- Email: [email protected]
Using the metabolism of PAHs in a human cell line to characterize environmental samples
Environmental Toxicology and Pharmacology 8 (2000) 119 – 126
www.elsevier.com/locate/etap
Using the metabolism of PAHs in a human cell line to characterize environmental samples Jennifer M. Jones a, Jack W. Anderson a,*, Robert H. Tukey b a
Columbia Analytical Ser6ices, 1185 Park Center Dri6e, Suite A, Vista, CA 92083, USA b Department of Pharmacology, Uni6ersity of California, San Diego, CA, USA
Received 22 October 1999; received in revised form 26 January 2000; accepted 3 February 2000
Abstract P450 reporter gene system (RGS) is an in vitro assay to detect compounds that activate the Ah receptor and induce cytochrome P450 (CYP1A1). This system utilizes a human cell (101L) stably transfected with a luciferase reporter downstream of human CYP1A1 promoter sequences. When CYP1A1-inducing compounds are present, luciferase is produced as well as endogenous CYP1A1 enzymes. Polycyclic aromatic hydrocarbons (PAHs) are more readily degraded than chlorinated compounds including dioxins, furans, and coplanar polychlorinated biphenyls (PCBs). PAH-induced luciferase production begins to decrease between 6 and 16 h, while chlorinated compounds produce a more sustained response. Individual and mixtures of CYP1A1-inducing compounds were tested at both 6 and 16 h. Extracts of soils containing both PAHs and dioxins were also tested, before and after cleanup to remove PAHs. Results indicate that RGS testing at 6 and 16 h is a promising tool to differentiate between PAHs and chlorinated hydrocarbons often co-occurring in environmental samples. © 2000 Elsevier Science B.V. All rights reserved. Keywords: CYP1A1; Ah receptor; PAHs; Chlorinated hydrocarbons; Co-occurrence
1. Introduction Polycyclic aromatic hydrocarbons (PAHs), as well as the polychlorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls (PCBs) are known to induce the CYP1A gene subfamily via the Ah-receptor (Whitlock, 1990). The subsequent production of CYP1A1 enzymes is an important pathway for metabolism and removal of these compounds from the cell. Although they share this common mode of action, the chlorinated inducers (dioxins, furans, and PCBs) have a lower rate of metabolism than PAHs. Shorter duration of CYP1A1 induction by PAHs compared to chlorinated compounds has been reported with in vitro studies using a fish hepatoma cell line (PLHC-1) (Celander et al., 1997), rainbow trout hepatocytes (Pesonen et al., 1992), and porcine aorta endothelial cells (Stegeman et al., 1995), as well as in vivo studies with marine killifish (Kloepper-Sams and Stegeman, * Corresponding author. Tel.: +1-760-7341450; fax: + 1-7607341409. E-mail address: [email protected] (J.W. Anderson)
1989). Hahn and Stegeman (1994) reported sustained induction of CYP1A1 mRNA by 2,3,7,8-TCDF in scup, contrasting with their previous findings of transient induction by nonhalogenated inducers. The prolonged induction of CYP1A1 mRNA by chlorinated hydrocarbons has been attributed to their slower metabolism and removal, leading to continued synthesis of mRNA (Stegeman et al., 1992). Postlind et al. (1993) reported similar results using 101L cells, which are human hepatoma cells stably integrated with a plasmid containing CYP1A1 sequences fused to the luciferase reporter gene. The production of luciferase in these cells is regulated by the dioxin-responsive elements of the CYP1A1 sequences, and is not controlled by endogenous feedback mechanisms in the cell that may affect CYP1A1 protein levels. Because luciferase has a short half-life in mammalian cells (0.5–3 h) (Thomson et al., 1991; Bronstein, 1994), the levels of luciferase activity may reflect short-term changes in the activity of the dioxin-responsive elements upstream. Postlind et al. (1993) reported that CYP1A1-mediated luciferase activity reached maximal levels and did not decrease between 12 and 24 h of
1382-6689/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 2 - 6 6 8 9 ( 0 0 ) 0 0 0 3 3 - 8
120
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
exposure to tetrachlorodibenzo-p-dioxin (TCDD). In contrast, luciferase activity reached maximal levels by 6 – 12 h of exposure to two PAHs (benzo[a]anthracene and benzo[a]pyrene), at which point levels dropped sharply. This decrease was presumably caused by metabolism of the PAHs and decrease of the concentrations of parent PAH present in the cells. The reporter gene system (RGS) is an application of the 101L cell line for use by industry and government agencies to screen environmental samples for the presence of CYP1A1-inducing compounds (Anderson et al., Table 1 Composition of polycyclic aromatic hydrocarbon (PAH) mixture PAH compound
Conc. (mg/ml)
Benzo[k]flouranthene Dibenzo[a,h]anthracene Benzo[b]flourathene Ideno[1,2,3-cd]pyrene Benzo[a]pyrene Benzo[a]anthracene Chrysene Benzo[g,h,i ]perylene Acenapthene Acenaphthylene Anthracene Benzo[e]pyrene Biphenyl Dibenzothiophene Fluoranthene Fluorene Naphthalene Perylene Phenanthrene Pyrene 1-Methylnaphthalene 2-Methylnaphthalene
13.7 4.7 13.7 10.5 21.2 14.8 15.9 13.5 1.7 4.2 5.3 16.4 14.7 2.3 16.1 3.8 15.0 22.4 13.6 29.6 45.7 33.5
Table 2 Composition of dioxin/furan mixture Analyte
Conc. (ng/ml)
2,3,7,8-Tetrachlorodibenzo-p-dioxin 1,2,3,7,8-Pentachlorodibenzo-p-dioxin 1,2,3,4,7,8-Tetrachlorodibenzo-p-dioxin 1,2,3,6,7,8-Hexrachlorodibenzo-p-dioxin 1,2,3,7,8,9-Hexachlorodibenzo-p-dioxin 1,2,3,4,6,7,8-Heptachlorodibenzo-p-dioxin Octachlorodibenzo-p-dioxin 2,3,7,8-Tetrachlorodibenzofuran 1,2,3,7,8-Pentachlorodibenzofuran 2,3,4,7,8-Pentachlorodibenzofuran 1,2,3,4,7,8-Hexachlorodibenzofuran 1,2,3,6,7,8-Hexachlorodibenzofuran 1,2,3,7,8,9-Hexachlorodibenzofuran 2,3,4,6,7,8-Hexachlorodibenzofuran 1,2,3,4,6,7,8-Hexachlorodibenzofuran 1,2,3,4,7,8,9-Hexachlorodibenzofuran Octachlorodibenzofuran
40 200 200 200 200 200 400 40 200 200 200 200 200 200 200 200 400
1995; Kim et al., 1997; Anderson et al., 1999; Jones and Anderson, 1999). Because PAHs and chlorinated compounds often co-occur in environmental mixtures, it is advantageous to be able to differentiate between these contaminant types in extracts of sediment, soil, and tissue. In a study of soils containing dioxins, a decrease in luciferase production by the RGS cell was evident from 6 to 16 h when PAHs co-occurred at high concentrations in the samples (Jones and Anderson, 1998). This study employed the P450 Reporter Gene System (RGS) to investigate the differences in luciferase production by the RGS cell at 6 and 16 h in response to individual CYP1A1-inducing compounds, including PAHs and PCBs, as well as mixtures of PAHs and dioxins and furans alone and in combination. Ellipticine, which inhibits the metabolic activity of CYP1A1 enzymes, was added to the cells to test the hypothesis that the metabolism of PAHs is causing the decrease in luciferase production from 6–16 h. Soil samples collected near a municipal incinerator and containing low levels of dioxins were also analyzed, before and after cleanup to remove PAHs, to examine the use of RGS testing at 6 and 16 h to characterize the types of contaminants present in environmental samples.
2. Materials and methods
2.1. Test toxicants Individual PAHs, PCB congeners, TCDD, and the Aroclor 1260 solution were obtained from Ultra Scientific (North Kingstown, RI). The PAH mixture (Table 1) was obtained from Supelco (Bellefonte, PA), and the dioxin/furan mixture (Table 2) from Cambridge Isotope Laboratories (Andover, MA). Ellipticine (5,11– Dimethyl-6H-pyrido[4,3-b]carbazole) was obtained from Sigma (St. Louis, MO). All solutions were prepared in HPLC-grade dimethylsulfoxide (DMSO) (Fisher, Pittsburgh, PA).
2.2. En6ironmental samples Two soil samples that were suspected of containing ash contaminated with dioxins and furans were collected from near a municipal incinerator. Samples (100 g) were extracted using EPA Method 3540C: soxhlet extraction using dichloromethane for 16 h. Dichloromethane extracts were then filtered through a sodium sulfate drying column and concentrated in a Kuderna Danish at 70°C. Following concentration, the extracts were solvent exchanged to hexane and adjusted to a 5 ml volume. Extracts were then washed with concentrated sulfuric acid, 5% sodium chloride, 20% sodium hydroxide, and a final wash with 5% sodium chloride. After completion of this initial cleanup, the
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
extracts were split in two equal portions. One portion was concentrated to 0.1 ml for RGS analysis. These extracts are referred to as c1-NC and c 2-NC to indicate that a column cleanup was not performed. The remaining portions were cleaned using a twocolumn system stacked in series following EPA Method 8280A. The first column contained multilayers of silica gel and sulfuric acid deactivated silica gel. The second column contained activated alumina. The extracts were applied to the columns and eluted with hexane. The silica gel column was used to remove polycyclic aromatic hydrocarbons (PAH) and polar breakdown products from the sulfuric acid/sodium hydroxide rinses. The alumina column was used to retain dioxins and furans during the initial elution of the columns. Following elution with hexane, the silica gel column was taken out of the elution sequence. Dioxins and furans were eluted from the alumina column using 20% dichloromethane/80% hexane. The eluent was collected and concentrated to 0.1 ml for RGS analysis. These extracts are referred to as c1-C and c 2-C.
121
MgCl2. Reactions were initiated by injection of 100 ml of luciferin, dissolved in 0.1 M potassium phosphate buffer, pH 7.8. Luminescence in relative light units (RLUs) was measured using a ML2250 Luminometer (Dynatech Laboratories, Chantilly, VA). Luciferase assay buffers were purchased from PharMingen (San Diego, CA). With each test run, a solvent blank (using a volume of DCM equal to the sample volume being tested) and a reference toxicant (1 ng/ml TCDD) were also applied to two separate replicate wells. Mean fold induction of the solvent blank was set equal to 1, and the fold induction of each standard solution and sample were determined by dividing the mean RLUs produced by that solution by the mean RLUs produced by the solvent blank. If fold induction of a sample exceeded 100, the test was repeated using serial dilutions, and the dilution factor was incorporated into the equivalency calculations described below. The coefficient of variation among replicates was acceptable if B20%. Thus, all data described in the figures are means of replicates varying by less than 20%.
2.3. Cell culture and application of test toxicants 2.5. Equi6alency calculations The 101L cells were grown as monolayers in an atmosphere of 5% CO2 and 100% humidity at 37°C in Eagle’s Minimum Essential Media (Mediatech, Herndon, VA), supplemented with 10% fetal bovine serum, 2% L-glutamine, 1% sodium pyruvate, and 0.4 mg/ml Geneticin (all from Sigma, St. Louis, MO). Cells were trypsinized upon reaching a maximum of 90% confluence, and discarded after 24 passages. In preparation for testing, cells were subcultured into six-well plates at a density of 2.5 ×105 cells/well and grown for 36 h in the environment described above to reach a density of approximately 1×106 cells/well. Standard solutions and environmental sample extracts were applied at volumes of 2 – 10 ml to two replicate wells each containing 2 ml of culture media. The concentrations of these solutions never exceeded 0.5% (v/ v). In most cases, duplicate plates were dosed, so that one plate was incubated for 6 h, and the other for 16 h.
The dioxin/furan mixture used in this study is composed of seventeen analytes (Table 2). Using the concentration of each analyte and its Toxic Equivalency Factor (TEF), a Chemical Toxic Equivalency (Chem TEQ) was calculated for the mixture (Van den Berg et al., 1998). From the concentration-response curve of this mixture, the RGS fold induction response is approximately equal to the mixture Chem TEQ in pg/ml. Based on this, an RGS TEQ (termed so because it is based on the RGS response, not the chemical concentration) for environmental samples is calculated using the following equation, where Ve is the total extract volume, Va is the volume of extract applied to cells, and Wd is the dry weight of the sample. A factor of 1000 yields RGS TEQ in ng/g.
2.4. Luciferase assay
2.6. Chemical analyses of en6ironmental samples
The detailed methodology used in this study has been described elsewhere (APHA, 1996; ASTM, 1999). After 6 or 16 h incubation with the test solutions, the cells were washed with Hank’s Balanced Salt Solution (Mediatech, Herndon, VA), and lysed with 200 ml of buffer containing 1% Triton, 25 mM Tricine, pH 7.8, 15 mM MgSO4, 4 mM EDTA, and 1 mM dithiothreitol (DTT). Cell lysates were centrifuged at 6000 rpm for 10 s, and 50 ml of the supernatant was applied to a 96-well plate, followed by 100 ml of 0.1 M potassium phosphate buffer, pH 7.8, containing 5 mM ATP and 10 mM
Analyses of dioxins and furans in soil samples were performed by Columbia Analytical Services in Houston, TX, using EPA Method 8290.
RGS TEQ=(fold induction/1000) × ((Ve/Va)/Wd)
3. Results
3.1. Analyses of standards and standard mixtures RGS fold induction responses to TCDD (2 ng/ml) and a standard PAH mixture (83 ng/ml) were de-
122
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
Fig. 1. RGS fold induction responses to TCDD and a PAH mixture over time. Each point represents the mean of at least two replicate analyses.
Fig. 2. RGS responses to PAH mixture and TCDD with and without the inhibitor ellipticine (100 mM). Each column represents the mean of at least two analyses. Responses to the solvent dimethyl sulfoxide (DMSO)+ellipticine are given as a negative control.
peared to induce maximum responses from PAHs and a representative chlorinated compound, TCDD, respectively, these two time periods were chosen for further study. When the CYP1A1-inhibitor ellipticine was added (100 mM), the 16 h fold induction response to the PAH mixture increased, while the 6 h response was unchanged (Fig. 2). The 16 h fold induction response to TCDD decreased somewhat in the presence of inhibitor. The 6 h fold induction responses to three high molecular weight PAHs (dibenz[a,h]anthracene, benzo[a]pyrene, and indeno[1,2,3-cd]pyrene) were around four times greater than at 16 h (Fig. 3). Benzo[k]fluoranthene, the most potent PAH to RGS tested thus far (Jones and Anderson, 1999), decreased less than two times from 6–16 h. In contrast, fold induction responses to three coplanar PCB congeners, 3,3%,4,4%TetraCB (c 77), 3,4,4%,5-TetraCB (c 81) and 3,3%,4,4%,5-PentaCB (c 126) and Aroclor 1260 were greater at 16h than at 6h in every case, with the most potent PCB congener (c 81) increasing around four times from 6 to 16 h (Fig. 3). A PAH mixture and a dioxin/furan mixture were tested individually and in combination (Fig. 4). The response to the dioxin/furan mixture alone (0.04 ng/ml) increased to reach a maximum at 16 h that was sustained at 20 h, while the response to the PAH mixture alone (42 ng/ml) was maximal at 6 h, and then began to decrease. When the two mixtures were applied to the cells concurrently at the previous concentrations, the response was maximal at 6 h, dropped off to a low at 12 h, and then increased again by 16–20 h, though not reaching the level exhibited at 6 h. Different proportions of the mixture combinations were then tested at 6 and 16 h (Fig. 5). The first pair of bars resulted from the addition of the PAH mixture alone, while the other pairs show the responses to increasing concentrations of the dioxin/furan mixture with the concentration of the PAH mixture consistent throughout. When the dioxin/furan mixture concentra-
Fig. 3. Responses to individual PAHs, PCBs, and the PCB mixture Aroclor 1260 at 6 and 16 h. Bars represent the mean of at least two analyses.
tectable (five-fold induction) by 2 h (Fig. 1). The response to TCDD steadily increased to reach a maximum fold induction of around 100 at 16 h. In contrast, the response to the PAH mixture increased to a maximum fold induction of around 50 at 6 h, then steadily decreased. Because 6 and 16 h exposure periods ap-
Fig. 4. Responses to a mixture of dioxins and furans, a PAH mixture, and the two mixtures in combination (at the same concentrations shown individually). Columns represent the mean of at least two analyses. Mixture compositions are available as supporting information.
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
123
16 h, typical of the response to chlorinated compounds alone. This procedure was repeated, maintaining the dioxin/ furan mixture at 0.09 ng/ml and applying increasing concentrations of the PAH mixture (Fig. 6). In this case, the fold induction responses were greater at 16 h until the PAH concentration was increased to 41.5 ng/ml. Typical of the response to PAHs alone, the 6 h response was greater than the 16 h response with this combination.
3.2. Analyses of en6ironmental samples
Fig. 5. Responses to the PAH mixture and dioxin/furan mixture in combination. The concentration of the PAH mixture is held constant, with increasing concentrations of the dioxin/furan mixture as indicated. Columns represent the mean of at least two analyses.
Fig. 6. Responses to the mixture combinations, with the dioxin/furan concentration held constant and increasing concentrations of the PAH mixture as indicated. Columns represent the mean of at least two analyses.
Analyses of the cleaned and uncleaned extracts of two soil samples are shown in Fig. 7. The 16 h responses to both c 1-NC and c 1-C are greater than the 6 h responses, which indicated that these extracts contained primarily chlorinated compounds. In contrast, the 16 h response to c2-NC was much less than the 6 h response, indicating that this extract contained primarily PAHs. Extract c2-C, which had gone through column cleanup, produced much lower responses at both times, with no difference between 6 and 16 h responses. After removal of the PAHs, this extract produced responses that indicated the presence of only small amounts of chlorinated compounds. As described in Section 2.5, the 16 h RGS responses were used to calculate RGS TEQ for Extracts c1-C and c2-C as 48.5 and 1.5 pg/g, respectively. From the EPA Method 8290 chemical analyses, the Chemical TEQ was calculated as the sum of each analyte concentration multiplied by its TEF (Table 3). The calculated Chem TEQ for Samples c 1 and c 2 were 61 and 0.8 pg/g, respectively.
4. Discussion
Fig. 7. Responses to extracts of two environmental soil/ash samples. ‘NC’ indicates that no cleanup was performed, ‘C’ extracts were cleaned using EPA Method 8280A to remove PAHs. Extracts were applied to the cells in equal volumes. Columns represent the mean of at least two analyses.
tion increased from 0.04 to 0.09 ng/ml, the ratio of 6/16 h responses was altered. At the 0.09 ng/ml concentration and higher, greater responses were observed at
The induction of the CYP1A1 gene is mediated by ligand binding at the cytosolic Ah-receptor, which is associated with other proteins, and further controlled by the translocation of the receptor–inducer complex to the nucleus and binding to dioxin-responsive elements (Whitlock, 1990; Pongratz et al., 1992). Both PAHs and chlorinated compounds, including coplanar PCBs, dioxins and furans, act via this pathway. In assessment of environmental mixtures, which may contain both PAHs and chlorinated compounds, it is beneficial to be able to differentiate between them. The human cell line, 101L, is used to study the mechanisms of Ah-receptor-mediated induction of CYP1A1 because of its sensitivity and facility (Postlind et al., 1993; Chen and Tukey, 1996). The P450 Reporter Gene System (RGS) applies the 101L cell assay to screen environmental samples for the presence of con-
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
124
taminants which induce CYP1A1 gene expression. In this study, RGS was used to investigate the temporal differences in luciferase production through CYP1A1 promoter activity by PAHs and chlorinated compounds, and how this may be used in assessment of environmental samples. It is apparent that the RGS fold induction response to PAHs is temporally different from that of the reference inducer TCDD (Fig. 1). The response to PAHs reaches maximal at an earlier time and then decreases, while the TCDD response reaches maximal at a later time and is more sustained. Individual PAHs exhibit the same response trends as the PAH mixture, and individual PCB congeners, as well as the Aroclor 1260 mixture, act in the same manner as TCDD (Fig. 3). Ellipticine, a reversible-type (competitive) inhibitor of human CYP1A1, inhibits the metabolic activity of CYP1A1 enzymes by binding strongly to the reduced heme iron and thereby preventing the fixation and activation of molecular dioxygen (Lesca et al., 1979). Inhibition occurs after activation of the CYP1A1 promoter region, so that it does not affect the production of luciferase in these cells, but rather inhibits the endogenous metabolic activity of CYP1A1 enzymes. The addition of ellipticine to the cell media increased the production of PAH-induced luciferase expression at 16 h, while the 6 h responses with and without ellipticine were equivalent (Fig. 2). This supports the assumption that the metabolism of PAHs in these cells is causing the luciferase production to decrease after 6 h. The response to TCDD was not increased at either time period.
A complex mixture of both PAHs and dioxins and furans exhibits additivity of response over time (Fig. 4), supporting the use of RGS as a measure of the total amount of CYP1A1-inducing contaminants present. The 6/16 h response patterns vary depending on the concentrations of PAHs compared to dioxins and furans. When the dioxin/furan mixture concentration was increased from 0.04 to 0.09 ng/ml, while the PAH mixture concentration was maintained at 16.6 ng/ml, the response ratio (6 h response divided by 16 h response) decreased from \ 1 to B 1 (Fig. 5). This occurred not only because the 16 h response was increased with increased dioxin/furan concentration, but also because the 6h response decreased. When higher concentrations of CYP1A1-inducing compounds were present, the metabolism of PAHs was apparently also increased, so that less was available to induce the production of luciferase at 6 h. Alternatively, the presence of increasing amounts of dioxins and furans, which have higher receptor and binding affinity, may have masked the 6 h response by PAHs. Fig. 6 shows the reverse effect, so that when the PAH mixture concentration was increased from 16.6 to 41.5 ng/ml while the dioxin/furan mixture concentration was held constant at 0.09 ng/ml, the 6/16 h response ratio increased from B 1 to \ 1. In addition to the increase in 6 h response with this combination, the 16 h response was decreased. The reason for the decrease is not known, but it is apparent that as the PAH concentration increases, the 6 h response will also increase and at some point will dominate the response pattern. Further studies using ellipticine to inhibit the metabolism
Table 3 Chemical analysis of environmental samples by EPA method 8290 Analyte
2,3,7,8-Tetrachlorodibenzo-p-dioxin 1,2,3,7,8-Pentachlorodibenzo-p-dioxin 1,2,3,4,7,8-Hexachlorodibenzo-p-dioxin 1,2,3,6,7,8-Hexrachlorodibenzo-p-dioxin 1,2,3,7,8,9-Hexachlorodibenzo-p-dioxin 1,2,3,4,6,7,8-Heptachlorodibenzo-p-dioxin Octachlorodibenzo-p-dioxin 2,3,7,8-Tetrachlorodibenzofuran 1,2,3,7,8-Pentachlorodibenzofuran 2,3,4,7,8-Pentachlorodibenzofuran 1,2,3,4,7,8-Hexachlorodibenzofuran 1,2,3,6,7,8-Hexachlorodibenzofuran 1,2,3,7,8,9-Hexachlorodibenzofuran 2,3,4,6,7,8-Hexachlorodibenzofuran 1,2,3,4,6,7,8-Hexachlorodibenzofuran 1,2,3,4,7,8,9-Hexachlorodibenzofuran Octachlorodibenzofuran Chem TEQ (pg/g) a
TEFa
1 0.1 0.1 0.1 0.1 0.01 0.001 0.1 0.05 0.5 0.1 0.1 0.1 0.1 0.01 0.01 0.001
Sample c1
Sample c2
(pg/g)
(pg/g)*TEF
(pg/g)
(pg/g)*TEF
0 14 26.5 60.7 74.7 595.7 2332.1 22.9 8.4 17.5 65.3 22.8 65.2 0 200.8 25.5 349.4
0 14.0 2.7 6.1 7.5 6.0 2.3 2.3 0.4 8.8 6.5 2.3 6.5 0.0 2.0 0.3 0.3
0 0 0 0 0 12.1 98.6 0 0 0 3 1.7 0 0 12.1 0 13.8
0 0 0 0 0 0.1 0.1 0 0 0 0.3 0.2 0 0 0.1 0 0.01
68
Toxic equivalency factors, humans and mammals (Van den Berg et al., 1998).
0.8
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
of PAHs may help to elucidate the mechanisms involved. The RGS response patterns to standard mixtures individually and in combination have important implications for the assessment of the type(s) of contaminants present in unknown environmental samples. Unpublished laboratory data have demonstrated that the RGS responses are 70 – 90% of what would be predicted from the combination of two to four inducing PAHs or coplanar PCBs. RGS analyses of sediment samples from coastal areas of the US, tested for the National Oceanic and Atmospheric Administration, have shown high correlations with subsequent chemical determinations for total high molecular weight PAHs (Anderson et al., 1999). If the RGS responses were not a function of the additive induction from multiple inducers, such strong correlations with chemical analyses would not be observed. The two soil samples analyzed in this study exhibit straightforward response patterns. Sample c 1 showed a response pattern typical of chlorinated compounds, with the 16 h response greater than the 6 h in both the original c 1-NC and cleaned extract c1-C. In contrast, PAHs were identified in Sample c 2 because the 6 h response to extract c2-NC was greater than at 16 h. When PAHs were removed, the response pattern of c 2-C changed dramatically to indicate the presence of only small amounts of inducers. The 16 h responses to extracts c1-C and c 2-C were used to calculate RGS TEQ values of 48.5 and 1.5 pg/g, respectively, corresponding well to the Chem TEQ derived by EPA Method 8290 (68 and 0.8 pg/g, respectively). When environmental samples contain more complex mixtures of both PAHs and chlorinated compounds, the use of RGS analysis at 6 and 16 h to differentiate between contaminant types is more difficult. Recent testing of sediment extracts from New York Harbor, which contained as high as 32 mg/g of PAHs and pg/g levels of dioxins and furans, consistently exhibited response patterns typical of PAHs. Nonetheless, the 16 h responses yielded RGS TEQ values that were highly correlated (R 2 =0.81) with the Chem TEQ values calculated from EPA Method 8290 confirmation of dioxins and furans. The use of the cleanup procedures described here in Section 2.2 for removal of PAHs is advantageous when RGS is employed specifically to screen for chlorinated compounds. However, the extra time and expense necessary for cleanup procedures is often not warranted in projects where a rapid and inexpensive screening test such as RGS is used. We recommend a tiered approach in such investigations, since the responses of the RGS assay will identify the presence of many of the organic chemicals of concern. After the first tier of RGS testing (testing all samples at 16 h), it is possible to eliminate some stations from further analysis and focus attention on
125
the most contaminated sites. The next tier would then utilize the 6/16 h time intervals to identify which classes of inducing compounds are likely to be present. Results of these tests will help to direct subsequent chemical analyses. The analytical results would then be compared to RGS findings, and a strong correlation would demonstrate that the levels of key contaminants in samples not chemically characterized could be estimated from the regression curve. It should be cautioned that RGS is highly specific only for those contaminants with affinity for the Ah receptor; other types of toxic chemicals would not be detected. RGS was used in a study of 75 sediment samples from two rivers in the Pacific Northwest for the US Army Corps of Engineers (US ACOE, 1999). Most samples produced greater RGS responses at 6 than at 16 h, indicating the presence of PAHs, which were then determined by GC/MS analysis. One sample produced a significantly greater RGS response at 16 h than at 6 h, and this sample was selected for further chemical analysis for the presence of chlorinated compounds. As a result, PCBs were identified in this sample. In conclusion, the analysis of environmental samples at both 6 and 16 h using RGS yields important information regarding the type of CYP1A1-inducing compounds present. When responses differ considerably at the two time periods, characterization between PAHs and chlorinated compounds is more straightforward. In contrast, 6 and 16 h responses that are more similar are indicative of the presence of both PAHs and chlorinated compounds, which is also important information in environmental assessment. Further, by examining the 6/16 h response patterns of all samples from a site, differences among samples are often discovered, so that PAHs and/or chlorinated inducers are identified where they were not suspected. Chemical confirmation of a representative number of these samples (: 15–20%) can then be advised. Acknowledgements We thank Joe Wiegel for sample extract preparation and clean-up. We are also grateful to Hiyoshi Corporation, Shiga, Japan, for providing soil samples. This work was supported by Columbia Analytical Services, Kelso, WA. References Anderson, J.W., Rossi, S.S., Tukey, R.H., et al., 1995. A biomarker, P450RGS, for assessing the potential toxicity of organic compounds in environmental samples. Environ. Toxicol. Chem. 14, 1159. Anderson, J.W., Jones, J.M., Hameedi, J., et al., 1999. Comparative analysis of sediment extracts from NOAA’s bioeffects studies by the biomarker, P450 RGS. Mar. Environ. Res. 48, 407.
126
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
APHA, P450 reporter gene response to dioxin-like organics. Method 8070, Standard Methods for the Examination of Water and Wastewater, 19th ed., Supplement, American Public Health Association, Washington, DC, 1996, p. 24. ASTM, E 1853M-96 Standard Guide for measuring the presence of planar organic compounds which induce CYP1A, using reporter gene test systems, Biological Effects and Environmental Fate; Biotechnology; Pesticides, 1997 Annual Book of ASTM Standards, vol. 11.05-Water and Environmental Technology, American Society for Testing and Materials, West Conshohocken, PA, 1999, p. 1400. Bronstein, I., 1994. Chemiluminescent and bioluminescent reporter gene assays. Anal. Biochem. 219, 169. Celander, M., Bremer, J., Hahn, M.E., et al., 1997. Glucocorticoid– xenobiotic interactions: dexamethasone-mediated potentiation of cytochrome P4501A induction by (b-naphthoflavone in a fish hepatoma cell line (PLHC-1). Environ. Toxicol. Chem. 16, 900. Chen, Y., Tukey, R.H., 1996. Protein kinase C modulates regulation of the CYP1A1 gene by the aryl hydrocarbon receptor. J. Biol. Chem. 271, 26261. Hahn, M.E., Stegeman, J.J., 1994. Regulation of cytochrome P4501A1 in teleosts: sustained induction of CYP1A1 mRNA, protein, and catalytic activity by 2,3,7,8-tetrachlorodibenzofuran in the marine fish Stenotomus chrysops. Toxicol. Appl. Pharmacol. 127, 187. Jones, J.M., Anderson, J.W., 1998. Cost-effective site characterization using P450 reporter gene system (RGS). In: Wickramanayake, G.B., Hinchee, R.E. (Eds.), Risk, Resource, and Regulatory Issues. Battelle, Columbus, OH, p. 1. Jones, J.M., Anderson, J.W., 1999. Relative potencies of PAHs and PCBs based on the response of human cells. Environ. Toxicol. Pharmacol. 7, 19. Kim, G.B., Anderson, J.W., Bothner, K., et al., 1997. Application of P450RGS (reporter gene system) as a bioindicator of sediment PAH contamination in the vicinity of Incheon Harbor, Korea. Biomarkers 2, 181. Kloepper-Sams, P., Stegeman, J.J., 1989. The temporal relationship between P450E protein content, catalytic activity, and mRNA levels in the teleost Fundulus heteroclitus following treatment with b-naphthoflavone. Arch. Biochem. Biophys. 268, 525.
.
Lesca, P., Rafidinarivo, E., Lecointe, P., et al., 1979. A class of strong inhibitors of microsomal monooxygenases: the ellipticines. Chem. Biol. Interactions 24, 189. Pesonen, M., Goksoyr, A., Andersson, T., 1992. Expressions of P4501A1 in a primary culture of rainbow trout hepatocytes exposed to (b-naphthoflavone or 2,3,7,8-tetrachlorodibenzo-pdioxin. Arch. Biochem. Biophys. 292, 228. Pongratz, I., Mason, G., Poellinger, L., 1992. Dual roles of the 90 kDa heat shock protein hsp90 in modulating functional activities of the dioxin receptor. J. Biol. Chem. 267, 13728. Postlind, H., Vu, T., Tukey, R.H., et al., 1993. Response of human CYP1-luciferase plasmids to 2,3,7,8-tetrachlorodibenzo-p-dioxin and polycyclic aromatic hydrocarbons. Toxicol. Appl. Pharmacol. 118, 255. Stegeman, J.J., Brouwer, M., DiGiulio, R.T., et al., 1992. Molecular responses to environmental contamination: enzyme and protein systems as indicators of chemical exposure and effect. In: Huggett, R.J., Kimerle, R.A., Mehrle, P.M., Bergman, H.L. (Eds.), Biomarkers: Biochemical, Physiological, and Histological Markers of Anthropogenic Stress. Lewis, Boca Rotan, FL, p. 235. Stegeman, J.J., Hahn, M.E., Weisbrod, R., et al., 1995. Induction of cytochrome P4501A1 by aryl hydrocarbon receptor agonists in porcine aorta endothelial cells in culture and cytochrome P4501A1 activity in intact cells. Mol. Pharmacol. 47, 296. Thomson, J.F., Hayes, L.S., Lloyd, D.B., 1991. Modulation of firefly luciferase stability and impact of studies of gene regulation. Gene 103, 171. US Army Corps of Engineers, Integrated Feasibility Report for Channel Improvements and Environmental Impact Statement; Columbia and Lower Willamette River Federal Navigation Channel, Final Report of the US ACOE, Portland District, August, 1999. Van den Berg, M., et al., 1998. Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environmental Health Perspectives 106, 775. Whitlock, J.P. Jr, 1990. Genetic and molecular aspects of 2,3,7,8-tetrachlorodibenzo-p-dioxin action. Annu. Rev. Pharmacol. Toxicol. 30, 251.
www.elsevier.com/locate/etap
Using the metabolism of PAHs in a human cell line to characterize environmental samples Jennifer M. Jones a, Jack W. Anderson a,*, Robert H. Tukey b a
Columbia Analytical Ser6ices, 1185 Park Center Dri6e, Suite A, Vista, CA 92083, USA b Department of Pharmacology, Uni6ersity of California, San Diego, CA, USA
Received 22 October 1999; received in revised form 26 January 2000; accepted 3 February 2000
Abstract P450 reporter gene system (RGS) is an in vitro assay to detect compounds that activate the Ah receptor and induce cytochrome P450 (CYP1A1). This system utilizes a human cell (101L) stably transfected with a luciferase reporter downstream of human CYP1A1 promoter sequences. When CYP1A1-inducing compounds are present, luciferase is produced as well as endogenous CYP1A1 enzymes. Polycyclic aromatic hydrocarbons (PAHs) are more readily degraded than chlorinated compounds including dioxins, furans, and coplanar polychlorinated biphenyls (PCBs). PAH-induced luciferase production begins to decrease between 6 and 16 h, while chlorinated compounds produce a more sustained response. Individual and mixtures of CYP1A1-inducing compounds were tested at both 6 and 16 h. Extracts of soils containing both PAHs and dioxins were also tested, before and after cleanup to remove PAHs. Results indicate that RGS testing at 6 and 16 h is a promising tool to differentiate between PAHs and chlorinated hydrocarbons often co-occurring in environmental samples. © 2000 Elsevier Science B.V. All rights reserved. Keywords: CYP1A1; Ah receptor; PAHs; Chlorinated hydrocarbons; Co-occurrence
1. Introduction Polycyclic aromatic hydrocarbons (PAHs), as well as the polychlorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls (PCBs) are known to induce the CYP1A gene subfamily via the Ah-receptor (Whitlock, 1990). The subsequent production of CYP1A1 enzymes is an important pathway for metabolism and removal of these compounds from the cell. Although they share this common mode of action, the chlorinated inducers (dioxins, furans, and PCBs) have a lower rate of metabolism than PAHs. Shorter duration of CYP1A1 induction by PAHs compared to chlorinated compounds has been reported with in vitro studies using a fish hepatoma cell line (PLHC-1) (Celander et al., 1997), rainbow trout hepatocytes (Pesonen et al., 1992), and porcine aorta endothelial cells (Stegeman et al., 1995), as well as in vivo studies with marine killifish (Kloepper-Sams and Stegeman, * Corresponding author. Tel.: +1-760-7341450; fax: + 1-7607341409. E-mail address: [email protected] (J.W. Anderson)
1989). Hahn and Stegeman (1994) reported sustained induction of CYP1A1 mRNA by 2,3,7,8-TCDF in scup, contrasting with their previous findings of transient induction by nonhalogenated inducers. The prolonged induction of CYP1A1 mRNA by chlorinated hydrocarbons has been attributed to their slower metabolism and removal, leading to continued synthesis of mRNA (Stegeman et al., 1992). Postlind et al. (1993) reported similar results using 101L cells, which are human hepatoma cells stably integrated with a plasmid containing CYP1A1 sequences fused to the luciferase reporter gene. The production of luciferase in these cells is regulated by the dioxin-responsive elements of the CYP1A1 sequences, and is not controlled by endogenous feedback mechanisms in the cell that may affect CYP1A1 protein levels. Because luciferase has a short half-life in mammalian cells (0.5–3 h) (Thomson et al., 1991; Bronstein, 1994), the levels of luciferase activity may reflect short-term changes in the activity of the dioxin-responsive elements upstream. Postlind et al. (1993) reported that CYP1A1-mediated luciferase activity reached maximal levels and did not decrease between 12 and 24 h of
1382-6689/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 2 - 6 6 8 9 ( 0 0 ) 0 0 0 3 3 - 8
120
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
exposure to tetrachlorodibenzo-p-dioxin (TCDD). In contrast, luciferase activity reached maximal levels by 6 – 12 h of exposure to two PAHs (benzo[a]anthracene and benzo[a]pyrene), at which point levels dropped sharply. This decrease was presumably caused by metabolism of the PAHs and decrease of the concentrations of parent PAH present in the cells. The reporter gene system (RGS) is an application of the 101L cell line for use by industry and government agencies to screen environmental samples for the presence of CYP1A1-inducing compounds (Anderson et al., Table 1 Composition of polycyclic aromatic hydrocarbon (PAH) mixture PAH compound
Conc. (mg/ml)
Benzo[k]flouranthene Dibenzo[a,h]anthracene Benzo[b]flourathene Ideno[1,2,3-cd]pyrene Benzo[a]pyrene Benzo[a]anthracene Chrysene Benzo[g,h,i ]perylene Acenapthene Acenaphthylene Anthracene Benzo[e]pyrene Biphenyl Dibenzothiophene Fluoranthene Fluorene Naphthalene Perylene Phenanthrene Pyrene 1-Methylnaphthalene 2-Methylnaphthalene
13.7 4.7 13.7 10.5 21.2 14.8 15.9 13.5 1.7 4.2 5.3 16.4 14.7 2.3 16.1 3.8 15.0 22.4 13.6 29.6 45.7 33.5
Table 2 Composition of dioxin/furan mixture Analyte
Conc. (ng/ml)
2,3,7,8-Tetrachlorodibenzo-p-dioxin 1,2,3,7,8-Pentachlorodibenzo-p-dioxin 1,2,3,4,7,8-Tetrachlorodibenzo-p-dioxin 1,2,3,6,7,8-Hexrachlorodibenzo-p-dioxin 1,2,3,7,8,9-Hexachlorodibenzo-p-dioxin 1,2,3,4,6,7,8-Heptachlorodibenzo-p-dioxin Octachlorodibenzo-p-dioxin 2,3,7,8-Tetrachlorodibenzofuran 1,2,3,7,8-Pentachlorodibenzofuran 2,3,4,7,8-Pentachlorodibenzofuran 1,2,3,4,7,8-Hexachlorodibenzofuran 1,2,3,6,7,8-Hexachlorodibenzofuran 1,2,3,7,8,9-Hexachlorodibenzofuran 2,3,4,6,7,8-Hexachlorodibenzofuran 1,2,3,4,6,7,8-Hexachlorodibenzofuran 1,2,3,4,7,8,9-Hexachlorodibenzofuran Octachlorodibenzofuran
40 200 200 200 200 200 400 40 200 200 200 200 200 200 200 200 400
1995; Kim et al., 1997; Anderson et al., 1999; Jones and Anderson, 1999). Because PAHs and chlorinated compounds often co-occur in environmental mixtures, it is advantageous to be able to differentiate between these contaminant types in extracts of sediment, soil, and tissue. In a study of soils containing dioxins, a decrease in luciferase production by the RGS cell was evident from 6 to 16 h when PAHs co-occurred at high concentrations in the samples (Jones and Anderson, 1998). This study employed the P450 Reporter Gene System (RGS) to investigate the differences in luciferase production by the RGS cell at 6 and 16 h in response to individual CYP1A1-inducing compounds, including PAHs and PCBs, as well as mixtures of PAHs and dioxins and furans alone and in combination. Ellipticine, which inhibits the metabolic activity of CYP1A1 enzymes, was added to the cells to test the hypothesis that the metabolism of PAHs is causing the decrease in luciferase production from 6–16 h. Soil samples collected near a municipal incinerator and containing low levels of dioxins were also analyzed, before and after cleanup to remove PAHs, to examine the use of RGS testing at 6 and 16 h to characterize the types of contaminants present in environmental samples.
2. Materials and methods
2.1. Test toxicants Individual PAHs, PCB congeners, TCDD, and the Aroclor 1260 solution were obtained from Ultra Scientific (North Kingstown, RI). The PAH mixture (Table 1) was obtained from Supelco (Bellefonte, PA), and the dioxin/furan mixture (Table 2) from Cambridge Isotope Laboratories (Andover, MA). Ellipticine (5,11– Dimethyl-6H-pyrido[4,3-b]carbazole) was obtained from Sigma (St. Louis, MO). All solutions were prepared in HPLC-grade dimethylsulfoxide (DMSO) (Fisher, Pittsburgh, PA).
2.2. En6ironmental samples Two soil samples that were suspected of containing ash contaminated with dioxins and furans were collected from near a municipal incinerator. Samples (100 g) were extracted using EPA Method 3540C: soxhlet extraction using dichloromethane for 16 h. Dichloromethane extracts were then filtered through a sodium sulfate drying column and concentrated in a Kuderna Danish at 70°C. Following concentration, the extracts were solvent exchanged to hexane and adjusted to a 5 ml volume. Extracts were then washed with concentrated sulfuric acid, 5% sodium chloride, 20% sodium hydroxide, and a final wash with 5% sodium chloride. After completion of this initial cleanup, the
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
extracts were split in two equal portions. One portion was concentrated to 0.1 ml for RGS analysis. These extracts are referred to as c1-NC and c 2-NC to indicate that a column cleanup was not performed. The remaining portions were cleaned using a twocolumn system stacked in series following EPA Method 8280A. The first column contained multilayers of silica gel and sulfuric acid deactivated silica gel. The second column contained activated alumina. The extracts were applied to the columns and eluted with hexane. The silica gel column was used to remove polycyclic aromatic hydrocarbons (PAH) and polar breakdown products from the sulfuric acid/sodium hydroxide rinses. The alumina column was used to retain dioxins and furans during the initial elution of the columns. Following elution with hexane, the silica gel column was taken out of the elution sequence. Dioxins and furans were eluted from the alumina column using 20% dichloromethane/80% hexane. The eluent was collected and concentrated to 0.1 ml for RGS analysis. These extracts are referred to as c1-C and c 2-C.
121
MgCl2. Reactions were initiated by injection of 100 ml of luciferin, dissolved in 0.1 M potassium phosphate buffer, pH 7.8. Luminescence in relative light units (RLUs) was measured using a ML2250 Luminometer (Dynatech Laboratories, Chantilly, VA). Luciferase assay buffers were purchased from PharMingen (San Diego, CA). With each test run, a solvent blank (using a volume of DCM equal to the sample volume being tested) and a reference toxicant (1 ng/ml TCDD) were also applied to two separate replicate wells. Mean fold induction of the solvent blank was set equal to 1, and the fold induction of each standard solution and sample were determined by dividing the mean RLUs produced by that solution by the mean RLUs produced by the solvent blank. If fold induction of a sample exceeded 100, the test was repeated using serial dilutions, and the dilution factor was incorporated into the equivalency calculations described below. The coefficient of variation among replicates was acceptable if B20%. Thus, all data described in the figures are means of replicates varying by less than 20%.
2.3. Cell culture and application of test toxicants 2.5. Equi6alency calculations The 101L cells were grown as monolayers in an atmosphere of 5% CO2 and 100% humidity at 37°C in Eagle’s Minimum Essential Media (Mediatech, Herndon, VA), supplemented with 10% fetal bovine serum, 2% L-glutamine, 1% sodium pyruvate, and 0.4 mg/ml Geneticin (all from Sigma, St. Louis, MO). Cells were trypsinized upon reaching a maximum of 90% confluence, and discarded after 24 passages. In preparation for testing, cells were subcultured into six-well plates at a density of 2.5 ×105 cells/well and grown for 36 h in the environment described above to reach a density of approximately 1×106 cells/well. Standard solutions and environmental sample extracts were applied at volumes of 2 – 10 ml to two replicate wells each containing 2 ml of culture media. The concentrations of these solutions never exceeded 0.5% (v/ v). In most cases, duplicate plates were dosed, so that one plate was incubated for 6 h, and the other for 16 h.
The dioxin/furan mixture used in this study is composed of seventeen analytes (Table 2). Using the concentration of each analyte and its Toxic Equivalency Factor (TEF), a Chemical Toxic Equivalency (Chem TEQ) was calculated for the mixture (Van den Berg et al., 1998). From the concentration-response curve of this mixture, the RGS fold induction response is approximately equal to the mixture Chem TEQ in pg/ml. Based on this, an RGS TEQ (termed so because it is based on the RGS response, not the chemical concentration) for environmental samples is calculated using the following equation, where Ve is the total extract volume, Va is the volume of extract applied to cells, and Wd is the dry weight of the sample. A factor of 1000 yields RGS TEQ in ng/g.
2.4. Luciferase assay
2.6. Chemical analyses of en6ironmental samples
The detailed methodology used in this study has been described elsewhere (APHA, 1996; ASTM, 1999). After 6 or 16 h incubation with the test solutions, the cells were washed with Hank’s Balanced Salt Solution (Mediatech, Herndon, VA), and lysed with 200 ml of buffer containing 1% Triton, 25 mM Tricine, pH 7.8, 15 mM MgSO4, 4 mM EDTA, and 1 mM dithiothreitol (DTT). Cell lysates were centrifuged at 6000 rpm for 10 s, and 50 ml of the supernatant was applied to a 96-well plate, followed by 100 ml of 0.1 M potassium phosphate buffer, pH 7.8, containing 5 mM ATP and 10 mM
Analyses of dioxins and furans in soil samples were performed by Columbia Analytical Services in Houston, TX, using EPA Method 8290.
RGS TEQ=(fold induction/1000) × ((Ve/Va)/Wd)
3. Results
3.1. Analyses of standards and standard mixtures RGS fold induction responses to TCDD (2 ng/ml) and a standard PAH mixture (83 ng/ml) were de-
122
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
Fig. 1. RGS fold induction responses to TCDD and a PAH mixture over time. Each point represents the mean of at least two replicate analyses.
Fig. 2. RGS responses to PAH mixture and TCDD with and without the inhibitor ellipticine (100 mM). Each column represents the mean of at least two analyses. Responses to the solvent dimethyl sulfoxide (DMSO)+ellipticine are given as a negative control.
peared to induce maximum responses from PAHs and a representative chlorinated compound, TCDD, respectively, these two time periods were chosen for further study. When the CYP1A1-inhibitor ellipticine was added (100 mM), the 16 h fold induction response to the PAH mixture increased, while the 6 h response was unchanged (Fig. 2). The 16 h fold induction response to TCDD decreased somewhat in the presence of inhibitor. The 6 h fold induction responses to three high molecular weight PAHs (dibenz[a,h]anthracene, benzo[a]pyrene, and indeno[1,2,3-cd]pyrene) were around four times greater than at 16 h (Fig. 3). Benzo[k]fluoranthene, the most potent PAH to RGS tested thus far (Jones and Anderson, 1999), decreased less than two times from 6–16 h. In contrast, fold induction responses to three coplanar PCB congeners, 3,3%,4,4%TetraCB (c 77), 3,4,4%,5-TetraCB (c 81) and 3,3%,4,4%,5-PentaCB (c 126) and Aroclor 1260 were greater at 16h than at 6h in every case, with the most potent PCB congener (c 81) increasing around four times from 6 to 16 h (Fig. 3). A PAH mixture and a dioxin/furan mixture were tested individually and in combination (Fig. 4). The response to the dioxin/furan mixture alone (0.04 ng/ml) increased to reach a maximum at 16 h that was sustained at 20 h, while the response to the PAH mixture alone (42 ng/ml) was maximal at 6 h, and then began to decrease. When the two mixtures were applied to the cells concurrently at the previous concentrations, the response was maximal at 6 h, dropped off to a low at 12 h, and then increased again by 16–20 h, though not reaching the level exhibited at 6 h. Different proportions of the mixture combinations were then tested at 6 and 16 h (Fig. 5). The first pair of bars resulted from the addition of the PAH mixture alone, while the other pairs show the responses to increasing concentrations of the dioxin/furan mixture with the concentration of the PAH mixture consistent throughout. When the dioxin/furan mixture concentra-
Fig. 3. Responses to individual PAHs, PCBs, and the PCB mixture Aroclor 1260 at 6 and 16 h. Bars represent the mean of at least two analyses.
tectable (five-fold induction) by 2 h (Fig. 1). The response to TCDD steadily increased to reach a maximum fold induction of around 100 at 16 h. In contrast, the response to the PAH mixture increased to a maximum fold induction of around 50 at 6 h, then steadily decreased. Because 6 and 16 h exposure periods ap-
Fig. 4. Responses to a mixture of dioxins and furans, a PAH mixture, and the two mixtures in combination (at the same concentrations shown individually). Columns represent the mean of at least two analyses. Mixture compositions are available as supporting information.
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
123
16 h, typical of the response to chlorinated compounds alone. This procedure was repeated, maintaining the dioxin/ furan mixture at 0.09 ng/ml and applying increasing concentrations of the PAH mixture (Fig. 6). In this case, the fold induction responses were greater at 16 h until the PAH concentration was increased to 41.5 ng/ml. Typical of the response to PAHs alone, the 6 h response was greater than the 16 h response with this combination.
3.2. Analyses of en6ironmental samples
Fig. 5. Responses to the PAH mixture and dioxin/furan mixture in combination. The concentration of the PAH mixture is held constant, with increasing concentrations of the dioxin/furan mixture as indicated. Columns represent the mean of at least two analyses.
Fig. 6. Responses to the mixture combinations, with the dioxin/furan concentration held constant and increasing concentrations of the PAH mixture as indicated. Columns represent the mean of at least two analyses.
Analyses of the cleaned and uncleaned extracts of two soil samples are shown in Fig. 7. The 16 h responses to both c 1-NC and c 1-C are greater than the 6 h responses, which indicated that these extracts contained primarily chlorinated compounds. In contrast, the 16 h response to c2-NC was much less than the 6 h response, indicating that this extract contained primarily PAHs. Extract c2-C, which had gone through column cleanup, produced much lower responses at both times, with no difference between 6 and 16 h responses. After removal of the PAHs, this extract produced responses that indicated the presence of only small amounts of chlorinated compounds. As described in Section 2.5, the 16 h RGS responses were used to calculate RGS TEQ for Extracts c1-C and c2-C as 48.5 and 1.5 pg/g, respectively. From the EPA Method 8290 chemical analyses, the Chemical TEQ was calculated as the sum of each analyte concentration multiplied by its TEF (Table 3). The calculated Chem TEQ for Samples c 1 and c 2 were 61 and 0.8 pg/g, respectively.
4. Discussion
Fig. 7. Responses to extracts of two environmental soil/ash samples. ‘NC’ indicates that no cleanup was performed, ‘C’ extracts were cleaned using EPA Method 8280A to remove PAHs. Extracts were applied to the cells in equal volumes. Columns represent the mean of at least two analyses.
tion increased from 0.04 to 0.09 ng/ml, the ratio of 6/16 h responses was altered. At the 0.09 ng/ml concentration and higher, greater responses were observed at
The induction of the CYP1A1 gene is mediated by ligand binding at the cytosolic Ah-receptor, which is associated with other proteins, and further controlled by the translocation of the receptor–inducer complex to the nucleus and binding to dioxin-responsive elements (Whitlock, 1990; Pongratz et al., 1992). Both PAHs and chlorinated compounds, including coplanar PCBs, dioxins and furans, act via this pathway. In assessment of environmental mixtures, which may contain both PAHs and chlorinated compounds, it is beneficial to be able to differentiate between them. The human cell line, 101L, is used to study the mechanisms of Ah-receptor-mediated induction of CYP1A1 because of its sensitivity and facility (Postlind et al., 1993; Chen and Tukey, 1996). The P450 Reporter Gene System (RGS) applies the 101L cell assay to screen environmental samples for the presence of con-
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
124
taminants which induce CYP1A1 gene expression. In this study, RGS was used to investigate the temporal differences in luciferase production through CYP1A1 promoter activity by PAHs and chlorinated compounds, and how this may be used in assessment of environmental samples. It is apparent that the RGS fold induction response to PAHs is temporally different from that of the reference inducer TCDD (Fig. 1). The response to PAHs reaches maximal at an earlier time and then decreases, while the TCDD response reaches maximal at a later time and is more sustained. Individual PAHs exhibit the same response trends as the PAH mixture, and individual PCB congeners, as well as the Aroclor 1260 mixture, act in the same manner as TCDD (Fig. 3). Ellipticine, a reversible-type (competitive) inhibitor of human CYP1A1, inhibits the metabolic activity of CYP1A1 enzymes by binding strongly to the reduced heme iron and thereby preventing the fixation and activation of molecular dioxygen (Lesca et al., 1979). Inhibition occurs after activation of the CYP1A1 promoter region, so that it does not affect the production of luciferase in these cells, but rather inhibits the endogenous metabolic activity of CYP1A1 enzymes. The addition of ellipticine to the cell media increased the production of PAH-induced luciferase expression at 16 h, while the 6 h responses with and without ellipticine were equivalent (Fig. 2). This supports the assumption that the metabolism of PAHs in these cells is causing the luciferase production to decrease after 6 h. The response to TCDD was not increased at either time period.
A complex mixture of both PAHs and dioxins and furans exhibits additivity of response over time (Fig. 4), supporting the use of RGS as a measure of the total amount of CYP1A1-inducing contaminants present. The 6/16 h response patterns vary depending on the concentrations of PAHs compared to dioxins and furans. When the dioxin/furan mixture concentration was increased from 0.04 to 0.09 ng/ml, while the PAH mixture concentration was maintained at 16.6 ng/ml, the response ratio (6 h response divided by 16 h response) decreased from \ 1 to B 1 (Fig. 5). This occurred not only because the 16 h response was increased with increased dioxin/furan concentration, but also because the 6h response decreased. When higher concentrations of CYP1A1-inducing compounds were present, the metabolism of PAHs was apparently also increased, so that less was available to induce the production of luciferase at 6 h. Alternatively, the presence of increasing amounts of dioxins and furans, which have higher receptor and binding affinity, may have masked the 6 h response by PAHs. Fig. 6 shows the reverse effect, so that when the PAH mixture concentration was increased from 16.6 to 41.5 ng/ml while the dioxin/furan mixture concentration was held constant at 0.09 ng/ml, the 6/16 h response ratio increased from B 1 to \ 1. In addition to the increase in 6 h response with this combination, the 16 h response was decreased. The reason for the decrease is not known, but it is apparent that as the PAH concentration increases, the 6 h response will also increase and at some point will dominate the response pattern. Further studies using ellipticine to inhibit the metabolism
Table 3 Chemical analysis of environmental samples by EPA method 8290 Analyte
2,3,7,8-Tetrachlorodibenzo-p-dioxin 1,2,3,7,8-Pentachlorodibenzo-p-dioxin 1,2,3,4,7,8-Hexachlorodibenzo-p-dioxin 1,2,3,6,7,8-Hexrachlorodibenzo-p-dioxin 1,2,3,7,8,9-Hexachlorodibenzo-p-dioxin 1,2,3,4,6,7,8-Heptachlorodibenzo-p-dioxin Octachlorodibenzo-p-dioxin 2,3,7,8-Tetrachlorodibenzofuran 1,2,3,7,8-Pentachlorodibenzofuran 2,3,4,7,8-Pentachlorodibenzofuran 1,2,3,4,7,8-Hexachlorodibenzofuran 1,2,3,6,7,8-Hexachlorodibenzofuran 1,2,3,7,8,9-Hexachlorodibenzofuran 2,3,4,6,7,8-Hexachlorodibenzofuran 1,2,3,4,6,7,8-Hexachlorodibenzofuran 1,2,3,4,7,8,9-Hexachlorodibenzofuran Octachlorodibenzofuran Chem TEQ (pg/g) a
TEFa
1 0.1 0.1 0.1 0.1 0.01 0.001 0.1 0.05 0.5 0.1 0.1 0.1 0.1 0.01 0.01 0.001
Sample c1
Sample c2
(pg/g)
(pg/g)*TEF
(pg/g)
(pg/g)*TEF
0 14 26.5 60.7 74.7 595.7 2332.1 22.9 8.4 17.5 65.3 22.8 65.2 0 200.8 25.5 349.4
0 14.0 2.7 6.1 7.5 6.0 2.3 2.3 0.4 8.8 6.5 2.3 6.5 0.0 2.0 0.3 0.3
0 0 0 0 0 12.1 98.6 0 0 0 3 1.7 0 0 12.1 0 13.8
0 0 0 0 0 0.1 0.1 0 0 0 0.3 0.2 0 0 0.1 0 0.01
68
Toxic equivalency factors, humans and mammals (Van den Berg et al., 1998).
0.8
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
of PAHs may help to elucidate the mechanisms involved. The RGS response patterns to standard mixtures individually and in combination have important implications for the assessment of the type(s) of contaminants present in unknown environmental samples. Unpublished laboratory data have demonstrated that the RGS responses are 70 – 90% of what would be predicted from the combination of two to four inducing PAHs or coplanar PCBs. RGS analyses of sediment samples from coastal areas of the US, tested for the National Oceanic and Atmospheric Administration, have shown high correlations with subsequent chemical determinations for total high molecular weight PAHs (Anderson et al., 1999). If the RGS responses were not a function of the additive induction from multiple inducers, such strong correlations with chemical analyses would not be observed. The two soil samples analyzed in this study exhibit straightforward response patterns. Sample c 1 showed a response pattern typical of chlorinated compounds, with the 16 h response greater than the 6 h in both the original c 1-NC and cleaned extract c1-C. In contrast, PAHs were identified in Sample c 2 because the 6 h response to extract c2-NC was greater than at 16 h. When PAHs were removed, the response pattern of c 2-C changed dramatically to indicate the presence of only small amounts of inducers. The 16 h responses to extracts c1-C and c 2-C were used to calculate RGS TEQ values of 48.5 and 1.5 pg/g, respectively, corresponding well to the Chem TEQ derived by EPA Method 8290 (68 and 0.8 pg/g, respectively). When environmental samples contain more complex mixtures of both PAHs and chlorinated compounds, the use of RGS analysis at 6 and 16 h to differentiate between contaminant types is more difficult. Recent testing of sediment extracts from New York Harbor, which contained as high as 32 mg/g of PAHs and pg/g levels of dioxins and furans, consistently exhibited response patterns typical of PAHs. Nonetheless, the 16 h responses yielded RGS TEQ values that were highly correlated (R 2 =0.81) with the Chem TEQ values calculated from EPA Method 8290 confirmation of dioxins and furans. The use of the cleanup procedures described here in Section 2.2 for removal of PAHs is advantageous when RGS is employed specifically to screen for chlorinated compounds. However, the extra time and expense necessary for cleanup procedures is often not warranted in projects where a rapid and inexpensive screening test such as RGS is used. We recommend a tiered approach in such investigations, since the responses of the RGS assay will identify the presence of many of the organic chemicals of concern. After the first tier of RGS testing (testing all samples at 16 h), it is possible to eliminate some stations from further analysis and focus attention on
125
the most contaminated sites. The next tier would then utilize the 6/16 h time intervals to identify which classes of inducing compounds are likely to be present. Results of these tests will help to direct subsequent chemical analyses. The analytical results would then be compared to RGS findings, and a strong correlation would demonstrate that the levels of key contaminants in samples not chemically characterized could be estimated from the regression curve. It should be cautioned that RGS is highly specific only for those contaminants with affinity for the Ah receptor; other types of toxic chemicals would not be detected. RGS was used in a study of 75 sediment samples from two rivers in the Pacific Northwest for the US Army Corps of Engineers (US ACOE, 1999). Most samples produced greater RGS responses at 6 than at 16 h, indicating the presence of PAHs, which were then determined by GC/MS analysis. One sample produced a significantly greater RGS response at 16 h than at 6 h, and this sample was selected for further chemical analysis for the presence of chlorinated compounds. As a result, PCBs were identified in this sample. In conclusion, the analysis of environmental samples at both 6 and 16 h using RGS yields important information regarding the type of CYP1A1-inducing compounds present. When responses differ considerably at the two time periods, characterization between PAHs and chlorinated compounds is more straightforward. In contrast, 6 and 16 h responses that are more similar are indicative of the presence of both PAHs and chlorinated compounds, which is also important information in environmental assessment. Further, by examining the 6/16 h response patterns of all samples from a site, differences among samples are often discovered, so that PAHs and/or chlorinated inducers are identified where they were not suspected. Chemical confirmation of a representative number of these samples (: 15–20%) can then be advised. Acknowledgements We thank Joe Wiegel for sample extract preparation and clean-up. We are also grateful to Hiyoshi Corporation, Shiga, Japan, for providing soil samples. This work was supported by Columbia Analytical Services, Kelso, WA. References Anderson, J.W., Rossi, S.S., Tukey, R.H., et al., 1995. A biomarker, P450RGS, for assessing the potential toxicity of organic compounds in environmental samples. Environ. Toxicol. Chem. 14, 1159. Anderson, J.W., Jones, J.M., Hameedi, J., et al., 1999. Comparative analysis of sediment extracts from NOAA’s bioeffects studies by the biomarker, P450 RGS. Mar. Environ. Res. 48, 407.
126
J.M. Jones et al. / En6ironmental Toxicology and Pharmacology 8 (2000) 119–126
APHA, P450 reporter gene response to dioxin-like organics. Method 8070, Standard Methods for the Examination of Water and Wastewater, 19th ed., Supplement, American Public Health Association, Washington, DC, 1996, p. 24. ASTM, E 1853M-96 Standard Guide for measuring the presence of planar organic compounds which induce CYP1A, using reporter gene test systems, Biological Effects and Environmental Fate; Biotechnology; Pesticides, 1997 Annual Book of ASTM Standards, vol. 11.05-Water and Environmental Technology, American Society for Testing and Materials, West Conshohocken, PA, 1999, p. 1400. Bronstein, I., 1994. Chemiluminescent and bioluminescent reporter gene assays. Anal. Biochem. 219, 169. Celander, M., Bremer, J., Hahn, M.E., et al., 1997. Glucocorticoid– xenobiotic interactions: dexamethasone-mediated potentiation of cytochrome P4501A induction by (b-naphthoflavone in a fish hepatoma cell line (PLHC-1). Environ. Toxicol. Chem. 16, 900. Chen, Y., Tukey, R.H., 1996. Protein kinase C modulates regulation of the CYP1A1 gene by the aryl hydrocarbon receptor. J. Biol. Chem. 271, 26261. Hahn, M.E., Stegeman, J.J., 1994. Regulation of cytochrome P4501A1 in teleosts: sustained induction of CYP1A1 mRNA, protein, and catalytic activity by 2,3,7,8-tetrachlorodibenzofuran in the marine fish Stenotomus chrysops. Toxicol. Appl. Pharmacol. 127, 187. Jones, J.M., Anderson, J.W., 1998. Cost-effective site characterization using P450 reporter gene system (RGS). In: Wickramanayake, G.B., Hinchee, R.E. (Eds.), Risk, Resource, and Regulatory Issues. Battelle, Columbus, OH, p. 1. Jones, J.M., Anderson, J.W., 1999. Relative potencies of PAHs and PCBs based on the response of human cells. Environ. Toxicol. Pharmacol. 7, 19. Kim, G.B., Anderson, J.W., Bothner, K., et al., 1997. Application of P450RGS (reporter gene system) as a bioindicator of sediment PAH contamination in the vicinity of Incheon Harbor, Korea. Biomarkers 2, 181. Kloepper-Sams, P., Stegeman, J.J., 1989. The temporal relationship between P450E protein content, catalytic activity, and mRNA levels in the teleost Fundulus heteroclitus following treatment with b-naphthoflavone. Arch. Biochem. Biophys. 268, 525.
.
Lesca, P., Rafidinarivo, E., Lecointe, P., et al., 1979. A class of strong inhibitors of microsomal monooxygenases: the ellipticines. Chem. Biol. Interactions 24, 189. Pesonen, M., Goksoyr, A., Andersson, T., 1992. Expressions of P4501A1 in a primary culture of rainbow trout hepatocytes exposed to (b-naphthoflavone or 2,3,7,8-tetrachlorodibenzo-pdioxin. Arch. Biochem. Biophys. 292, 228. Pongratz, I., Mason, G., Poellinger, L., 1992. Dual roles of the 90 kDa heat shock protein hsp90 in modulating functional activities of the dioxin receptor. J. Biol. Chem. 267, 13728. Postlind, H., Vu, T., Tukey, R.H., et al., 1993. Response of human CYP1-luciferase plasmids to 2,3,7,8-tetrachlorodibenzo-p-dioxin and polycyclic aromatic hydrocarbons. Toxicol. Appl. Pharmacol. 118, 255. Stegeman, J.J., Brouwer, M., DiGiulio, R.T., et al., 1992. Molecular responses to environmental contamination: enzyme and protein systems as indicators of chemical exposure and effect. In: Huggett, R.J., Kimerle, R.A., Mehrle, P.M., Bergman, H.L. (Eds.), Biomarkers: Biochemical, Physiological, and Histological Markers of Anthropogenic Stress. Lewis, Boca Rotan, FL, p. 235. Stegeman, J.J., Hahn, M.E., Weisbrod, R., et al., 1995. Induction of cytochrome P4501A1 by aryl hydrocarbon receptor agonists in porcine aorta endothelial cells in culture and cytochrome P4501A1 activity in intact cells. Mol. Pharmacol. 47, 296. Thomson, J.F., Hayes, L.S., Lloyd, D.B., 1991. Modulation of firefly luciferase stability and impact of studies of gene regulation. Gene 103, 171. US Army Corps of Engineers, Integrated Feasibility Report for Channel Improvements and Environmental Impact Statement; Columbia and Lower Willamette River Federal Navigation Channel, Final Report of the US ACOE, Portland District, August, 1999. Van den Berg, M., et al., 1998. Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environmental Health Perspectives 106, 775. Whitlock, J.P. Jr, 1990. Genetic and molecular aspects of 2,3,7,8-tetrachlorodibenzo-p-dioxin action. Annu. Rev. Pharmacol. Toxicol. 30, 251.