Comparison of TCDD and PCB CYP1A Induction Sensitivities in Fresh Hepatocytes from Human Donors, Sprague-Dawley Rats, and Rhesus Monkeys and
http://www.100md.com
《毒物学科学杂志》
General Electric Company, Global Research Center, One Research Circle, Niskayuna, New York 12309
In Vitro Technologies, Inc., Baltimore, Maryland 21227
General Electric Company, Fairfield, Connecticut 06431
ABSTRACT
2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and related chemicals induce cytochrome P450 1A (CYP1A) gene expression and, at sufficient exposures, cause toxicity. Human health risks from such exposures are typically estimated from animal studies. We tested whether animal models predict human sensitivity by characterizing CYP1A gene expression in cultures of fresh hepatocytes from human donors, rats, and rhesus monkeys and HepG2 human hepatoma cells. We exposed the cells to three aryl hydrocarbon receptor (AhR) ligands of current environmental interest and measured 7-ethoxyresorufin-O-deethylase (EROD) activity and concentrations of CYP1A1 and CYP1A2 mRNA. We found that human cells are about 10–1000 times less sensitive to TCDD, 3,3',4,4',5-pentachlorobiphenyl (PCB 126), and Aroclor 1254 than rat and monkey cells, that relative potencies among these chemicals are different across species, and that gene expression thresholds exist for these chemicals. Newly calculated rat–human interspecies relative potency factors for PCB 126 were more than 100 times lower than the current rodent-derived value. We propose that human-derived values be used to improve the accuracy of estimates of human health risks.
Key Words: polychlorinated biphenyls; PCB; TCDD; dioxin; dioxin equivalents; TEQ; relative potency factors; human hepatocytes; HepG2 cells; cytochrome P450 1A; CYP1A1; CYP1A2; risk assessment.
INTRODUCTION
Polychlorinated dibenzo-p-dioxins, dibenzofurans, and PCBs comprise a group of highly regulated environmental contaminants of global concern. A subset of these compounds can bind to, and activate, a natural cellular receptor, the aryl hydrocarbon receptor (AhR), which is a transcription factor (Carlson and Perdew, 2002; Nebert et al., 1993). This can result in the up-regulation of many genes (Boutros et al., 2004; Okey et al., 1994), including those encoding metabolic enzymes such as the cytochrome P450 (CYP) isozymes (Rowlands and Gustafsson, 1997). The induction of CYP1A1 correlates well with exposure in animals (Tritscher et al., 1992; Vanden Heuvel et al., 1994) and is used as an important basis for estimating human health risk (Birnbaum and DeVito, 1995). The World Health Organization (WHO) has established toxic equivalency factors (TEFs) for each of 29 AhR ligands, including 7 dioxins, 10 polychlorinated dibenzofurans, and 12 PCBs. Because 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is the highest affinity ligand for the AhR, and by far the most potent at eliciting biologic responses, it was assigned a TEF of 1. TEFs for other chemicals are based on relative potency (REP) values relative to that of TCDD using both in vivo and in vitro animal studies, and it is assumed that these TEFs (WHO98 TEFs) are appropriate for use in human risk assessment (van den Berg et al., 1998). We tested this assumption by comparing the responses of fresh hepatocytes from rats, rhesus monkeys, and human donors, and also a human-derived tumor cell line, HepG2, to TCDD, PCB 126 (3,3',4,4',5-pentachlorobiphenyl), and Aroclor 1254, a commercial mixture of PCBs. PCB 126 has a TEF of 0.1. Aroclor 1254 has a calculated WHO98 toxic equivalency (TEQ) of 0.000047 (Mayes et al., 1998), largely contributed by PCB 126.
Studies that compared CYP1A1 induction in mammalian cells or cell lines suggest that humans are less sensitive than rats to TCDD. Some studies measured TCDD-induced 7-ethoxyresorufin-O-deethylase (EROD) activity in the human HepG2 cell line (Lipp et al., 1992), while other studies measured EROD activity in primary human hepatocytes (Schrenk et al., 1995). When the EC50s calculated in these studies were compared to those of an earlier study with rat primary hepatocytes and H4IIE cells (Schrenk et al., 1991), the human cells were found to be 8–19 times less sensitive than the rat cells to TCDD. Wiebel et al. (1996) compared TCDD-induced CYP1A activity in human HepG2 cells and rat H4IIE cells and found that the human cells were 20 times less sensitive. Xu et al. (2000) demonstrated that primary human hepatocytes were less sensitive to TCDD than primary rat hepatocytes and also reported species differences in the expression of CYP1A1 and CYP1A2 mRNA.
Few studies have measured CYP1A induction by TCDD and PCB 126 in more than one species using the same methodology. Vamvakas et al. (1996) tested TCDD and several PCB congeners in the HepG2 and MCF-7 human cell lines and in the rat H4IIE cell line. Zeiger et al. (2001) compared the induction by TCDD and several PCB congeners in primary rat hepatocytes and the rat H4IIE and human HepG2 cell lines. Recently, Peters et al. (2004) reported the induction of EROD activity by TCDD and PCB 126 in human MCF-7 and HepG2 cells and in rat H4IIE cells. Each of these studies showed that the human cells were less sensitive to TCDD and PCB 126 than rat-derived cells.
In addition to understanding the relative sensitivity between species to these chemicals, it is also important to determine if the REPs among these chemicals are consistent across species, especially if REPs derived from data for one species are used to extrapolate effects to other species. In fact, studies with mice (DeVito et al., 2000) and human-derived cells (Zeiger et al., 2001) have suggested that the REPs for several PCB congeners were inconsistent with their current WHO98 TEFs. More studies are needed to characterize how human responses to AhR ligands differ from the responses of other species while also characterizing the consistency of the relative potencies of AhR ligands across species.
Thus, to evaluate species differences in the CYP1A responses to TCDD, PCB 126, and Aroclor 1254, and to determine the REPs for PCB 126 and Aroclor 1254 in each species, we compared the responses of all cells under identical conditions. This included the use of a single lot number of each test chemical throughout the study and the preparation of only one or two stock solutions. Because immortal cell lines may not accurately reflect responses of normal human hepatocytes, we tested both fresh human hepatocytes and the HepG2 cell line. Use of donor cells also permitted an evaluation of differences among five humans. Fresh rat hepatocytes and HepG2 cells were tested to compare our methodology and results with earlier work (Lipp et al., 1992; Schrenk et al., 1995; Zeiger et al., 2001). Hepatocytes from rhesus monkeys (Macaca mulatta) also were included, since both reproductive and immunological effects have been reported for this primate following exposure to PCBs (Arnold et al., 1995). All four of these cell types are known to express an AhR (Roberts et al., 1985, 1989, 1990). To sample the diversity of humans, we tested hepatocytes from five organ donors, two Caucasians of each gender and one African-American male.
Cells were treated over a wide concentration range of each chemical in serum-free culture medium (Hestermann et al., 2000). CYP1A induction was determined after 48 h of exposure by measuring EROD activity and CYP1A mRNA. EROD activity was not obtained for one of the five donors due to experimental error. CYP1A1 and CYP1A2 mRNAs were measured in two to seven experiments for each species. Species differences were evaluated by comparing thresholds, EC50s, and maximal responses for each chemical. The potencies of PCB 126 and Aroclor 1254, relative to TCDD, were calculated for each measurement and compared across species to derive interspecies relative potency factors.
MATERIALS AND METHODS
Chemicals.
TCDD (molecular weight = 322) was obtained from Accustandard (New Haven, CT; catalog no. D404N; CAS no. 1746–01–6; Lot no. 970401R-AC; 99.1% pure). The single contaminant was a pentachloro-hydroxydiphenyl ether by GC/MS.
PCB 126 (molecular weight = 326.4) was obtained from Accustandard (Catalog no. C-126N; CAS no. 57465–28–8; Lot no. 081699MT-AC; 99.2% pure). The single contaminant was identified as a tetrachlorobiphenyl by GC/MS.
Aroclor 1254, lot no. 122–078 (molecular weight = 326.2) was from the same lot of material used in an earlier chronic bioassay conducted for General Electric Company (Mayes et al., 1998). The calculated WHO98 TEQ (i.e., the toxic equivalency to TCDD found by summing the products of the toxic equivalency factor of each of the 12 PCB congeners with an assigned TEF multiplied by its respective measured concentration) is 47 ppm.
Hepatocyte sources.
Human hepatocytes were prepared from nontransplantable human tissue acquired after informed consent for use in research by In Vitro Technologies, Inc. (IVT). An external FDA-certified Institutional Review Board approved the use of nontransplantable human tissue for ADME-Tox research at IVT. Donor 1, (IID), IVT Lot MHU-L-012303, was a 66-year-old Caucasian male who died from an intracranial hemorrhage. Donor 2, (KZO), IVT Lot FHU-L-020203, was a 42-year-old Caucasian female who died from a cerebrovascular accident. Donor 3, (WRG), IVT Lot MHU-L-052004, was a 41-year-old Caucasian male who died from an astrocytoma. Donor 4, (RFA) IVT Lot FHU-L-072004, was a 56-year-old Caucasian female who died from a cerebrovascular accident. EROD data were not collected for this donor because of experimental error but CYP1A mRNA data were obtained. Donor 5, (ZYZ) IVT Lot MHU-L-0730044, was a 46-year-old African-American male who died from anoxia. Serologies for all donors were negative for HIV, HBV, and HCV, but positive for cytomegalovirus. Urinalyses and blood chemistries for all donors were within normal limits. See www.invitrotech.com/characterizationtab.cfm for additional donor information.
Rhesus monkey hepatocytes were isolated from liver tissue from a single chemically naive young adult female for each of three trials. The tissue was purchased by IVT from approved sources that comply with all appropriate laws and guidelines.
Rat hepatocytes were isolated by IVT, from two rats (Female Crl:CD (SD)IGS BR, Charles River Laboratories, Wilmington, MA) and pooled for each of five experiments. Rats were treated in accordance with the Animal Welfare Act.
The HepG2 human hepatoma cell line was obtained from the American Type Culture Collection, Manassas, VA.
Hepatocyte cultures.
Hepatocytes were isolated according to the two-step collagenase perfusion procedure of Li et al. (1992). Isolated hepatocytes were counted using Trypan blue exclusion to determine yield and viability. Only hepatocyte preparations with 70% viability were used. Freshly isolated hepatocytes from rat, monkey, or human donors were plated, in triplicate, onto collagen-coated 24-well plates at a cell density of 3.5 x 105 cells per well in Plating Medium (Dulbecco's modified Eagle's medium (DMEM) supplemented with bovine serum albumin, fructose, HEPES, sodium bicarbonate, L-glutamine (2.4 mM), hydrocortisone (2.38 μM), insulin (135 nM), MEM nonessential amino acids (1.2%), amikacin, penicillin (200,000 U/l), streptomycin (200 mg/l), gentamycin, and Fungizone). Fungizone was excluded from HepG2 cell Plating Medium. Cultures were placed in a 37°C/5% CO2 incubator for 2 days before use to establish the primary hepatocyte and HepG2 cell monolayers. Confluency was visually checked each day of the culture period and was generally 90–100%, except for Donor 1, for which it was 75–90%.
Chemical treatments.
Each test chemical was prepared in dimethyl sulfoxide (DMSO) at 200 times (200x) the final concentration and diluted with incubation medium to the required concentrations. TCDD was soluble in DMSO at 60 μM, but not 200 μM. PCB 126 was soluble in DMSO at 2000 μM, but not 6000 μM. Aroclor 1254 was soluble in DMSO up to at least 60 mM. Two 200x stock solutions were prepared for TCDD and PCB 126. Eleven concentrations were used for each trial, in addition to the vehicle control. TCDD was used at 0.00001, 0.0001, 0.001, 0.003, 0.01, 0.03, 0.1, 0.3, 1, 10, and 100 nM. PCB 126 was used at 0.001, 0.01, 0.03, 0.1, 0.3, 1, 3, 10, 100, 1000, and 10,000 nM. Aroclor 1254 was used at 1, 3, 10, 30, 100, 300, 1000, 3000, 10,000, 30,000, and 300,000 nM.
Established primary hepatocyte and HepG2 cell cultures were treated with 500 μl of serum-free Incubation Medium (Plating Medium without serum) containing TCDD, PCB 126, or Aroclor 1254. Serum-free medium was used, since serum can significantly reduce the cellular uptake of these compounds (Hestermann et al., 2000). The final concentration of DMSO in the incubations was 0.5%, consistent with previous studies with these chemicals and cells (Hestermann et al., 2000; Lipp et al., 1992; Zeiger et al., 2001). There were no visible indications that any chemicals (or any subsets of congeners in the Aroclor 1254) had precipitated at any of the incubation concentrations, but this was not analytically confirmed. Cultures were incubated at 37°C/5% CO2 with test chemicals for a total of 48 h. The incubation medium containing the test chemical was replaced at 24 h.
Culture viability was assessed in a replicate set of cultures for each cell type and chemical treatment by measuring the metabolic conversion of 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) to the formazan product in triplicate wells (Mosmann, 1983). TCDD and PCB 126 did not affect culture viability at any concentration tested. Aroclor 1254 reduced MTT conversion in rhesus cells to approximately 50% of control levels at concentrations 10–5 M. At 3 x 10–4 M, Aroclor 1254 reduced MTT conversion to <2%, 41%, 2%, and 26% in rhesus, rat, HepG2, and 1 donor, respectively. Three donors were not affected at this concentration (data not shown).
CYP1A enzyme activity assay.
EROD activity was determined using modifications of the methods of Burke et al. (1985) and Donato et al. (1993). Medium containing the test chemical was removed from the culture plates and replaced with 300 μl of Krebs-Henseleit buffer (supplemented with amikacin, calcium chloride, gentamicin, heptanoic acid, N-(2-hydroxyethyl)piperazine-N'-(2-ethanesulfonate), and sodium bicarbonate) containing 10 μM ethoxyresorufin and 3 mM salicylamide. Cultures were incubated at 37°C/5% CO2 for 30 min. Incubations were terminated by removing the supernatant and adding it to 300 μl of methanol containing 2% DMSO. The amount of resorufin formed in the incubations was quantitated against a standard curve of resorufin as measured by a fluorometric assay (excitation: 530 nm; emission: 590 nm; Wallac Victor2 plate reader, Perkin Elmer). The quantitation limit was 0.1 pmol resorufin/(min x mg protein).
Protein concentrations of the cultures were determined using a BCA protein assay kit (Pierce Chemical Co., Rockford, ILL). Cells were lysed by adding 200 μl of 0.1 N NaOH and incubating at 37°C/5% CO2 for 30 min. Cell lysates were scraped and harvested. Ten microliters of each cell lysate was mixed with 200 μl of BCA working reagent (reagent A:reagent B (49:1)) and incubated for 30 min at 37°C. Protein concentrations were quantified against a standard curved using absorbance at 572 nm (Wallac Victor2 plate reader) and were generally about 0.1 mg protein/culture well.
RNA isolation.
Medium containing the test chemical was removed from each well, and cultures were washed with 500 μl of phosphate buffered saline. Phosphate buffered saline was then replaced with 200 μl of TRIzol reagent (Invitrogen, Carlsbad, CA). Hepatocytes were scraped from each well with a wide-bore pipette tip, and homogenized by gently pipetting up and down. Cell homogenates were transferred to microcentrifuge tubes, and 200 μl of chloroform was added to each tube. Tubes were mixed for 15 s. Homogenates were placed on ice for 5 min and then centrifuged at 12,000 x g for 15 min at 4°C. The aqueous layer was carefully removed and transferred to a fresh microcentrifuge tube, and 200 μl of phenol:chloroform:isoamyl alcohol (25:24:1) was added. Tubes were mixed for 15 s. Mixtures were placed on ice for 5 min and then centrifuged at 12,000 x g for 15 min at 4°C. The aqueous layer was removed to a fresh microcentrifuge tube. RNA from the aqueous layer was precipitated by adding 200 μl of isopropanol and placing the tubes at –20°C overnight. Following overnight precipitation, the tubes were stored at –70°C until analysis.
RNA quantitation.
The RNA samples were centrifuged at 12,000 x g for 15 min. Supernatants were decanted, and the RNA pellets were washed with 70% ethanol in Tris EDTA buffer. Samples were centrifuged at 12,000 x g for 30 min. Supernatants were decanted, and the pellets were allowed to air dry. RNA pellets were reconstituted with 50 μl of TE buffer. RNA was quantitated using a Ribogreen RNA quantitation kit (Molecular Probes, Eugene, OR). In brief, 5 μl of each RNA sample was diluted 50-fold with TE buffer. A 100-μl aliquot of the diluted RNA was mixed with 100 μl of the diluted RNA quantitation reagent and incubated for 5 min at room temperature. RNA in the samples was quantitated against a standard curve using fluorometric assay (excitation: 485 nm; emission: 535 nm; Wallac Victor2 plate reader).
Quantitation of specific mRNA—RNA Invader invasive cleavage assay.
The RNA Invader invasive cleavage assay is an invasive cleavage amplification assay developed by Third Wave Technologies, Inc., Madison, WI (Burczynski et al., 2001; Hall et al., 2000). In this method, two oligonucleotides, an upstream oligonucleotide and the probe, anneal to the target sequence. The probe contains both a target-specific and a noncomplimentary region. When the probe overlaps the upstream oligonucleotide a reporter sequence can be cleaved. The rapid turnover and production of the cleaved sequence permits a secondary reaction and the linear amplification of a fluorescent signal. This method has the ability to discriminate between two highly homologous sequences such as found with the P450s (Eis et al., 2001). Total RNA isolated from each triplicate culture for each concentration and for each species was hybridized against human/rat CYP1A1 (h/rCYP1A1) or h/rCYP1A2 probes. CYP1A1 probes targeted human GeneBank Accession no. K03191 sequence (5'-CCTGATTGAGCACTGTCAGGA-3') and rat GeneBank Accession no. NM_012540 sequence (5'-CCTCATTGAGCATTGTCTGGA-3'). CYP1A2 probes targeted human GeneBank Accession no. M55053 sequence (5'-AGGAGCACTATCAGGACTTTGACAAG-3') and rat GeneBank Accession no. K02422 sequence (5'-AGGAACACTATCAAGACTTCAACAAG-3'). The human CYP1A1 probe was also used with rhesus hepatocytes because these two species share a consensus sequence at the targeted region. The human CYP1A2 probe could not be used for the rhesus hepatocytes because the targeted sequence is not shared, as we confirmed. CYP1A1 and CYP1A2 mRNAs are expressed as amol (10–18 moles) specific mRNA/ng total RNA. Human and rat GAPDH mRNAs were also measured, as reference mRNAs, and found to be consistently expressed for each chemical, all treatment concentrations, and all species (data not shown).
Each RNA sample was diluted to a maximum concentration of 10 ng/μl. If the sample concentration was below 10 ng/μl, it was used without further dilution. The amount of specific mRNA in each RNA sample was quantified using Invader assay kits (1A1, 1A2, hGAPDH, positive control sequences, and generic reagent kits from Third Wave Technologies, Inc., Madison, WI). Specific RNA in the samples was quantified against standard curves generated using the Invader oligo sequence test probes against positive control transcripts that included the targeted sequence in rats and humans. Fluorescence was measured as described above.
Dose-response modeling.
Freshly isolated hepatocytes from the three species or the human-derived HepG2 hepatoma cell line were treated in vitro for 48 h. The dose-responses for EROD activity and CYP1A mRNA were modeled by combining unsummarized triplicate culture data across all experiments for each cell type at each concentration tested using the variable slope sigmoid Hill equation (GraphPad, 2005). We defined threshold as the concentration at which the response first exceeds the model's estimated constitutive or background expression level. We estimated the threshold for each curve by determining the concentration at which a line tangent to the modeled dose-response curve at the EC05 intersects the bottom of the modeled dose-response curve, as determined by the Hill equation (Fig. 2) (GraphPad, 2005). This is an objective method to quantify the dosage at which the curve is no longer attached to its lower asymptote (i.e., the constitutive expression level). The EROD EC50 is the concentration at which the induced enzyme activity is halfway between the calculated bottom and top of each dose-response curve (GraphPad, 2005).
RESULTS
EROD Activity
EROD dose-response curves are shown in Figure 1. Species differences for each chemical are clearly evident (Fig. 1A). We conducted additional analyses of the modeled curves by examining three features of each curve: the threshold, the EC50, and the maximal response. The threshold and EC50 provide complementary information on cell sensitivity, whereas the maximal response provides information on efficacy.
For TCDD, we found the lowest thresholds in fresh rat and rhesus liver cells. Thresholds in both fresh human liver cells (p 0.05) and HepG2 cells (p 0.05) (Fig. 2) were about 10 times higher than in rat and rhesus liver cells. Thus, both human cell types were about 0.1 as sensitive to TCDD as were rat and rhesus cells. The EC50s show that the human cells were about 0.10 to 0.27 as sensitive to TCDD as either rat or rhesus cells (Table 1). These observations are contrary to the assumption normally used in risk assessment that humans are more sensitive than experimental animals. They also indicate that the current application of factor multipliers to compensate for uncertainties regarding species sensitivities may result in overestimating human risk by several orders of magnitude.
The maximal heights of the TCDD response curves, which measure the ability of an AhR ligand to elicit a full response, were similar for all three species (i.e., within a factor of three). This suggests that the sensitivity for EROD induction, measured by either threshold or EC50, differentiates well among species while the maximal enzyme activity level does not. This also suggests that measures of sensitivity, because they are very different among species, are more important than maximal activity levels in estimating risk across species. Because the concentrations at which thresholds, EC50s, and the maximal responses occur are similar for HepG2 and donor cells, we suggest that either type of human cells can be used to study interspecies sensitivities to chemicals, but fresh human cells have the added advantage of providing information on the extent of individual variability.
Even more pronounced differences between human cells and both rat and rhesus cells were seen when the PCB 126 EROD dose responses were compared. Rhesus cells responded in a manner more similar to rats than human cells (Fig. 1A). The thresholds and EC50s indicated that both donor and HepG2 cells were between 0.01 and 0.001 as sensitive to PCB 126 as were rat and rhesus cells (Fig. 2, Table 1). Similar species differences were observed with Aroclor 1254. Donor and HepG2 cells were 0.01 as sensitive to Aroclor 1254 as were rhesus and rat cells, based on either thresholds (Figs. 1A and 2) or estimated EC50 values (Table 1), although precise EC50s could not be calculated.
By comparing the heights of the PCB 126 response curves with those of TCDD, we see that PCB 126 is about 0.9 and 0.8 as efficacious as TCDD in rhesus and rat cells, respectively, but only 0.5 (p 0.05) and 0.6 (p 0.05) as efficacious as TCDD in HepG2 and donor cells, respectively. This suggests that there are species-dependent factors other than those that determine sensitivity that influence differential CYP1A1 gene expression.
In comparing the EROD response of each species to their respective TCDD response, we found that PCB 126 is about 0.1 as potent as TCDD in rat and rhesus cells. This value is consistent with the WHO98 TEF for PCB 126 of 0.1. However, PCB 126 is between 0.01 and 0.001 as potent as TCDD in donor and HepG2 cells, based on thresholds (Figs. 1A and 2) and EC50 values (Table 1).
Our data show that human cells are about 0.1 as sensitive to TCDD as rats and rhesus cells. Our data also show that human cells are 0.001 or less as sensitive to PCB 126 as rat and rhesus cells are to TCDD and less than 0.000001 as sensitive to Aroclor 1254 as rats are to TCDD (Table 1). The human EROD responses to TCDD, PCB 126, and Aroclor 1254 are compared to only the responses of rats to TCDD in Figure 3 to facilitate the direct comparison of their EROD dose-response curves and EC50s in Tables 1 and 2.
Human Diversity in EROD Response
The diversity of human responsiveness, characterized by both sensitivity and maximal response, is an important concern for risk managers responsible for protecting sensitive populations. An earlier report suggests that humans have similar sensitivities but variable maximal responses to TCDD (Schrenk et al., 1995). In our study, we found similar thresholds for Donors 1, 2, 3, and 5 that were within an 11-fold concentration range (Fig. 1A) (calculation not shown). The EC50s of Donors 1, 2, 3, and 5 were within a seven-fold range, i.e., 1.1 x 10–10 M (8.9 x 10–11 M to 1.3 x 10–10 M, r2 = 0.95); 7.3 x 10–10 M (5.7 x 10–10 M to 9.3 x 10–10 M, r2 = 0.98); 1.2 x 10–8 (1.9 x 10–9 to 7.1 x 10–8, r2 = 0.95 [estimated because the maximal response could not be established with certainty]); and 1.1 x 10–10 M (8.5 x 10–11 M to 1.5 x 10–10 M, r2 = 0.97), respectively. The maximal responses of Donors 1, 2, and 5 varied by less than 2-fold for each chemical (Fig. 1B). Donor 3 was a poor responder to both TCDD and PCB 126 and had no measurable response to Aroclor 1254. This donor was excluded from the aggregate model shown in Fig. 1A (thus avoiding a shift in the curve further to the right) but is shown separately in Fig. 1B. Compared to rats, all donors responded poorly to Aroclor 1254. Our findings that sensitivity differences, measured by either threshold or EC50s, span over three orders of magnitude between human and rat cells, but only vary by a factor of about ten among the human samples, suggest that species differences are a more significant source than individual differences for the uncertainty in risk estimation.
CYP1A mRNA Induction
Measurement of CYP1A1 mRNA serves as an indicator of Ah gene expression before any posttranslational modifications or interferences occur that could affect protein levels and/or function (Staskal et al., 2005). Furthermore, mRNA levels provide a way to measure the competency of ligands with different AhR binding affinities to activate Ah gene expression (Chen et al., 2004). Thus, mRNA levels could provide additional information on species differences not provided by EROD activity alone. Since both CYP1A1 and CYP1A2 express EROD activity (Nerurkar et al., 1993), we measured CYP1A1 and CYP1A2 mRNA in a replicate set of cultures treated as described above using the RNA Invader invasive cleavage assay (Eis et al., 2001).
The induction profile of CYP1A1 mRNA for each chemical (Fig. 4A) was generally consistent with that of EROD activity for each species (Fig. 1). Interestingly, the induction of CYP1A1 mRNA by TCDD in rats, the most sensitive responder, was observed at concentrations about 10 times lower than seen for EROD activity. This has been observed by others who suggested that it was due to differences in the detection sensitivities of mRNA and enzyme activity (Vanden Heuvel et al., 1994). Because this was not observed in human and rhesus monkey cells for TCDD, or for PCB 126 or Aroclor 1254 in any species, our data suggest that there are aspects of TCDD induction and of the regulation of CYP1A1 that are unique to the rat that require further investigation. Consistent with the species differences observed for EROD activity, human cells required 100 to 1000 times higher concentrations of each chemical than rat and rhesus cells to induce CYP1A1 mRNA (Fig. 4A). This suggests that such differences are due to early events in AhR binding and/or DNA activation occurring up through mRNA transcription. Our results are also consistent with a recent investigation that showed a diminished CYP1A mRNA response when the AhR gene in mice was replaced with human AhR cDNA (Moriguchi et al., 2003), which the authors suggested might be due to structural differences in the AhR itself. Regardless of the explanation, the absence of such early induction events in human cells at exposure levels that elicit responses in other species suggests that animal models are overly sensitive when predicting human responses associated with Ah gene activation.
Individual variations in the AhR structure and function have also been described for several human cells (Cook et al., 1987; Micka et al., 1997). However, in our study, the CYP1A1 mRNA responses of four of the five donors were similar (Fig. 4B). These donors responded to TCDD at approximately 10–10 M and to PCB 126 between 10–8 and 10–7 M, and the respective responses reached amplitudes that were generally within a factor of three of one another. In contrast, the CYP1A1 mRNA responses of Donor 5 to all three chemicals were very low—but the CYP1A2 mRNA responses were robust (Fig. 5). Consistent with the Aroclor 1254 EROD results, only Donor 1 had a robust Aroclor 1254 CYP1A1 mRNA response, and the responses of the other four donors were barely detectable.
The regulation of both CYP1A1 and CYP1A2 mRNA induction seem to be similar in donor cells but dissimilar in rat cells (Figs. 4 and 5, Table 1). Donor cells responded to TCDD with nearly equivalent amounts of both CYP1A1 and CYP1A2 mRNA over the same concentration range. In rats, however, CYP1A2 mRNA induction was about 0.1 as sensitive as CYP1A1 mRNA induction, and the maximal CYP1A2 mRNA levels were only about 25% of CYP1A1 levels. This predominance of CYP1A2 over CYP1A1 in cultures of rat cells is consistent with earlier work (Xu et al., 2000). Interestingly, both proteins have been shown to be equally induced by TCDD in a long-term rat in vivo study (Walker et al., 1999). The CYP1A2 mRNA EC50s shown in Table 1 indicate that rat cells were about six and three times more sensitive than donor cells to CYP1A2 induction by both TCDD and PCB 126, respectively. These findings are consistent with reports that CYP1A1 and CYP1A2 may be differently regulated in rats (Drahushuk et al., 1999; Santostefano et al., 1997), but also indicate that there are important differences in their respective regulation between humans and rats.
CYP1A2 mRNA was barely detectable in HepG2 cells, which is consistent with earlier reports (Chung et al., 1994; Li et al., 1998; Vakharia et al., 2001). Our finding that CYP1A1 and CYP1A2 mRNAs are strongly and similarly induced in fresh human hepatocytes demonstrates the value of using primary human cells, rather than cell lines, in studying the expression of these two AhR-associated genes.
We calculated new EROD- and CYP1A1 mRNA-based REPs for PCB 126 using the data from our study (Table 2). REPs are typically calculated by dividing the EC50 for TCDD, the reference ligand, by the EC50 for the AhR ligand of interest with both values from the same species. For PCB 126, the EROD-based REPs in rhesus monkeys and rats were 0.13 and 0.12, respectively, consistent with the WHO98 TEF of 0.1. However, the EROD-based REP for PCB 126 for human donor cells was 0.003 and only 0.002 for HepG2 cells. These values are consistent with earlier work with HepG2 cells (Zeiger et al., 2001) and much lower than the WHO98 TEF of 0.1. This clearly demonstrates the inadequacy of the current TEQ approach to account for the possibility that each species may have its own unique set of REPs for different AhR ligands.
To account for the accumulating evidence not only that humans and rats may have different sensitivities to TCDD (Lipp et al., 1992; Schrenk et al., 1995; Vamvakas et al., 1996; Wiebel et al., 1996; Xu et al., 2000; Zeiger et al., 2001), but also that humans may have a unique set of REPs, we then calculated human REPs relative to the rat TCDD response (r-h REP) (Table 2, Fig. 6). This approach uses data from both species and is more consistent with the empirical evidence that humans and rats have different sensitivities to TCDD and other AhR ligands than using only rodent-derived data. The results show, based on EROD induction, that an appropriate r-h REP for PCB 126 is 0.0009, compared to the WHO98 TEF of 0.1. Even lower REP values were observed for CYP1A1 mRNA (Table 2).
Using the same approach for Aroclor 1254, we calculated an EROD-based r-h REP for human donor liver cells that is at least 54 times lower than predicted using the WHO98 TEFs (Table 2, Fig. 6). Comparable differences were also seen for Aroclor 1254 CYP1A1 mRNA data, with human donor cells 600 times less sensitive than predicted by the WHO98 TEF. Both assays also suggest that rhesus monkeys are more sensitive than rats to Aroclor 1254 and indicate that neither rhesus monkeys, although primates, nor rats are good models for humans for AhR ligand-based risk assessments for this complex mixture.
DISCUSSION
Our findings show orders of magnitude species differences in sensitivity to TCDD and PCBs and highlight the substantial uncertainties that arise when using experimental animal data to extrapolate to potential human health risks. These findings may help to explain the lack of conclusive evidence that PCBs have affected human health (Kimbrough et al., 2003). To the extent that AhR-mediated events are used to predict human risk, our data demonstrate a need to compensate for the species differences in sensitivity. Interspecies relative potency factors can partially compensate for our observations that humans may be less sensitive than animals to TCDD and PCBs and that the relative potencies of these chemicals in humans may be quite different than those observed in animals. Hepatocytes from ten additional human donors have been tested for EROD responsiveness to TCDD and PCB 126. The additional results are consistent with the findings reported in this study and support our conclusions (Koganti, personal communication). Additional work with human cells is clearly needed.
NOTES
Portions of these findings were presented before the National Research Council's Committee on "EPA's Exposure and Human Health Reassessment of TCDD and Related Compounds" on March 21, 2005, at the National Academy of Sciences auditorium, Washington, D.C. and to the World Health Organization Expert Panel for the Re-evaluation of Mammalian Toxic Equivalency Factors (TEFs) of Dioxins and Dioxin-like Compounds, Public Session, June 27, 2005, WHO, Geneva, Switzerland.
ACKNOWLEDGMENTS
Conflict of interest: Five of the six authors are employed by General Electric Company. This work was wholly funded by General Electric Company.
REFERENCES
Arnold, D. L., Bryce, F., McGuire, P. F., Stapley, R., Tanner, J. R., Wrenshall, E., Mes, J., Fernie, S., Tryphonas, H., Hayward, S., et al. (1995). Toxicological consequences of Aroclor 1254 ingestion by female rhesus (Macaca mulatta) monkeys. Part 2. Reproduction and infant findings. Food Chem. Toxicol. 33, 457–474.
Birnbaum, L. S., and DeVito, M. J. (1995). Use of toxic equivalency factors for risk assessment for dioxins and related compounds. Toxicology 105, 391–401.
Boutros, P. C., Moffat, I. D., Franc, M. A., Tijet, N., Tuomisto, J., Pohjanvirta, R., and Okey, A. B. (2004). Dioxin-responsive AHRE-II gene battery: Identification by phylogenetic footprinting. Biochem. Biophys. Res. Commun. 321, 707–715.
Burczynski, M. E., McMillian, M., Parker, J. B., Bryant, S., Leone, A., Grant, E. R., Thorne, J. M., Zhong, Z., Zivin, R. A., and Johnson, M. D. (2001). Cytochrome P450 induction in rat hepatocytes assessed by quantitative real-time reverse-transcription polymerase chain reaction and the RNA invasive cleavage assay. Drug Metab. Dispos. 29, 1243–1250.
Burke, M. D., Thompson, S., Elcombe, C. R., Halpert, J., Haaparanta, T., and Mayer, R. T. (1985). Ethoxy-, pentoxy- and benzyloxyphenoxazones and homologues: A series of substrates to distinguish between different induced cytochromes P-450. Biochem. Pharmacol. 34, 3337–3345.
Carlson, D. B., and Perdew, G. H. (2002). A dynamic role for the Ah receptor in cell signaling Insights from a diverse group of Ah receptor interacting proteins. J. Biochem. Mol. Toxicol. 16, 317–325.
Chen, G. S., and Bunce, N. J. (2004). Interaction between halogenated aromatic compounds in the Ah receptor signal transduction pathway. Environ. Toxicol. 19, 480–489.
Chung, I., and Bresnick, E. (1994). 3-Methylcholanthrene-mediated induction of cytochrome P4501A2 in human hepatoma HepG2 cells as quantified by the reverse transcription-polymerase chain reaction. Arch. Biochem. Biophys. 314, 75–81.
Cook, J. C., Dold, K. M., and Greenlee, W. F. (1987). An in vitro model for studying the toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin to human thymus. Toxicol. Appl. Pharmacol. 89, 256–268.
DeVito, M. J., Ménache, M. G., Diliberto, J. J., Ross, D. G., and Birnbaum, L. S. (2000). Dose-response relationships for induction of CYP1A1 and CYP1A2 enzyme activity in liver, lung, and skin in female mice following subchronic exposure to polychlorinated biphenyls. Toxicol. Appl. Pharmacol. 167, 157–172.
Donato, M. T., Gomez-Lechon, M. J., and Castell, J. V. (1993). A microassay for measuring cytochrome P450IA1 and P450IIB1 activities in intact human and rat hepatocytes cultured on 96-well plates. Anal. Biochem. 213, 29–33.
Drahushuk, A. T., McGarrigle, B. P., Slezak, B. P., Stegeman, J. J., and Olson, J. R. (1999). Time- and concentration-dependent induction of CYP1A1 and CYP1A2 in precision-cut rat liver slices incubated in dynamic organ culture in the presence of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 155, 127–138.
Eis, P. S., Olson, M. C., Takova, T., Curtis, M. L., Olson, S. M., Vener, T. I., Ip, H. S., Vedvik, K. L., Bartholomay, C. T., Allawi, H. T., et al. (2001). An invasive cleavage assay for direct quantitation of specific RNAs. Nat. Biotechnol. 19, 673–676.
GraphPad (2005). GraphPad Prism version 4.02 for Windows, GraphPad Software, San Diego, CA, www.graphpad.com.
Hall, J. G., Eis, P. S., Law, S. M., Reynaldo, L. P., Prudent, J. R., Marshall, D. J., Allawi, H. T., Mast, A. L., Dahlberg, J. E., Kwiatkowski, R. W., et al. (2000). Sensitive detection of DNA polymorphisms by the serial invasive signal amplification reaction. Proc. Natl. Acad. Sci. U.S.A. 97, 8272–8277.
Hestermann, E. V., Stegeman, J. J., and Hahn, M. E. (2000). Serum alters the uptake and relative potencies of halogenated aromatic hydrocarbons in cell culture bioassays. Toxicol. Sci. 53, 316–325.
Kimbrough, R. D., and Krouskas, C. A. (2003). Human exposure to polychlorinated biphenyls and health effects: A critical synopsis. Toxicol. Rev. 22, 217–233.
Li, A. P., Roque, M. A., Beck, D. J., and Kaminski, D. L. (1992). Isolation and culturing of hepatocytes from human livers. J. Tissue Cult. Methods 14, 139–146.
Li, W., Harper, P. A., Tang, B. K., and Okey, A. B. (1998). Regulation of cytochrome P450 enzymes by aryl hydrocarbon receptor in human cells: CYP1A2 expression in the LS180 colon carcinoma cell line after treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin or 3-methylcholanthrene. Biochem. Pharmacol. 56, 599–612.
Lipp, H.-P., Schrenk, D., Wiesmüller, T., Hagenmaier, H., and Bock, K. W. (1992). Assessment of biological activities of mixtures of polychlorinated dibenzo-p-dioxins (PCDDs) and their constituents in human HepG2 cells. Arch. Toxicol. 66, 220–223.
Mayes, B. A., McConnell, E. E., Neal, B. H., Brunner, M. J., Hamilton, S. B., Sullivan, T. M., Peters, A. C., Ryan, M. J., Toft, J. D., Singer, A. W., et al. (1998). Comparative carcinogenicity in Sprague-Dawley rats of the polychlorinated biphenyl mixtures Aroclors 1016, 1242, 1254, and 1260. Toxicol. Sci. 41, 62–76.
Micka, J., Milatovich, A., Menon, A., Grabowski, G. A., Puga, A., and Nebert, D. W. (1997). Human Ah receptor (AHR) gene: Localization to 7p15 and suggestive correlation of polymorphism with CYP1A1 inducibility. Pharmacogenetics 7, 95–101.
Moriguchi, T., Motohashi, H., Hosoya, T., Nakajima, O., Takahashi, S., Ohsako, S., Aoki, Y., Nishimura, N., Tohyama, C., Fujii-Kuriyama, Y., et al. (2003). Distinct response to dioxin in an arylhydrocarbon receptor (AHR)-humanized mouse. Proc. Natl. Acad. Sci. U.S.A. 100, 5652–5657.
Mosmann, T. (1983). Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63.
Nebert, D. W., Puga, A., and Vasiliou, V. (1993). Role of the Ah receptor and the dioxin-inducible [Ah] gene battery in toxicity, cancer, and signal transduction. Ann. N.Y. Acad. Sci. 685, 624–640.
Nerurkar, P. V., Park, S. S., Thomas, P. E., Nims, R. W., and Lubet, R. A. (1993). Methoxyresorufin and benzyloxyresorufin: Substrates preferentially metabolized by cytochromes P4501A2 and 2B, respectively, in the rat and mouse. Biochem. Pharmacol. 46, 933–943.
Okey, A. B., Riddick, D. S., and Harper, P. A. (1994). The Ah receptor: Mediator of the toxicity of 2,3,7,8-tetrachlorodibenzo- p-dioxin (TCDD) and related compounds. Toxicol. Lett. 70, 1–22.
Peters, A. K., van Londen, K., Bergman, A., Bohonowych, J., Denison, M. S., van den, B. M., and Sanderson, J. T. (2004). Effects of polybrominated diphenyl ethers on basal and TCDD-induced ethoxyresorufin activity and cytochrome P450–1A1 expression in MCF-7, HepG2, and H4IIE cells. Toxicol. Sci. 82, 488–496.
Roberts, E. A., Johnson, K. C., Harper, P. A., and Okey, A. B. (1990). Characterization of the Ah receptor mediating aryl hydrocarbon hydroxylase induction in the human liver cell line Hep G2. Arch. Biochem. Biophys. 276, 442–450.
Roberts, E. A., Shear, N. H., Okey, A. B., and Manchester, D. K. (1985). The Ah receptor and dioxin toxicity: From rodent to human tissues. Chemosphere 14, 661–674.
Roberts, E. A., Vella, L. M., Golas, C. L., Dafoe, L. A., and Okey, A. B. (1989). Ah receptor in spleen of rodent and primate species: Detection by binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Can. J. Physiol. Pharmacol. 67, 594–600.
Rowlands, J. C., and Gustafsson, J. A. (1997). Aryl hydrocarbon receptor-mediated signal transduction. Crit. Rev. Toxicol. 27, 109–134.
Santostefano, M. J., Ross, D. G., Savas, U., Jefcoate, C. R., and Birnbaum, L. S. (1997). Differential time-course and dose-response relationships of TCDD- induced CYP1B1, CYP1A1, and CYP1A2 proteins in rats. Biochem. Biophys. Res. Commun. 233, 20–24.
Schrenk, D., Lipp, H. P., Wiesmüller, T., Hagenmaier, H., and Bock, K. W. (1991). Assessment of biological activities of mixtures of polychlorinated dibenzo-p-dioxins: Comparison between defined mixtures and their constituents. Arch. Toxicol. 65, 114–118.
Schrenk, D., Stüven, T., Gohl, G., Viebahn, R., and Bock, K. W. (1995). Induction of CYP1A and glutathione S-transferase activities by 2,3,7,8-tetrachlorodibenzo-p-dioxin in human hepatocyte cultures. Carcinogenesis 16, 943–946.
Staskal, D. F., Diliberto, J. J., DeVito, M. J., and Birnbaum, L. S. (2005). Inhibition of Human and Rat CYP1A2 by TCDD and Dioxin-like Chemicals. Toxicol. Sci. 84, 225–231.
Tritscher, A. M., Goldstein, J. A., Portier, C. J., McCoy, Z., Clark, G. C., and Lucier, G. W. (1992). Dose-response relationships for chronic exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin in a rat tumor promotion model: Quantification and immunolocalization of CYP1A1 and CYP1A2 in the liver. Cancer Res. 52, 3436–3442.
Vakharia, D. D., Liu, N., Pause, R., Fasco, M., Bessette, E., Zhang, Q. Y., and Kaminsky, L. S. (2001). Polycyclic aromatic hydrocarbon/metal mixtures: Effect on PAH induction of CYP1A1 in human HEPG2 cells. Drug Metab. Dispos. 29, 999–1006.
Vamvakas, A., Keller, J., and Dufresne, M. (1996). In vitro induction of CYP 1A1-associated activities in human and rodent cell lines by commercial and tissue-extracted halogenated aromatic hydrocarbons. Environ. Toxicol. Chem. 15, 814–823.
van den Berg, M., Birnbaum, L., Bosveld, A. T. C., Brunstrm, B., Cook, P., Feeley, M., Giesy, J. P., Hanberg, A., Hasegawa, R., Kennedy, S. W., et al. (1998). Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ. Health Perspect. 106, 775–792.
Vanden Heuvel, J. P., Clark, G. C., Kohn, M. C., Tritscher, A. M., Greenlee, W. F., Lucier, G. W., and Bell, D. A. (1994). Dioxin-responsive genes: Examination of dose-response relationships using quantitative reverse transcriptase-polymerase chain reaction. Cancer Res. 54, 62–68.
Walker, N. J., Portier, C. J., Lax, S. F., Crofts, F. G., Li, Y., Lucier, G. W., and Sutter, T. R. (1999). Characterization of the dose-response of CYP1B1, CYP1A1, and CYP1A2 in the liver of female Sprague-Dawley rats following chronic exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 154, 279–286.
Wiebel, F. J., Wegenke, M., and Kiefer, F. (1996). Bioassay for determining 2,3,7,8-tetrachlorodibenzo-p-dioxin equivalents (TEs) in human hepatoma HepG2 cells. Toxicol. Lett. 88, 335–338.
Xu, L., Li, A. P., Kaminski, D. L., and Ruh, M. F. (2000). 2,3,7,8 tetrachlorodibenzo-p-dioxin induction of cytochrome P4501A in cultured rat and human hepatocytes. Chem. Biol. Interact. 124, 173–189.
Zeiger, M., Haag, R., Hockel, J., Schrenk, D., and Schmitz, H.-J. (2001). Inducing effects of dioxin-like polychlorinated biphenyls on CYP1A in the human hepatoblastoma cell line HepG2, the rat hepatoma cell line H4IIE, and rat primary hepatocytes: Comparison of relative potencies. Toxicol. Sci. 63, 65–73.(Jay B. Silkworth, Aruna K)
In Vitro Technologies, Inc., Baltimore, Maryland 21227
General Electric Company, Fairfield, Connecticut 06431
ABSTRACT
2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and related chemicals induce cytochrome P450 1A (CYP1A) gene expression and, at sufficient exposures, cause toxicity. Human health risks from such exposures are typically estimated from animal studies. We tested whether animal models predict human sensitivity by characterizing CYP1A gene expression in cultures of fresh hepatocytes from human donors, rats, and rhesus monkeys and HepG2 human hepatoma cells. We exposed the cells to three aryl hydrocarbon receptor (AhR) ligands of current environmental interest and measured 7-ethoxyresorufin-O-deethylase (EROD) activity and concentrations of CYP1A1 and CYP1A2 mRNA. We found that human cells are about 10–1000 times less sensitive to TCDD, 3,3',4,4',5-pentachlorobiphenyl (PCB 126), and Aroclor 1254 than rat and monkey cells, that relative potencies among these chemicals are different across species, and that gene expression thresholds exist for these chemicals. Newly calculated rat–human interspecies relative potency factors for PCB 126 were more than 100 times lower than the current rodent-derived value. We propose that human-derived values be used to improve the accuracy of estimates of human health risks.
Key Words: polychlorinated biphenyls; PCB; TCDD; dioxin; dioxin equivalents; TEQ; relative potency factors; human hepatocytes; HepG2 cells; cytochrome P450 1A; CYP1A1; CYP1A2; risk assessment.
INTRODUCTION
Polychlorinated dibenzo-p-dioxins, dibenzofurans, and PCBs comprise a group of highly regulated environmental contaminants of global concern. A subset of these compounds can bind to, and activate, a natural cellular receptor, the aryl hydrocarbon receptor (AhR), which is a transcription factor (Carlson and Perdew, 2002; Nebert et al., 1993). This can result in the up-regulation of many genes (Boutros et al., 2004; Okey et al., 1994), including those encoding metabolic enzymes such as the cytochrome P450 (CYP) isozymes (Rowlands and Gustafsson, 1997). The induction of CYP1A1 correlates well with exposure in animals (Tritscher et al., 1992; Vanden Heuvel et al., 1994) and is used as an important basis for estimating human health risk (Birnbaum and DeVito, 1995). The World Health Organization (WHO) has established toxic equivalency factors (TEFs) for each of 29 AhR ligands, including 7 dioxins, 10 polychlorinated dibenzofurans, and 12 PCBs. Because 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is the highest affinity ligand for the AhR, and by far the most potent at eliciting biologic responses, it was assigned a TEF of 1. TEFs for other chemicals are based on relative potency (REP) values relative to that of TCDD using both in vivo and in vitro animal studies, and it is assumed that these TEFs (WHO98 TEFs) are appropriate for use in human risk assessment (van den Berg et al., 1998). We tested this assumption by comparing the responses of fresh hepatocytes from rats, rhesus monkeys, and human donors, and also a human-derived tumor cell line, HepG2, to TCDD, PCB 126 (3,3',4,4',5-pentachlorobiphenyl), and Aroclor 1254, a commercial mixture of PCBs. PCB 126 has a TEF of 0.1. Aroclor 1254 has a calculated WHO98 toxic equivalency (TEQ) of 0.000047 (Mayes et al., 1998), largely contributed by PCB 126.
Studies that compared CYP1A1 induction in mammalian cells or cell lines suggest that humans are less sensitive than rats to TCDD. Some studies measured TCDD-induced 7-ethoxyresorufin-O-deethylase (EROD) activity in the human HepG2 cell line (Lipp et al., 1992), while other studies measured EROD activity in primary human hepatocytes (Schrenk et al., 1995). When the EC50s calculated in these studies were compared to those of an earlier study with rat primary hepatocytes and H4IIE cells (Schrenk et al., 1991), the human cells were found to be 8–19 times less sensitive than the rat cells to TCDD. Wiebel et al. (1996) compared TCDD-induced CYP1A activity in human HepG2 cells and rat H4IIE cells and found that the human cells were 20 times less sensitive. Xu et al. (2000) demonstrated that primary human hepatocytes were less sensitive to TCDD than primary rat hepatocytes and also reported species differences in the expression of CYP1A1 and CYP1A2 mRNA.
Few studies have measured CYP1A induction by TCDD and PCB 126 in more than one species using the same methodology. Vamvakas et al. (1996) tested TCDD and several PCB congeners in the HepG2 and MCF-7 human cell lines and in the rat H4IIE cell line. Zeiger et al. (2001) compared the induction by TCDD and several PCB congeners in primary rat hepatocytes and the rat H4IIE and human HepG2 cell lines. Recently, Peters et al. (2004) reported the induction of EROD activity by TCDD and PCB 126 in human MCF-7 and HepG2 cells and in rat H4IIE cells. Each of these studies showed that the human cells were less sensitive to TCDD and PCB 126 than rat-derived cells.
In addition to understanding the relative sensitivity between species to these chemicals, it is also important to determine if the REPs among these chemicals are consistent across species, especially if REPs derived from data for one species are used to extrapolate effects to other species. In fact, studies with mice (DeVito et al., 2000) and human-derived cells (Zeiger et al., 2001) have suggested that the REPs for several PCB congeners were inconsistent with their current WHO98 TEFs. More studies are needed to characterize how human responses to AhR ligands differ from the responses of other species while also characterizing the consistency of the relative potencies of AhR ligands across species.
Thus, to evaluate species differences in the CYP1A responses to TCDD, PCB 126, and Aroclor 1254, and to determine the REPs for PCB 126 and Aroclor 1254 in each species, we compared the responses of all cells under identical conditions. This included the use of a single lot number of each test chemical throughout the study and the preparation of only one or two stock solutions. Because immortal cell lines may not accurately reflect responses of normal human hepatocytes, we tested both fresh human hepatocytes and the HepG2 cell line. Use of donor cells also permitted an evaluation of differences among five humans. Fresh rat hepatocytes and HepG2 cells were tested to compare our methodology and results with earlier work (Lipp et al., 1992; Schrenk et al., 1995; Zeiger et al., 2001). Hepatocytes from rhesus monkeys (Macaca mulatta) also were included, since both reproductive and immunological effects have been reported for this primate following exposure to PCBs (Arnold et al., 1995). All four of these cell types are known to express an AhR (Roberts et al., 1985, 1989, 1990). To sample the diversity of humans, we tested hepatocytes from five organ donors, two Caucasians of each gender and one African-American male.
Cells were treated over a wide concentration range of each chemical in serum-free culture medium (Hestermann et al., 2000). CYP1A induction was determined after 48 h of exposure by measuring EROD activity and CYP1A mRNA. EROD activity was not obtained for one of the five donors due to experimental error. CYP1A1 and CYP1A2 mRNAs were measured in two to seven experiments for each species. Species differences were evaluated by comparing thresholds, EC50s, and maximal responses for each chemical. The potencies of PCB 126 and Aroclor 1254, relative to TCDD, were calculated for each measurement and compared across species to derive interspecies relative potency factors.
MATERIALS AND METHODS
Chemicals.
TCDD (molecular weight = 322) was obtained from Accustandard (New Haven, CT; catalog no. D404N; CAS no. 1746–01–6; Lot no. 970401R-AC; 99.1% pure). The single contaminant was a pentachloro-hydroxydiphenyl ether by GC/MS.
PCB 126 (molecular weight = 326.4) was obtained from Accustandard (Catalog no. C-126N; CAS no. 57465–28–8; Lot no. 081699MT-AC; 99.2% pure). The single contaminant was identified as a tetrachlorobiphenyl by GC/MS.
Aroclor 1254, lot no. 122–078 (molecular weight = 326.2) was from the same lot of material used in an earlier chronic bioassay conducted for General Electric Company (Mayes et al., 1998). The calculated WHO98 TEQ (i.e., the toxic equivalency to TCDD found by summing the products of the toxic equivalency factor of each of the 12 PCB congeners with an assigned TEF multiplied by its respective measured concentration) is 47 ppm.
Hepatocyte sources.
Human hepatocytes were prepared from nontransplantable human tissue acquired after informed consent for use in research by In Vitro Technologies, Inc. (IVT). An external FDA-certified Institutional Review Board approved the use of nontransplantable human tissue for ADME-Tox research at IVT. Donor 1, (IID), IVT Lot MHU-L-012303, was a 66-year-old Caucasian male who died from an intracranial hemorrhage. Donor 2, (KZO), IVT Lot FHU-L-020203, was a 42-year-old Caucasian female who died from a cerebrovascular accident. Donor 3, (WRG), IVT Lot MHU-L-052004, was a 41-year-old Caucasian male who died from an astrocytoma. Donor 4, (RFA) IVT Lot FHU-L-072004, was a 56-year-old Caucasian female who died from a cerebrovascular accident. EROD data were not collected for this donor because of experimental error but CYP1A mRNA data were obtained. Donor 5, (ZYZ) IVT Lot MHU-L-0730044, was a 46-year-old African-American male who died from anoxia. Serologies for all donors were negative for HIV, HBV, and HCV, but positive for cytomegalovirus. Urinalyses and blood chemistries for all donors were within normal limits. See www.invitrotech.com/characterizationtab.cfm for additional donor information.
Rhesus monkey hepatocytes were isolated from liver tissue from a single chemically naive young adult female for each of three trials. The tissue was purchased by IVT from approved sources that comply with all appropriate laws and guidelines.
Rat hepatocytes were isolated by IVT, from two rats (Female Crl:CD (SD)IGS BR, Charles River Laboratories, Wilmington, MA) and pooled for each of five experiments. Rats were treated in accordance with the Animal Welfare Act.
The HepG2 human hepatoma cell line was obtained from the American Type Culture Collection, Manassas, VA.
Hepatocyte cultures.
Hepatocytes were isolated according to the two-step collagenase perfusion procedure of Li et al. (1992). Isolated hepatocytes were counted using Trypan blue exclusion to determine yield and viability. Only hepatocyte preparations with 70% viability were used. Freshly isolated hepatocytes from rat, monkey, or human donors were plated, in triplicate, onto collagen-coated 24-well plates at a cell density of 3.5 x 105 cells per well in Plating Medium (Dulbecco's modified Eagle's medium (DMEM) supplemented with bovine serum albumin, fructose, HEPES, sodium bicarbonate, L-glutamine (2.4 mM), hydrocortisone (2.38 μM), insulin (135 nM), MEM nonessential amino acids (1.2%), amikacin, penicillin (200,000 U/l), streptomycin (200 mg/l), gentamycin, and Fungizone). Fungizone was excluded from HepG2 cell Plating Medium. Cultures were placed in a 37°C/5% CO2 incubator for 2 days before use to establish the primary hepatocyte and HepG2 cell monolayers. Confluency was visually checked each day of the culture period and was generally 90–100%, except for Donor 1, for which it was 75–90%.
Chemical treatments.
Each test chemical was prepared in dimethyl sulfoxide (DMSO) at 200 times (200x) the final concentration and diluted with incubation medium to the required concentrations. TCDD was soluble in DMSO at 60 μM, but not 200 μM. PCB 126 was soluble in DMSO at 2000 μM, but not 6000 μM. Aroclor 1254 was soluble in DMSO up to at least 60 mM. Two 200x stock solutions were prepared for TCDD and PCB 126. Eleven concentrations were used for each trial, in addition to the vehicle control. TCDD was used at 0.00001, 0.0001, 0.001, 0.003, 0.01, 0.03, 0.1, 0.3, 1, 10, and 100 nM. PCB 126 was used at 0.001, 0.01, 0.03, 0.1, 0.3, 1, 3, 10, 100, 1000, and 10,000 nM. Aroclor 1254 was used at 1, 3, 10, 30, 100, 300, 1000, 3000, 10,000, 30,000, and 300,000 nM.
Established primary hepatocyte and HepG2 cell cultures were treated with 500 μl of serum-free Incubation Medium (Plating Medium without serum) containing TCDD, PCB 126, or Aroclor 1254. Serum-free medium was used, since serum can significantly reduce the cellular uptake of these compounds (Hestermann et al., 2000). The final concentration of DMSO in the incubations was 0.5%, consistent with previous studies with these chemicals and cells (Hestermann et al., 2000; Lipp et al., 1992; Zeiger et al., 2001). There were no visible indications that any chemicals (or any subsets of congeners in the Aroclor 1254) had precipitated at any of the incubation concentrations, but this was not analytically confirmed. Cultures were incubated at 37°C/5% CO2 with test chemicals for a total of 48 h. The incubation medium containing the test chemical was replaced at 24 h.
Culture viability was assessed in a replicate set of cultures for each cell type and chemical treatment by measuring the metabolic conversion of 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) to the formazan product in triplicate wells (Mosmann, 1983). TCDD and PCB 126 did not affect culture viability at any concentration tested. Aroclor 1254 reduced MTT conversion in rhesus cells to approximately 50% of control levels at concentrations 10–5 M. At 3 x 10–4 M, Aroclor 1254 reduced MTT conversion to <2%, 41%, 2%, and 26% in rhesus, rat, HepG2, and 1 donor, respectively. Three donors were not affected at this concentration (data not shown).
CYP1A enzyme activity assay.
EROD activity was determined using modifications of the methods of Burke et al. (1985) and Donato et al. (1993). Medium containing the test chemical was removed from the culture plates and replaced with 300 μl of Krebs-Henseleit buffer (supplemented with amikacin, calcium chloride, gentamicin, heptanoic acid, N-(2-hydroxyethyl)piperazine-N'-(2-ethanesulfonate), and sodium bicarbonate) containing 10 μM ethoxyresorufin and 3 mM salicylamide. Cultures were incubated at 37°C/5% CO2 for 30 min. Incubations were terminated by removing the supernatant and adding it to 300 μl of methanol containing 2% DMSO. The amount of resorufin formed in the incubations was quantitated against a standard curve of resorufin as measured by a fluorometric assay (excitation: 530 nm; emission: 590 nm; Wallac Victor2 plate reader, Perkin Elmer). The quantitation limit was 0.1 pmol resorufin/(min x mg protein).
Protein concentrations of the cultures were determined using a BCA protein assay kit (Pierce Chemical Co., Rockford, ILL). Cells were lysed by adding 200 μl of 0.1 N NaOH and incubating at 37°C/5% CO2 for 30 min. Cell lysates were scraped and harvested. Ten microliters of each cell lysate was mixed with 200 μl of BCA working reagent (reagent A:reagent B (49:1)) and incubated for 30 min at 37°C. Protein concentrations were quantified against a standard curved using absorbance at 572 nm (Wallac Victor2 plate reader) and were generally about 0.1 mg protein/culture well.
RNA isolation.
Medium containing the test chemical was removed from each well, and cultures were washed with 500 μl of phosphate buffered saline. Phosphate buffered saline was then replaced with 200 μl of TRIzol reagent (Invitrogen, Carlsbad, CA). Hepatocytes were scraped from each well with a wide-bore pipette tip, and homogenized by gently pipetting up and down. Cell homogenates were transferred to microcentrifuge tubes, and 200 μl of chloroform was added to each tube. Tubes were mixed for 15 s. Homogenates were placed on ice for 5 min and then centrifuged at 12,000 x g for 15 min at 4°C. The aqueous layer was carefully removed and transferred to a fresh microcentrifuge tube, and 200 μl of phenol:chloroform:isoamyl alcohol (25:24:1) was added. Tubes were mixed for 15 s. Mixtures were placed on ice for 5 min and then centrifuged at 12,000 x g for 15 min at 4°C. The aqueous layer was removed to a fresh microcentrifuge tube. RNA from the aqueous layer was precipitated by adding 200 μl of isopropanol and placing the tubes at –20°C overnight. Following overnight precipitation, the tubes were stored at –70°C until analysis.
RNA quantitation.
The RNA samples were centrifuged at 12,000 x g for 15 min. Supernatants were decanted, and the RNA pellets were washed with 70% ethanol in Tris EDTA buffer. Samples were centrifuged at 12,000 x g for 30 min. Supernatants were decanted, and the pellets were allowed to air dry. RNA pellets were reconstituted with 50 μl of TE buffer. RNA was quantitated using a Ribogreen RNA quantitation kit (Molecular Probes, Eugene, OR). In brief, 5 μl of each RNA sample was diluted 50-fold with TE buffer. A 100-μl aliquot of the diluted RNA was mixed with 100 μl of the diluted RNA quantitation reagent and incubated for 5 min at room temperature. RNA in the samples was quantitated against a standard curve using fluorometric assay (excitation: 485 nm; emission: 535 nm; Wallac Victor2 plate reader).
Quantitation of specific mRNA—RNA Invader invasive cleavage assay.
The RNA Invader invasive cleavage assay is an invasive cleavage amplification assay developed by Third Wave Technologies, Inc., Madison, WI (Burczynski et al., 2001; Hall et al., 2000). In this method, two oligonucleotides, an upstream oligonucleotide and the probe, anneal to the target sequence. The probe contains both a target-specific and a noncomplimentary region. When the probe overlaps the upstream oligonucleotide a reporter sequence can be cleaved. The rapid turnover and production of the cleaved sequence permits a secondary reaction and the linear amplification of a fluorescent signal. This method has the ability to discriminate between two highly homologous sequences such as found with the P450s (Eis et al., 2001). Total RNA isolated from each triplicate culture for each concentration and for each species was hybridized against human/rat CYP1A1 (h/rCYP1A1) or h/rCYP1A2 probes. CYP1A1 probes targeted human GeneBank Accession no. K03191 sequence (5'-CCTGATTGAGCACTGTCAGGA-3') and rat GeneBank Accession no. NM_012540 sequence (5'-CCTCATTGAGCATTGTCTGGA-3'). CYP1A2 probes targeted human GeneBank Accession no. M55053 sequence (5'-AGGAGCACTATCAGGACTTTGACAAG-3') and rat GeneBank Accession no. K02422 sequence (5'-AGGAACACTATCAAGACTTCAACAAG-3'). The human CYP1A1 probe was also used with rhesus hepatocytes because these two species share a consensus sequence at the targeted region. The human CYP1A2 probe could not be used for the rhesus hepatocytes because the targeted sequence is not shared, as we confirmed. CYP1A1 and CYP1A2 mRNAs are expressed as amol (10–18 moles) specific mRNA/ng total RNA. Human and rat GAPDH mRNAs were also measured, as reference mRNAs, and found to be consistently expressed for each chemical, all treatment concentrations, and all species (data not shown).
Each RNA sample was diluted to a maximum concentration of 10 ng/μl. If the sample concentration was below 10 ng/μl, it was used without further dilution. The amount of specific mRNA in each RNA sample was quantified using Invader assay kits (1A1, 1A2, hGAPDH, positive control sequences, and generic reagent kits from Third Wave Technologies, Inc., Madison, WI). Specific RNA in the samples was quantified against standard curves generated using the Invader oligo sequence test probes against positive control transcripts that included the targeted sequence in rats and humans. Fluorescence was measured as described above.
Dose-response modeling.
Freshly isolated hepatocytes from the three species or the human-derived HepG2 hepatoma cell line were treated in vitro for 48 h. The dose-responses for EROD activity and CYP1A mRNA were modeled by combining unsummarized triplicate culture data across all experiments for each cell type at each concentration tested using the variable slope sigmoid Hill equation (GraphPad, 2005). We defined threshold as the concentration at which the response first exceeds the model's estimated constitutive or background expression level. We estimated the threshold for each curve by determining the concentration at which a line tangent to the modeled dose-response curve at the EC05 intersects the bottom of the modeled dose-response curve, as determined by the Hill equation (Fig. 2) (GraphPad, 2005). This is an objective method to quantify the dosage at which the curve is no longer attached to its lower asymptote (i.e., the constitutive expression level). The EROD EC50 is the concentration at which the induced enzyme activity is halfway between the calculated bottom and top of each dose-response curve (GraphPad, 2005).
RESULTS
EROD Activity
EROD dose-response curves are shown in Figure 1. Species differences for each chemical are clearly evident (Fig. 1A). We conducted additional analyses of the modeled curves by examining three features of each curve: the threshold, the EC50, and the maximal response. The threshold and EC50 provide complementary information on cell sensitivity, whereas the maximal response provides information on efficacy.
For TCDD, we found the lowest thresholds in fresh rat and rhesus liver cells. Thresholds in both fresh human liver cells (p 0.05) and HepG2 cells (p 0.05) (Fig. 2) were about 10 times higher than in rat and rhesus liver cells. Thus, both human cell types were about 0.1 as sensitive to TCDD as were rat and rhesus cells. The EC50s show that the human cells were about 0.10 to 0.27 as sensitive to TCDD as either rat or rhesus cells (Table 1). These observations are contrary to the assumption normally used in risk assessment that humans are more sensitive than experimental animals. They also indicate that the current application of factor multipliers to compensate for uncertainties regarding species sensitivities may result in overestimating human risk by several orders of magnitude.
The maximal heights of the TCDD response curves, which measure the ability of an AhR ligand to elicit a full response, were similar for all three species (i.e., within a factor of three). This suggests that the sensitivity for EROD induction, measured by either threshold or EC50, differentiates well among species while the maximal enzyme activity level does not. This also suggests that measures of sensitivity, because they are very different among species, are more important than maximal activity levels in estimating risk across species. Because the concentrations at which thresholds, EC50s, and the maximal responses occur are similar for HepG2 and donor cells, we suggest that either type of human cells can be used to study interspecies sensitivities to chemicals, but fresh human cells have the added advantage of providing information on the extent of individual variability.
Even more pronounced differences between human cells and both rat and rhesus cells were seen when the PCB 126 EROD dose responses were compared. Rhesus cells responded in a manner more similar to rats than human cells (Fig. 1A). The thresholds and EC50s indicated that both donor and HepG2 cells were between 0.01 and 0.001 as sensitive to PCB 126 as were rat and rhesus cells (Fig. 2, Table 1). Similar species differences were observed with Aroclor 1254. Donor and HepG2 cells were 0.01 as sensitive to Aroclor 1254 as were rhesus and rat cells, based on either thresholds (Figs. 1A and 2) or estimated EC50 values (Table 1), although precise EC50s could not be calculated.
By comparing the heights of the PCB 126 response curves with those of TCDD, we see that PCB 126 is about 0.9 and 0.8 as efficacious as TCDD in rhesus and rat cells, respectively, but only 0.5 (p 0.05) and 0.6 (p 0.05) as efficacious as TCDD in HepG2 and donor cells, respectively. This suggests that there are species-dependent factors other than those that determine sensitivity that influence differential CYP1A1 gene expression.
In comparing the EROD response of each species to their respective TCDD response, we found that PCB 126 is about 0.1 as potent as TCDD in rat and rhesus cells. This value is consistent with the WHO98 TEF for PCB 126 of 0.1. However, PCB 126 is between 0.01 and 0.001 as potent as TCDD in donor and HepG2 cells, based on thresholds (Figs. 1A and 2) and EC50 values (Table 1).
Our data show that human cells are about 0.1 as sensitive to TCDD as rats and rhesus cells. Our data also show that human cells are 0.001 or less as sensitive to PCB 126 as rat and rhesus cells are to TCDD and less than 0.000001 as sensitive to Aroclor 1254 as rats are to TCDD (Table 1). The human EROD responses to TCDD, PCB 126, and Aroclor 1254 are compared to only the responses of rats to TCDD in Figure 3 to facilitate the direct comparison of their EROD dose-response curves and EC50s in Tables 1 and 2.
Human Diversity in EROD Response
The diversity of human responsiveness, characterized by both sensitivity and maximal response, is an important concern for risk managers responsible for protecting sensitive populations. An earlier report suggests that humans have similar sensitivities but variable maximal responses to TCDD (Schrenk et al., 1995). In our study, we found similar thresholds for Donors 1, 2, 3, and 5 that were within an 11-fold concentration range (Fig. 1A) (calculation not shown). The EC50s of Donors 1, 2, 3, and 5 were within a seven-fold range, i.e., 1.1 x 10–10 M (8.9 x 10–11 M to 1.3 x 10–10 M, r2 = 0.95); 7.3 x 10–10 M (5.7 x 10–10 M to 9.3 x 10–10 M, r2 = 0.98); 1.2 x 10–8 (1.9 x 10–9 to 7.1 x 10–8, r2 = 0.95 [estimated because the maximal response could not be established with certainty]); and 1.1 x 10–10 M (8.5 x 10–11 M to 1.5 x 10–10 M, r2 = 0.97), respectively. The maximal responses of Donors 1, 2, and 5 varied by less than 2-fold for each chemical (Fig. 1B). Donor 3 was a poor responder to both TCDD and PCB 126 and had no measurable response to Aroclor 1254. This donor was excluded from the aggregate model shown in Fig. 1A (thus avoiding a shift in the curve further to the right) but is shown separately in Fig. 1B. Compared to rats, all donors responded poorly to Aroclor 1254. Our findings that sensitivity differences, measured by either threshold or EC50s, span over three orders of magnitude between human and rat cells, but only vary by a factor of about ten among the human samples, suggest that species differences are a more significant source than individual differences for the uncertainty in risk estimation.
CYP1A mRNA Induction
Measurement of CYP1A1 mRNA serves as an indicator of Ah gene expression before any posttranslational modifications or interferences occur that could affect protein levels and/or function (Staskal et al., 2005). Furthermore, mRNA levels provide a way to measure the competency of ligands with different AhR binding affinities to activate Ah gene expression (Chen et al., 2004). Thus, mRNA levels could provide additional information on species differences not provided by EROD activity alone. Since both CYP1A1 and CYP1A2 express EROD activity (Nerurkar et al., 1993), we measured CYP1A1 and CYP1A2 mRNA in a replicate set of cultures treated as described above using the RNA Invader invasive cleavage assay (Eis et al., 2001).
The induction profile of CYP1A1 mRNA for each chemical (Fig. 4A) was generally consistent with that of EROD activity for each species (Fig. 1). Interestingly, the induction of CYP1A1 mRNA by TCDD in rats, the most sensitive responder, was observed at concentrations about 10 times lower than seen for EROD activity. This has been observed by others who suggested that it was due to differences in the detection sensitivities of mRNA and enzyme activity (Vanden Heuvel et al., 1994). Because this was not observed in human and rhesus monkey cells for TCDD, or for PCB 126 or Aroclor 1254 in any species, our data suggest that there are aspects of TCDD induction and of the regulation of CYP1A1 that are unique to the rat that require further investigation. Consistent with the species differences observed for EROD activity, human cells required 100 to 1000 times higher concentrations of each chemical than rat and rhesus cells to induce CYP1A1 mRNA (Fig. 4A). This suggests that such differences are due to early events in AhR binding and/or DNA activation occurring up through mRNA transcription. Our results are also consistent with a recent investigation that showed a diminished CYP1A mRNA response when the AhR gene in mice was replaced with human AhR cDNA (Moriguchi et al., 2003), which the authors suggested might be due to structural differences in the AhR itself. Regardless of the explanation, the absence of such early induction events in human cells at exposure levels that elicit responses in other species suggests that animal models are overly sensitive when predicting human responses associated with Ah gene activation.
Individual variations in the AhR structure and function have also been described for several human cells (Cook et al., 1987; Micka et al., 1997). However, in our study, the CYP1A1 mRNA responses of four of the five donors were similar (Fig. 4B). These donors responded to TCDD at approximately 10–10 M and to PCB 126 between 10–8 and 10–7 M, and the respective responses reached amplitudes that were generally within a factor of three of one another. In contrast, the CYP1A1 mRNA responses of Donor 5 to all three chemicals were very low—but the CYP1A2 mRNA responses were robust (Fig. 5). Consistent with the Aroclor 1254 EROD results, only Donor 1 had a robust Aroclor 1254 CYP1A1 mRNA response, and the responses of the other four donors were barely detectable.
The regulation of both CYP1A1 and CYP1A2 mRNA induction seem to be similar in donor cells but dissimilar in rat cells (Figs. 4 and 5, Table 1). Donor cells responded to TCDD with nearly equivalent amounts of both CYP1A1 and CYP1A2 mRNA over the same concentration range. In rats, however, CYP1A2 mRNA induction was about 0.1 as sensitive as CYP1A1 mRNA induction, and the maximal CYP1A2 mRNA levels were only about 25% of CYP1A1 levels. This predominance of CYP1A2 over CYP1A1 in cultures of rat cells is consistent with earlier work (Xu et al., 2000). Interestingly, both proteins have been shown to be equally induced by TCDD in a long-term rat in vivo study (Walker et al., 1999). The CYP1A2 mRNA EC50s shown in Table 1 indicate that rat cells were about six and three times more sensitive than donor cells to CYP1A2 induction by both TCDD and PCB 126, respectively. These findings are consistent with reports that CYP1A1 and CYP1A2 may be differently regulated in rats (Drahushuk et al., 1999; Santostefano et al., 1997), but also indicate that there are important differences in their respective regulation between humans and rats.
CYP1A2 mRNA was barely detectable in HepG2 cells, which is consistent with earlier reports (Chung et al., 1994; Li et al., 1998; Vakharia et al., 2001). Our finding that CYP1A1 and CYP1A2 mRNAs are strongly and similarly induced in fresh human hepatocytes demonstrates the value of using primary human cells, rather than cell lines, in studying the expression of these two AhR-associated genes.
We calculated new EROD- and CYP1A1 mRNA-based REPs for PCB 126 using the data from our study (Table 2). REPs are typically calculated by dividing the EC50 for TCDD, the reference ligand, by the EC50 for the AhR ligand of interest with both values from the same species. For PCB 126, the EROD-based REPs in rhesus monkeys and rats were 0.13 and 0.12, respectively, consistent with the WHO98 TEF of 0.1. However, the EROD-based REP for PCB 126 for human donor cells was 0.003 and only 0.002 for HepG2 cells. These values are consistent with earlier work with HepG2 cells (Zeiger et al., 2001) and much lower than the WHO98 TEF of 0.1. This clearly demonstrates the inadequacy of the current TEQ approach to account for the possibility that each species may have its own unique set of REPs for different AhR ligands.
To account for the accumulating evidence not only that humans and rats may have different sensitivities to TCDD (Lipp et al., 1992; Schrenk et al., 1995; Vamvakas et al., 1996; Wiebel et al., 1996; Xu et al., 2000; Zeiger et al., 2001), but also that humans may have a unique set of REPs, we then calculated human REPs relative to the rat TCDD response (r-h REP) (Table 2, Fig. 6). This approach uses data from both species and is more consistent with the empirical evidence that humans and rats have different sensitivities to TCDD and other AhR ligands than using only rodent-derived data. The results show, based on EROD induction, that an appropriate r-h REP for PCB 126 is 0.0009, compared to the WHO98 TEF of 0.1. Even lower REP values were observed for CYP1A1 mRNA (Table 2).
Using the same approach for Aroclor 1254, we calculated an EROD-based r-h REP for human donor liver cells that is at least 54 times lower than predicted using the WHO98 TEFs (Table 2, Fig. 6). Comparable differences were also seen for Aroclor 1254 CYP1A1 mRNA data, with human donor cells 600 times less sensitive than predicted by the WHO98 TEF. Both assays also suggest that rhesus monkeys are more sensitive than rats to Aroclor 1254 and indicate that neither rhesus monkeys, although primates, nor rats are good models for humans for AhR ligand-based risk assessments for this complex mixture.
DISCUSSION
Our findings show orders of magnitude species differences in sensitivity to TCDD and PCBs and highlight the substantial uncertainties that arise when using experimental animal data to extrapolate to potential human health risks. These findings may help to explain the lack of conclusive evidence that PCBs have affected human health (Kimbrough et al., 2003). To the extent that AhR-mediated events are used to predict human risk, our data demonstrate a need to compensate for the species differences in sensitivity. Interspecies relative potency factors can partially compensate for our observations that humans may be less sensitive than animals to TCDD and PCBs and that the relative potencies of these chemicals in humans may be quite different than those observed in animals. Hepatocytes from ten additional human donors have been tested for EROD responsiveness to TCDD and PCB 126. The additional results are consistent with the findings reported in this study and support our conclusions (Koganti, personal communication). Additional work with human cells is clearly needed.
NOTES
Portions of these findings were presented before the National Research Council's Committee on "EPA's Exposure and Human Health Reassessment of TCDD and Related Compounds" on March 21, 2005, at the National Academy of Sciences auditorium, Washington, D.C. and to the World Health Organization Expert Panel for the Re-evaluation of Mammalian Toxic Equivalency Factors (TEFs) of Dioxins and Dioxin-like Compounds, Public Session, June 27, 2005, WHO, Geneva, Switzerland.
ACKNOWLEDGMENTS
Conflict of interest: Five of the six authors are employed by General Electric Company. This work was wholly funded by General Electric Company.
REFERENCES
Arnold, D. L., Bryce, F., McGuire, P. F., Stapley, R., Tanner, J. R., Wrenshall, E., Mes, J., Fernie, S., Tryphonas, H., Hayward, S., et al. (1995). Toxicological consequences of Aroclor 1254 ingestion by female rhesus (Macaca mulatta) monkeys. Part 2. Reproduction and infant findings. Food Chem. Toxicol. 33, 457–474.
Birnbaum, L. S., and DeVito, M. J. (1995). Use of toxic equivalency factors for risk assessment for dioxins and related compounds. Toxicology 105, 391–401.
Boutros, P. C., Moffat, I. D., Franc, M. A., Tijet, N., Tuomisto, J., Pohjanvirta, R., and Okey, A. B. (2004). Dioxin-responsive AHRE-II gene battery: Identification by phylogenetic footprinting. Biochem. Biophys. Res. Commun. 321, 707–715.
Burczynski, M. E., McMillian, M., Parker, J. B., Bryant, S., Leone, A., Grant, E. R., Thorne, J. M., Zhong, Z., Zivin, R. A., and Johnson, M. D. (2001). Cytochrome P450 induction in rat hepatocytes assessed by quantitative real-time reverse-transcription polymerase chain reaction and the RNA invasive cleavage assay. Drug Metab. Dispos. 29, 1243–1250.
Burke, M. D., Thompson, S., Elcombe, C. R., Halpert, J., Haaparanta, T., and Mayer, R. T. (1985). Ethoxy-, pentoxy- and benzyloxyphenoxazones and homologues: A series of substrates to distinguish between different induced cytochromes P-450. Biochem. Pharmacol. 34, 3337–3345.
Carlson, D. B., and Perdew, G. H. (2002). A dynamic role for the Ah receptor in cell signaling Insights from a diverse group of Ah receptor interacting proteins. J. Biochem. Mol. Toxicol. 16, 317–325.
Chen, G. S., and Bunce, N. J. (2004). Interaction between halogenated aromatic compounds in the Ah receptor signal transduction pathway. Environ. Toxicol. 19, 480–489.
Chung, I., and Bresnick, E. (1994). 3-Methylcholanthrene-mediated induction of cytochrome P4501A2 in human hepatoma HepG2 cells as quantified by the reverse transcription-polymerase chain reaction. Arch. Biochem. Biophys. 314, 75–81.
Cook, J. C., Dold, K. M., and Greenlee, W. F. (1987). An in vitro model for studying the toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin to human thymus. Toxicol. Appl. Pharmacol. 89, 256–268.
DeVito, M. J., Ménache, M. G., Diliberto, J. J., Ross, D. G., and Birnbaum, L. S. (2000). Dose-response relationships for induction of CYP1A1 and CYP1A2 enzyme activity in liver, lung, and skin in female mice following subchronic exposure to polychlorinated biphenyls. Toxicol. Appl. Pharmacol. 167, 157–172.
Donato, M. T., Gomez-Lechon, M. J., and Castell, J. V. (1993). A microassay for measuring cytochrome P450IA1 and P450IIB1 activities in intact human and rat hepatocytes cultured on 96-well plates. Anal. Biochem. 213, 29–33.
Drahushuk, A. T., McGarrigle, B. P., Slezak, B. P., Stegeman, J. J., and Olson, J. R. (1999). Time- and concentration-dependent induction of CYP1A1 and CYP1A2 in precision-cut rat liver slices incubated in dynamic organ culture in the presence of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 155, 127–138.
Eis, P. S., Olson, M. C., Takova, T., Curtis, M. L., Olson, S. M., Vener, T. I., Ip, H. S., Vedvik, K. L., Bartholomay, C. T., Allawi, H. T., et al. (2001). An invasive cleavage assay for direct quantitation of specific RNAs. Nat. Biotechnol. 19, 673–676.
GraphPad (2005). GraphPad Prism version 4.02 for Windows, GraphPad Software, San Diego, CA, www.graphpad.com.
Hall, J. G., Eis, P. S., Law, S. M., Reynaldo, L. P., Prudent, J. R., Marshall, D. J., Allawi, H. T., Mast, A. L., Dahlberg, J. E., Kwiatkowski, R. W., et al. (2000). Sensitive detection of DNA polymorphisms by the serial invasive signal amplification reaction. Proc. Natl. Acad. Sci. U.S.A. 97, 8272–8277.
Hestermann, E. V., Stegeman, J. J., and Hahn, M. E. (2000). Serum alters the uptake and relative potencies of halogenated aromatic hydrocarbons in cell culture bioassays. Toxicol. Sci. 53, 316–325.
Kimbrough, R. D., and Krouskas, C. A. (2003). Human exposure to polychlorinated biphenyls and health effects: A critical synopsis. Toxicol. Rev. 22, 217–233.
Li, A. P., Roque, M. A., Beck, D. J., and Kaminski, D. L. (1992). Isolation and culturing of hepatocytes from human livers. J. Tissue Cult. Methods 14, 139–146.
Li, W., Harper, P. A., Tang, B. K., and Okey, A. B. (1998). Regulation of cytochrome P450 enzymes by aryl hydrocarbon receptor in human cells: CYP1A2 expression in the LS180 colon carcinoma cell line after treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin or 3-methylcholanthrene. Biochem. Pharmacol. 56, 599–612.
Lipp, H.-P., Schrenk, D., Wiesmüller, T., Hagenmaier, H., and Bock, K. W. (1992). Assessment of biological activities of mixtures of polychlorinated dibenzo-p-dioxins (PCDDs) and their constituents in human HepG2 cells. Arch. Toxicol. 66, 220–223.
Mayes, B. A., McConnell, E. E., Neal, B. H., Brunner, M. J., Hamilton, S. B., Sullivan, T. M., Peters, A. C., Ryan, M. J., Toft, J. D., Singer, A. W., et al. (1998). Comparative carcinogenicity in Sprague-Dawley rats of the polychlorinated biphenyl mixtures Aroclors 1016, 1242, 1254, and 1260. Toxicol. Sci. 41, 62–76.
Micka, J., Milatovich, A., Menon, A., Grabowski, G. A., Puga, A., and Nebert, D. W. (1997). Human Ah receptor (AHR) gene: Localization to 7p15 and suggestive correlation of polymorphism with CYP1A1 inducibility. Pharmacogenetics 7, 95–101.
Moriguchi, T., Motohashi, H., Hosoya, T., Nakajima, O., Takahashi, S., Ohsako, S., Aoki, Y., Nishimura, N., Tohyama, C., Fujii-Kuriyama, Y., et al. (2003). Distinct response to dioxin in an arylhydrocarbon receptor (AHR)-humanized mouse. Proc. Natl. Acad. Sci. U.S.A. 100, 5652–5657.
Mosmann, T. (1983). Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63.
Nebert, D. W., Puga, A., and Vasiliou, V. (1993). Role of the Ah receptor and the dioxin-inducible [Ah] gene battery in toxicity, cancer, and signal transduction. Ann. N.Y. Acad. Sci. 685, 624–640.
Nerurkar, P. V., Park, S. S., Thomas, P. E., Nims, R. W., and Lubet, R. A. (1993). Methoxyresorufin and benzyloxyresorufin: Substrates preferentially metabolized by cytochromes P4501A2 and 2B, respectively, in the rat and mouse. Biochem. Pharmacol. 46, 933–943.
Okey, A. B., Riddick, D. S., and Harper, P. A. (1994). The Ah receptor: Mediator of the toxicity of 2,3,7,8-tetrachlorodibenzo- p-dioxin (TCDD) and related compounds. Toxicol. Lett. 70, 1–22.
Peters, A. K., van Londen, K., Bergman, A., Bohonowych, J., Denison, M. S., van den, B. M., and Sanderson, J. T. (2004). Effects of polybrominated diphenyl ethers on basal and TCDD-induced ethoxyresorufin activity and cytochrome P450–1A1 expression in MCF-7, HepG2, and H4IIE cells. Toxicol. Sci. 82, 488–496.
Roberts, E. A., Johnson, K. C., Harper, P. A., and Okey, A. B. (1990). Characterization of the Ah receptor mediating aryl hydrocarbon hydroxylase induction in the human liver cell line Hep G2. Arch. Biochem. Biophys. 276, 442–450.
Roberts, E. A., Shear, N. H., Okey, A. B., and Manchester, D. K. (1985). The Ah receptor and dioxin toxicity: From rodent to human tissues. Chemosphere 14, 661–674.
Roberts, E. A., Vella, L. M., Golas, C. L., Dafoe, L. A., and Okey, A. B. (1989). Ah receptor in spleen of rodent and primate species: Detection by binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Can. J. Physiol. Pharmacol. 67, 594–600.
Rowlands, J. C., and Gustafsson, J. A. (1997). Aryl hydrocarbon receptor-mediated signal transduction. Crit. Rev. Toxicol. 27, 109–134.
Santostefano, M. J., Ross, D. G., Savas, U., Jefcoate, C. R., and Birnbaum, L. S. (1997). Differential time-course and dose-response relationships of TCDD- induced CYP1B1, CYP1A1, and CYP1A2 proteins in rats. Biochem. Biophys. Res. Commun. 233, 20–24.
Schrenk, D., Lipp, H. P., Wiesmüller, T., Hagenmaier, H., and Bock, K. W. (1991). Assessment of biological activities of mixtures of polychlorinated dibenzo-p-dioxins: Comparison between defined mixtures and their constituents. Arch. Toxicol. 65, 114–118.
Schrenk, D., Stüven, T., Gohl, G., Viebahn, R., and Bock, K. W. (1995). Induction of CYP1A and glutathione S-transferase activities by 2,3,7,8-tetrachlorodibenzo-p-dioxin in human hepatocyte cultures. Carcinogenesis 16, 943–946.
Staskal, D. F., Diliberto, J. J., DeVito, M. J., and Birnbaum, L. S. (2005). Inhibition of Human and Rat CYP1A2 by TCDD and Dioxin-like Chemicals. Toxicol. Sci. 84, 225–231.
Tritscher, A. M., Goldstein, J. A., Portier, C. J., McCoy, Z., Clark, G. C., and Lucier, G. W. (1992). Dose-response relationships for chronic exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin in a rat tumor promotion model: Quantification and immunolocalization of CYP1A1 and CYP1A2 in the liver. Cancer Res. 52, 3436–3442.
Vakharia, D. D., Liu, N., Pause, R., Fasco, M., Bessette, E., Zhang, Q. Y., and Kaminsky, L. S. (2001). Polycyclic aromatic hydrocarbon/metal mixtures: Effect on PAH induction of CYP1A1 in human HEPG2 cells. Drug Metab. Dispos. 29, 999–1006.
Vamvakas, A., Keller, J., and Dufresne, M. (1996). In vitro induction of CYP 1A1-associated activities in human and rodent cell lines by commercial and tissue-extracted halogenated aromatic hydrocarbons. Environ. Toxicol. Chem. 15, 814–823.
van den Berg, M., Birnbaum, L., Bosveld, A. T. C., Brunstrm, B., Cook, P., Feeley, M., Giesy, J. P., Hanberg, A., Hasegawa, R., Kennedy, S. W., et al. (1998). Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ. Health Perspect. 106, 775–792.
Vanden Heuvel, J. P., Clark, G. C., Kohn, M. C., Tritscher, A. M., Greenlee, W. F., Lucier, G. W., and Bell, D. A. (1994). Dioxin-responsive genes: Examination of dose-response relationships using quantitative reverse transcriptase-polymerase chain reaction. Cancer Res. 54, 62–68.
Walker, N. J., Portier, C. J., Lax, S. F., Crofts, F. G., Li, Y., Lucier, G. W., and Sutter, T. R. (1999). Characterization of the dose-response of CYP1B1, CYP1A1, and CYP1A2 in the liver of female Sprague-Dawley rats following chronic exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 154, 279–286.
Wiebel, F. J., Wegenke, M., and Kiefer, F. (1996). Bioassay for determining 2,3,7,8-tetrachlorodibenzo-p-dioxin equivalents (TEs) in human hepatoma HepG2 cells. Toxicol. Lett. 88, 335–338.
Xu, L., Li, A. P., Kaminski, D. L., and Ruh, M. F. (2000). 2,3,7,8 tetrachlorodibenzo-p-dioxin induction of cytochrome P4501A in cultured rat and human hepatocytes. Chem. Biol. Interact. 124, 173–189.
Zeiger, M., Haag, R., Hockel, J., Schrenk, D., and Schmitz, H.-J. (2001). Inducing effects of dioxin-like polychlorinated biphenyls on CYP1A in the human hepatoblastoma cell line HepG2, the rat hepatoma cell line H4IIE, and rat primary hepatocytes: Comparison of relative potencies. Toxicol. Sci. 63, 65–73.(Jay B. Silkworth, Aruna K)