,8-pentachlorodibenzofuran (PCDF); (B) 3,3',4,4'-tetrachlorobiphenyl (TCB) and 3,3',4,4',5-pentachlorobiphenyl (3-PeCB); or (C) benzo(k)fluoranthene (BKF), benzo(a)pyrene (BAP), and benz(a)anthracene (BAA) at the indicated concentrations for 4 h, after which luciferase activity (corrected for protein concentration) was determined as described in Materials and Methods. Values represent the means ± SD of triplicate determinations.

 

View this table:
[in this window]
[in a new window]
 
TABLE 2 Minimum Detection Levels (MDLs), Effective Concentrations for 50% and 20% Maximal 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) Luciferase Induction [EC50 (TCDD) and EC20 (TCDD), respectively], and Induction Equivalency Factors (I-EFs) for Those Polycyclic and Halogenated Aromatic Hydrocarbons in 50% Goat Serum (in {alpha}-MEM) That Showed Significant Increases Above the Control
 
Determination of AhR Dependence of Serum Induction
Because the analysis described here uses normal rather than solvent-extracted serum, it is possible that a given sample may contain substances that stimulate luciferase gene expression by an AhR-independent manner (possibly by endogenous cellular components and growth factors), resulting in false positives. In order to confirm that an observed increase in luciferase activity is due to an AhR-dependent mechanism, induction by samples were also examined in the presence of an AhR antagonist. In initial experiments (data not shown), we evaluated several different antagonists for their ability to block induction of luciferase activity by serum containing TCDD. Of the antagonists examined [{alpha}-NF (Gasiewicz and Rucci, 1991Go), MCDF (Astroff et al., 1988Go), AMF, or MNF (Lu et al., 1995Go)], AMF was the most effective inhibitor. This is not surprising given that AMF, unlike other AhR antagonists, appears to be a pure antagonist and does not exhibit any agonist activity even at high concentrations (Lu et al., 1995Go). Dose-response experiments (Fig. 8Go) further revealed that 100 µM AMF effectively inhibited induction of luciferase by serum containing 10 nM TCDD. Thus, when a serum sample is positive in the CALUX assay, the AhR-dependence of this induction can be confirmed by reanalysis of another aliquot of the sample in the presence of AMF. By the same argument, the presence of these and similar compounds (i.e., naturally occurring antagonists) in unknown serum samples might have the ability to falsely reduce luciferase activity, resulting in a false negative result. Although this occurrence would be very unlikely due the necessity of greater than micromolar concentrations of these compounds to cause even slight inhibition, it is still possible. To address this issue, a given sample could be spiked with a defined amount of an AhR agonist and the level of induction compared to that obtained using a control spiked with the same amount of chemical. A reduction in the amount of activation in the unknown sample could suggest the presence of an antagonist or inhibitor of the AhR signaling pathway.



View larger version (21K):
[in this window]
[in a new window]
 
FIG. 8. Dose-response curves for induction of luciferase activity of TCDD in the presence of DMSO or different concentrations of the AhR antagonist 4-amino-3-methoxyflavone (AMF). A 96-well microtiter plate containing H1L1.1c2 approximately 80–100% confluent was incubated with DMSO or 4A3MF at the indicated concentration for 15 min, followed by DMSO or TCDD at the indicated concentration for 4 h. Luciferase activity (corrected for protein concentration) was determined as described in Materials and Methods. Values represent the mean ± SD from triplicate determinations.

 
Overall Error Determination
In order to assess the effectiveness of these error correction methods to control for the different potential sources of variation within this assay system, several parallel and serial analyses of spiked serum samples were evaluated both before and after the use of postexposure correction methods (i.e., protein, TCDD, and antagonist correction). Two 96-well plates were prepared by two separate operators on two separate days following either the protocols above or the protocols without cell numerical correction, and 12 positive (1 nM TCDD in goat serum) and negative (1% DMSO in goat serum) samples from a single source were added in each. Luciferase activity for all exposed wells in each plate was assessed as described in the protocol, and the variation in results between plates, persons, and days for both positive and negative analyses prior to and postcorrection was evaluated using multivariate analysis of variance (MANOVA) methods. Overall, for those plates in which cell numbers were not adjusted before exposure, all effects (which include all 1-, 2-, and 3-way comparisons between operator, plate, and day) were statistically significantly different (p < 0.001) for both positive and negative samples. In those plates where cell numbers were corrected for prior to plating, statistically significant differences (p < 0.001) were observed between plates and with the day/plate effect in both DMSO and TCDD-exposed wells. Additional significant differences were noted in TCDD samples between different operators, days, and operator/day effects. After correction (protein quantification, TCDD standardization, and antagonist use), only the operator/day effect remained significant and only for the DMSO samples. This remaining significant result is most likely attributable to one pair of plates run by one operator being slightly lower in TCDD expression, thereby falsely skewing some negative samples toward higher values. Overall, however, the correction of results from serum sample analysis for sources of interplate (by cell number quantification, TCDD standardization, and antagonist use) and intraplate (by protein quantification using fluorescamine) variation effectively reduces the potential for confounding by non–ligand-dependent variables.

Application of the CALUX Serum Bioassay for Screening of Wildlife Samples
As an example of the type of information that this bioassay system could afford biomedical researchers, a subset of sera from several different marine wildlife species (with unknown xenobiotic exposure) was acquired and analyzed using the optimized method described above (Fig. 9Go). Results from these analyses show a spectrum of luciferase activity that can be interpreted based roughly on the natural histories of each. Samples collected from black-footed albatross, a seabird that primarily resides at a distance from shore-based activities, had minimal luciferase activity. Double-crested cormorants, a species that has more interaction with the mainland (therefore more potential contact with xenobiotics) exhibited statistically significant levels of induction but at relatively low levels (on the order of picomolar equivalents of TCDD). The final species examined, the Southern sea otter (Enhydra lutris nereis), is a species at high risk of pollutant exposure due to its residing exclusively in the nearshore habitat, its preying on organisms known to bioaccumulate environmental toxicants, and its reliance on a high metabolic rate of activity and thick layer of hair for thermoregulation. These data indicate that the animals examined here have been exposed to relatively high levels of unknown pollutants (approximately equivalent to 0.1 nM TCDD). Although it is currently unknown what specific chemical(s) in these serum samples is responsible for the observed induction response, our results demonstrate that this bioassay system is a highly effective method for screening large numbers of potentially exposed individuals. Subsequent studies will focus on final validation of this bioassay approach by comparing CALUX induction results with those of analytical chemical analysis (GC/MS) of the same unknown sample. By comparison of the bioassay results with activity obtained from concurrent analysis of several concentrations of TCDD, overall induction equivalencies can be determined for a particular sample. Positive screening test results could then be followed by chemical analysis in order to identify the specific chemicals responsible for the induction response and to assess specific exposure in the individuals. It should be noted, however, that we have observed relatively high background activity in some serum samples from humans (Ziccardi et al., 1997Go) and selected wildlife species, whereas serum from other species produces little or no background luciferase activity (Ziccardi et al., manuscript in preparation). Solvent extraction of serum eliminates this background activity (data not shown). These differences may be due to higher exposure of certain populations to AhR-inducing compounds or the presence of endogenous activators in the blood of species that exhibit high background activity. Studies are currently underway to identify the substance(s) present in these control serum samples responsible for activation of the reporter gene. Because this method utilizes an in vitro methodology to detect compounds of interest, interpretation of these data for predicting in vivo toxicologic effects for any species must be made carefully. Therefore, without a priori knowledge of background CALUX activity and endogenous AhR response in the population and species of interest, direct application of results may be limited.



View larger version (20K):
[in this window]
[in a new window]
 
FIG. 9. Overall induction of luciferase activity of selected wildlife samples. A 96-well microtiter plate containing H1L1.1c2 approximately 80–100% confluent was incubated with a 50% dilution of serum sample in the absence or presence of 10 µM 4-amino-3-methoxyflavone (AMF) for 4 h. Luciferase activity (corrected for protein concentration) was determined as described in Materials and Methods. Values represent the means ± SD from triplicate determinations. Asterisks indicate those values statistically significantly different from DMSO control (Student t-test, p < 0.05).

 
Applications and Limitations
The use of this bioassay for detection of xenobiotic exposure has many advantages over methods previously used. First and most important, because it is a direct measure of contaminant concentrations in blood, it does not rely on the individual's response (or lack of response) to the contaminant. Second, because this assay requires minute volumes of test substance (75 to 225 µl of serum versus the significant amounts of tissue necessary in other assays), it is easier to acquire samples. Third, because it is blood based, problems associated with acquiring samples from endangered or small-bodied species are less than when invasive collection of tissue is necessary. Fourth, while the use of serum applied directly on cells using these methods may not have the same degree of sensitivity as does direct application or extraction of ligand in other systems, this method is still extremely analytically sensitive, detecting TCDD in serum at the low parts per trillion level. Therefore, this bioassay could be a valuable tool for rapid and initial assessment of exposure both in human populations, where the analysis of pollutant exposure in statistically significant numbers of volunteers or cases is often limited by cost, as well as in wildlife studies, where the assessment of xenobiotic exposure is often limited both by cost plus sample availability and volume.

Although this assay technique has proven to be a valuable method to detect overall exposure to xenobiotic compounds, it cannot be used exclusively for diagnosis. Because of its lack of analytic specificity for individual compounds (due to the detection of any compound that binds and activates the AhR and AhR signal transduction pathway) and the potential for the presence of high basal levels of luciferase activity in some species, we believe that a primary application of this assay is as a rapid and inexpensive method for prescreening a large number of samples from selected species for the presence of AhR ligands. Once positive samples are identified, analytical chemical approaches such as GC/MS and/or HPLC can be used to identify the specific compounds or congeners. Work is currently underway, however, to further refine the methodology described here in order to separate the luciferase activity due to different classes of chemicals possibly present in serum. This combination of chemical and class-specific extraction procedures, coupled with the CALUX assay, will provide further information as to potential exposure to AhR ligands and activators in an inexpensive and relatively rapid manner.


    ACKNOWLEDGMENTS
 
The authors thank Drs. Jonna Mazet and David Jessup for their assistance and support of this project; Michael Sowby and the California Department of Fish and Game, Office of Oil Spill Prevention and Response for their continued support of the development of this system; Dr. John Giesy for the albatross samples and Dr. Jay Davies for the cormorant samples; Drs. George Clark and David Brown of Xenobiotic Detection Systems for technical assistance and advice; and W. Joe Rogers, Jack Lam, and William Brackney for laboratory assistance. This research was supported by the University of California (Robert Cessna Fellowship), California Department of Fish and Game, Office of Oil Spill Prevention and Response (FG-6404OS and FG-393OS), the University of California Toxic Substances Research and Training Program, the National Institute of Environmental Health Sciences (ES07685), and Superfund Basic Research program grant (ES04699), and the California Agricultural Experimental Station.


    NOTES
 
1 Present address: Lincoln Park Zoo, P. O. Box 14903, Chicago, IL 60614. Back

2 To whom correspondence should be addressed at Department of Environmental Toxicology, Meyer Hall, One Shields Avenue, University of California, Davis, CA 95616. Fax: (530) 752-3394. E-mail: msdenison{at}ucdavis.edu. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Aarts, J. M. M. J. G., Cenjin, P. H., Blankvoort, B. M. G., Murk, A., Brouwer, A., Bovee, T. F. H., Traag, W. A., Hoogenboom, L. A. P., Patandin, S., Weisglas-Kuperus, N., Sauer, P. J. J., and Denison, M. S. (1996). Application of the chemical-activated luciferase expression (CALUX) bioassay for quantification of dioxin-like compound in small samples of human milk and blood plasma. Organohal. Cmpds. 27, 285–290.

Aarts, J. M. M. J. G., Denison, M. S., Cox, M. A., Schalk, M. A. C., Garrison, P. M., Tullis, K., deHaan, L. H. J., and Brouwer, A. (1995). Species-specific antagonism of Ah receptor action by 2,2',5,5'-tetrachloro- and 2,2',3,3',4,4'-hexachlorobiphenyl. Eur. J. Pharmacol. 293, 463–474.[Medline]

Aarts, J. M. M. J. G., Denison, M. S., De Haan, L. H. J., Schalk, J. A. C., Cox, M. A., and Brouwer, A. (1993). Ah receptor-mediated luciferase expression: a tool for monitoring dioxin-like toxicity. Organohal. Cmpds. 13, 361–364.

Abraham, K., Krowke, R., and Neubert, D. (1988). Pharmacokinetics and biological activity of 2,3,7,8-tetrachlorodibenzo-p-dioxin. 1. Dose-dependent tissue distribution and induction of hepatic ethoxyresorufin O-deethylase in rats following a single injection. Arch. Toxicol. 62, 359–368.[ISI][Medline]

Anderson, J. W., Rossi, S. S., Tukey, R. H., Vu, T., and Quattrochi, L. C. (1995). A biomarker, P450 RGS, for assessing the induction potential of environmental samples. Environ. Toxicol. Chem. 14, 1159–1169.[ISI]

Astroff, B., Zacharewski, T., Safe, S., Arlotto, M. P., Parkinson, A., Thomas, P., and Levin, W. (1988). 6-Methyl-1,3,8-trichlorodibenzofuran as a 2,3,7,8-tetrachlorodibenzo-p-dioxin antagonist: inhibition of the induction of rat cytochrome P-450 isozymes and related monooxygenase activities. Mol. Pharmacol. 33, 231–236.[Abstract]

Bartosek, I., Guaitani, A., Modica, R., Fiume, M., and Urso, R. (1984). Comparative kinetics of oral benz(a)anthracene, chrysene and triphenylene in rats: study with hydrocarbon mixtures. Toxicol. Lett. 23, 333–339.[ISI][Medline]

Bradlaw, J. A., and Casterline, J. L. Jr. (1979). Induction of enzyme activity in cell culture: A rapid screen for detection of planar polychlorinated organic compounds. J. Assoc. Off. Anal. Chem. 62, 904–916.[ISI][Medline]

Burke, M. D., and Mayer, R. T. (1974). Ethoxyresorufin: direct fluorimetric assay of a microsomal O-dealkylation which is preferentially inducible by 3-methylcholanthrene. Drug Metab. Dispos. 2, 583–588.[ISI][Medline]

Denison, M. S., El-Fouly, M. H., Aarts, J. M. M. J. G., Brouwer, A., Richter, C., and Giesy, J. P. (1993). Production of novel recombinant cell line bioassay systems for detection of 2,3,7,8-tetrachlorodibenzo-p-dioxin-like chemicals. Organohal. Cmpds. 13, 365–368.

Denison, M. S., Garrison, P. M., Aarts, J. M. M. J. G., Tullis, K., Schalk, J. A. C., Cox, M. A., and Brouwer, A. (1995). Species-specific differences in Ah receptor ligand binding and transcriptional activation: Implications for bioassays for the detection of dioxin-like chemicals. Organohal. Cmpds. 23, 225–229.

Denison, M. S., Phelan, D., and Elferink, C. J. (1998a). The Ah receptor signal transduction pathway. In Toxicant-Receptor Interactions (M. S. Denison and W. G. Helferich, Eds.). pp. 3–33. Taylor and Francis, Philadelphia.

Denison, M. S., Phelan, D., Winter, G. M., and Ziccardi, M. H. (1998b). Carbaryl, a carbamate insecticide, is a ligand for the hepatic Ah (dioxin) receptor. Toxicol. Appl. Pharmacol. 152, 406–414.[ISI][Medline]

Denison, M. S., Rogers, W. J., Fair, M., Ziccardi, M., Clark, G., Murk, A. J., and Brouwer, A. (1996). Application of the CALUX bioassay system for the detection of dioxin-like chemicals (Ah receptor ligands) in whole serum samples and in extracts from commercial and consumer products. Organohal. Cmpds. 27, 280–284.

Denison, M. S., and Wilkinson, C. F. (1985). Identification of the Ah receptor in selected mammalian species and induction of aryl hydrocarbon hydroxylase. Eur. J. Biochem. 147, 429–435.[Abstract]

Garrison, P. M., Tullis, K., Aarts, J. M. M. J. G., Brouwer, A., Giesy, J. P., and Denison, M. S. (1996). Species-specific recombinant cell lines as bioassay systems for the detection of 2,3,7,8-tetrachlorodibenzo-p-dioxin-like chemicals. Fundam. Appl. Toxicol. 30, 194–203.[ISI][Medline]

Gasiewicz, T. A., and Rucci, G. (1991). Alpha-naphthoflavone acts as an antagonist of 2,3,7,8- tetrachlorodibenzo-p-dioxin by forming an inactive complex with the Ah receptor. Mol. Pharmacol. 40, 607–612.[Abstract]

Gebremichael, A., Tullis, K., Denison, M.S., Cheek, J. M., and Pinkerton, K. E. (1996). Ah-receptor-dependent modulation of gene expression by aged and diluted sidestream cigarette smoke. Toxicol. Appl. Pharmacol. 141, 76–83.[ISI][Medline]

Hahn, M. E., Lamb, T. M., Schultz, M. E., Smolowitz, R. M., and Stegeman, J. (1993). Cytochrome P4501A induction and inhibition by 3,3',4,4'-tetrachlorobiphenyl in an Ah receptor-containing fish hepatoma cell line (PLHC-1). Aquat. Toxicol. 26, 185–208.[ISI]

Heath-Pagliuso, S., Rogers, W. J., Tullis, K., Seidel, S. D., Cenijn, P. H., Brouwer, A., and Denison, M. S. (1998). Activation of the Ah receptor by tryptophan and tryptophan metabolites. Biochemistry 37, 11508–11515.[ISI][Medline]

Kennedy, S. W., Lorenzen, A., James, C. A., and Collins, B. T. (1993). Ethoxyresorufin-O-deethylase and porphyrin analysis in chicken embryo hepatocyte cultures with a fluorescence multiwell plate reader. Anal. Biochem. 211, 102–112.[ISI][Medline]

Kramer, S. M., Jewell, N. P., and Cremer, N. E. (1983). Discriminant analysis of data in enzyme immunoassay. J. Immunol. Methods 60, 243–255.[ISI][Medline]

Lu, Y. F., Santostefano, M., Cunningham, B. D., Threadgill, M. D., and Safe, S. (1995). Identification of 3'-methoxy-4'-nitroflavone as a pure aryl hydrocarbon (Ah) receptor antagonist and evidence for more than one form of the nuclear Ah receptor in MCF-7 human breast cancer cells. Arch. Biochem. Biophys. 316, 470–477.[ISI][Medline]

Murk, A. J., Jonas, A., Brouwer, A., Leonards, P. E. G., and Denison, M. S. (1996a). Application of the CALUX (chemical activated luciferase gene expression) assay for measuring TCDD-equivalents in sediment, pore water and blood plasma samples. Organohal. Cmpds. 27, 291–296.

Murk, A. J., Legler, J., Denison, M. S., Giesy, J. P., van de Guchte, C., and Brouwer, A. (1996b). Chemical-activated luciferase gene expression (CALUX): a novel in vitro bioassay for Ah receptor active compounds in sediments and pore water. Fundam. Appl. Toxicol. 33, 149–160.[ISI][Medline]

Murk, A. J., Leonards, P. E. G., van Hattum, B., Luit, R., van der Weiden, M. E. J., and Smit, M. (1998). Application of biomarkers for exposure and effect of polyhalogenated aromatic hydrocarbons in naturally exposed European otters (Lutra lutra). Environ. Toxicol. Pharmacol. 6, 91–102.[ISI]

Nebert, D. W., and Gelboin, H. V. (1968). Substrate-inducible microsomal aryl hydroxylase in mammalian cell culture. I. Assay and properties of induced enzyme. J. Biol. Chem. 243, 6242–6249.[Abstract/Free Full Text]

Nelson, D. R., Kamataki, T., Waxman, D. J., Guengerich, F. P., Estabrook, R. W., Feyereisen, R., Gonzales, F. J., Coon, M. J., Gunsalus, I. C., Gotoh, O., Okuda, K., and Nebert, D. W. (1993). The P450 superfamily: update on new sequences, gene mapping, accession numbers, early trivial names of enzymes and nomenclature. DNA Cell Biol. 12, 1–51.[ISI][Medline]

Patterson Jr., D. G., Fingerhut, M. A., Roberts, D. W., Needham, L. L., Sweeney, M. H., Marlow, D. A., Andrews J. S. Jr., and Halperin, W. E. (1989). Levels of polychlorinated dibenzo-p-dioxins and dibenzofurans in workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Am. J. Ind. Med. 16, 135–146.[ISI][Medline]

Phelan, D., Winter, G. M., Rogers, W. J., Lam, J. C., and Denison, M. S. (1998). Activation of the Ah receptor signal transduction pathway by bilirubin and biliverdin. Arch. Biochem. Biophys. 357, 155–163.[ISI][Medline]

Postlind, H., Vu, T. P., Tukey, R. H., and Quattrochi, L. C. (1993). Response of human CYP1-luciferase plasmids to 2,3,7,8-tetrachlorodibenzo-p-dioxin and polycyclic aromatic hydrocarbons. Toxicol. Appl. Pharmacol. 118, 255–262.[ISI][Medline]

Richter, C. A., Tieber, V. L., Denison, M. S., and Giesy, J. P. (1997). An in vitro rainbow trout cell bioassay for aryl hydrocarbon receptor-mediated toxins. Environ. Toxicol. Chem. 16, 543–550.[ISI]

Safe, S., Mason, G., Sawyer, T., Zacharewski, T., Harris, M., Yao, C., Keys, B., Farrell, K., Holcomb, M., Davis, D., Safe, L., Piskorska-Pliszczynska, J., Leece, B., Denomme, M., Hutzinger, O., Thoma, H., Chittim, B., and Madge, J. (1989). Development and validation of in vitro induction assays for toxic halogenated aromatic mixtures: a review. Toxicol. Ind. Health 5, 757–775.[ISI][Medline]

Sanderson, J. T., Aarts, J. M., Brouwer, A., Froese, K. L., Denison, M. S., and Giesy, J. P. (1996). Comparison of Ah receptor-mediated luciferase and ethoxyresorufin-O-deethylase induction in H4IIE cells: implications for their use as bioanalytical tools for the detection of polyhalogenated aromatic hydrocarbons. Toxicol. Appl. Pharmacol. 137, 316–325.[ISI][Medline]

Tillitt, D. E., Giesy, J. P., and Ankley, G. T. (1991). Characterization of the H4IIE rat hepatoma cell bioassay as a tool for assessing toxic potency of planar halogenated hydrocarbons in environmental samples. Environ. Sci. Tech. 25, 87–92.[ISI]

Vanden Heuvel, J. P., Clark, G. C., Thompson, C. L., McCoy, Z., Miller, C. R., Lucier, G. W., and Bell, D. A. (1993). CYP1A1 mRNA levels as a human exposure biomarker: use of quantitative polymerase chain reaction to measure CYP1A1 expression in human peripheral blood lymphocytes. Carcinogenesis 14, 2003–2006.[Abstract]

Whitlock, J. P. (1990). Genetic and molecular aspects of 2,3,7,8-tetrachlorodibenzo-p-dioxin action. Annu. Rev. Pharmacol. Toxicol. 30, 251–277.[ISI][Medline]

Willett, K. L., Gardinali, P. R., Sericano, J. L., Wade, T. L., and Safe, S. H. (1997). Characterization of the H4IIE rat hepatoma cell bioassay for evaluation of environmental samples containing polynuclear aromatic hydrocarbons (PAHs). Arch. Environ. Contam. Toxicol. 32, 442–448.[ISI][Medline]

Ziccardi, M., Rogers, W. J., Schecter, A. J., Papke, O., and Denison, M. S. (1997). Direct detection of dioxins and related chemicals in small quantities of human serum using a recombinant cell bioassay. Organohal. Cmpds. 31, 215–218.