Skip to content

Environmental Health Perspectives

Facebook Page EHP Twitter Feed Open Access icon  

Research July 2013 | Volume 121 | Issue 7

Email this to someoneShare on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Share on StumbleUpon
Environ Health Perspect; DOI:Environ Health Perspect; DOI:10.1289/ehp.1206336

Comparison of Intake and Systemic Relative Effect Potencies of Dioxin-like Compounds in Female Mice after a Single Oral Dose

Karin I. van Ede,1 Patrik L. Andersson,2 Konrad P.J. Gaisch,1 Martin van den Berg,1 and Majorie B.M. van Duursen1
Author Affiliations open
1Institute for Risk Assessment Sciences, Utrecht University, Utrecht, the Netherlands; 2Department of Chemistry, Umeå University, Umeå, Sweden

PDF icon PDF Version (442 KB)

  • Background: Risk assessment for mixtures of polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and polychlorinated biphenyls (PCBs) is performed using the toxic equivalency factor (TEF) approach. These TEF values are derived mainly from relative effect potencies (REPs) linking an administered dose to an in vivo toxic or biological effect, resulting in “intake” TEFs. At present, there is insufficient data available to conclude that intake TEFs are also applicable for systemic concentrations (e.g., blood and tissues).

    Objective: We compared intake and systemic REPs of 1,2,3,7,8-pentachlorodibenzodioxin (PeCDD), 2,3,4,7,8-pentachlorodibenzofuran (4-PeCDF), 3,3´,4,4´,5-pentachlorobiphenyl (PCB-126), 2,3´,4,4´,5-pentachlorobiphenyl (PCB-118), and 2,3,3´,4,4´,5-hexachlorobiphenyl (PCB-156) in female C57BL/6 mice 3 days after a single oral dose.

    Methods: We calculated intake REPs and systemic REPs based on administered dose and liver, adipose, or plasma concentrations relative to TCDD. Hepatic cytochrome P450 1A1–associated ethoxyresorufin-O-deethylase (EROD) activity and gene expression of Cyp1a1, 1a2 and 1b1 in the liver and peripheral blood lymphocytes (PBLs) were used as biological end points.

    Results: We observed up to one order of magnitude difference between intake REPs and systemic REPs. Two different patterns were discerned. Compared with intake REPs, systemic REPs based on plasma or adipose levels were higher for PeCDD, 4-PeCDF, and PCB-126 but lower for the mono-ortho PCBs 118 and 156.

    Conclusions: Based on these mouse data, the comparison between intake REPs and systemic REPs reveals significant congener-specific differences that warrants the development of systemic TEFs to calculate toxic equivalents (TEQs) in blood and body tissues.

    Key words: dibenzofurans, dioxins, PCBs, PCDDs, PCDFs, polychlorinated biphenyls, systemic REPs, TEF concept.

  • Environ Health Perspect 121:847-853 (2013).

    Address correspondence K. van Ede, Institute for Risk Assessment Sciences (IRAS), Utrecht University, P.O. Box 80177, NL-3508 TD Utrecht, the Netherlands. Telephone: 31 30 253 3570. E-mail:

    This work was financially supported by the European Union Seventh Framework Project SYSTEQ under grant agreement FP7-ENV-226694. The compounds were purchased by The Dow Chemical Company (Michigan, USA) and donated to the SYSTEQ consortium.

    The purchase of the compounds was financially made possible by The Dow Chemical Company. However, the authors’ freedom to design, conduct, interpret, and publish research was not compromised by The Dow Chemical Company as a condition of review and publication.

    We thank H. Avezaat, S. Versteeg, A. van der Sar, R. Timmermans, S. Nijmeijer, E. Janssen, W. de Wilde, and S. Bergek for excellent technical assistance. We also thank L. Aylward for all the helpful discussions.

    Received: 30 November 2012
    Accepted: 2 May 2013
    Advance Publication: 3 May 2013
    Final Publication: 1 July 2013

  • PDF icon Supplemental Material (1.3 MB) PDF

Polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and polychlorinated biphenyls (PCBs) are persistent and widespread contaminants. Of the 419 possible congeners that exist, 7 PCDDs, 10 PCDFs, and 12 non-ortho and mono-ortho PCBs are classified as having dioxin-like effects. Most, if not all, toxic effects of dioxin-like compounds (DLCs) are mediated through the aryl hydrocarbon receptor (AHR); the toxic effects of these DLCs include endocrine, developmental, immune, and carcinogenic effects, among others (Birnbaum 1994; Birnbaum and Tuomisto 2000; Safe 1990; White and Birnbaum 2009). Humans are exposed to a complex mixture of these DLCs mainly through the diet, with food of animal origin being the most important source. Although exposure has significantly decreased during the past decades (De Mul et al. 2008; Fürst 2006), current human exposure is still above the tolerable daily intake (TDI) or reference dose (RfD) levels for parts of the population in some countries (Bilau et al. 2008; De Mul et al. 2008; Llobet et al. 2008; Loutfy et al. 2006; Tard et al. 2007). Therefore, improving the risk assessment process for this class of compounds remains important and societally relevant.

Currently, risk assessment of DLCs is based on the toxic equivalency factor (TEF) approach (Safe 1990, 1994) endorsed by the World Health Organization (WHO) (Van den Berg et al. 1998, 2006). Each congener-specific TEF is derived from multiple relative effect potencies (REPs) determined from a range of AhR-specific end points [e.g., cytochrome P450 1A1 (CYP1A1) activity]. The toxic or biological potency of a congener is compared to that of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). A shortcoming of the TEF concept originates from the fact that the TEFs were established primarily from in vivo end points linking administered dose levels (via oral exposure) to toxic or biological effects, resulting in “intake” TEFs (intakeTEFs). Consequently, these intakeTEFs are applicable only for situations in which ingestion (e.g., food intake, consumption of breast milk) is known. However, because ingestion data for humans is often lacking or difficult to establish, blood or adipose tissue levels are frequently used to quantify the relative exposure to humans. Subsequently, regulatory authorities commonly calculate risks based on blood or adipose tissue (systemic) levels using these intakeTEFs. Unfortunately, even for the most relevant DLCs, experimental validation is in sufficient to either reject or accept this application of intakeTEFs for blood or tissue levels. There is limited evidence suggesting that the use of intakeTEFs instead of systemicTEFs may lead to inaccurate interpretation of the risk because of congener-specific toxicokinetic differences (Chen et al. 2001; DeVito et al. 1998; Hamm et al. 2003). Properties such as absorption, distribution, metabolism, and excretion can clearly contribute to the potency of a congener (Budinsky et al. 2006; DeVito and Birnbaum 1995; DeVito et al. 1997, 2000) and may be misinterpreted when relying solely on intakeTEFs. At the most recent WHO expert meeting (in 2005) where the TEFs were (re)evaluated, it was concluded that insufficient data were available to develop systemicTEFs, leaving a major gap in the risk assessment process for DLCs (Van den Berg et al. 2006). To fill this data gap, the European Union (EU) project SYSTEQ was initiated, with the main objectives of establishing in vivo systemicREPs in the mouse and rat, with special focus on effects in peripheral blood lymphocytes (PBLs) as potential biomarkers of exposure.

In the present study we compared intakeREPs and systemicREPs in female C57BL/6 mice based on the administered dose and liver, adipose, or plasma concentrations. We used 2,3,7,8-tetrachlorodibenzodioxin (TCDD), 1,2,3,7,8-pentachlorodibenzodioxin (PeCDD), 2,3,4,7,8-pentachlorodibenzofuran (4-PeCDF), 3,3´,4,4´,5-pentachlorobiphenyl (PCB-126), 2,3´,4,4´,5-pentachlorobiphenyl (PCB-118), and 2,3,3´,4,4´,5-hexachlorobiphenyl (PCB-156), which represent approximately 90% of the dioxin-like activity in the human food chain (Liem et al. 2000); we also included the non-dioxin-like 2,2´,4,4´,5,5´-hexachlorobiphenyl (PCB-153). Three days after exposure, we calculated intakeREPs and systemicREPs for hepatic CYP1A1-associated ethoxyresorufin-O-deethylase (EROD) activity and Cyp1a1, 1a2, and 1b1 gene expression in the mouse liver and PBLs.

Materials and Methods

Chemicals. TCDD, PeCDD, 4-PeCDF, and PCB-126 were purchased from Wellington Laboratories Inc. (Guelph, Ontario, Canada) and dissolved in corn oil (ACH Food Companies Inc., Oakbrook, IL, USA); concentrations were then checked and confirmed by Wellington Laboratories Inc. We purchased PCB-118, PCB-156, and PCB-153 from Cerilliant Corp. (Round Rock, TX, USA). These three PCBs and corn oil (Sigma-Aldrich, Stockholm, Sweden) were purity checked; and PCB-118 and PCB-156 were purified at the Department of Chemistry, Umeå University. Before purification, PCB-118 contained 85 ng toxic equivalents (TEQ)/g and PCB-156 contained 201 ng TEQ/g. The final toxic equivalent (TEQ) contributions of impurities were 6.6 ng TEQ/g (PCB-118), 36 ng TEQ/g (PCB-156), and 0.41 ng TEQ/g (PCB-153), levels we considered to have no influence on the final outcome of our results. PCBs were dissolved in corn oil after purification. All tested congeners were further diluted in corn oil (Sigma-Aldrich) at the Institute for Risk Assessment Sciences, Utrecht University).

Animals. Eight-week-old female C57BL/6 mice (Harlan laboratories, Venray, the Netherlands) were randomly assigned to treatment groups (six animals per group) and allowed to acclimate for 1.5 weeks. The animals were housed in groups in standard cages and conditions (23 ± 2°C, 50–60% relative humidity, 12-hr dark/light cycle) with free access to food and water. Mice received a single dose of test compound by oral gavage at a dosing volume of 10 mL/kg body weight (BW). Mice treated with corn oil vehicle (10 mL/kg BW) served as controls. For each congener, five different doses were administered, ranging from 0.5–100 μg/kg BW (TCDD) to 5,000–500,000 µg/kg BW (PCB-153). Detailed information on doses is provided in Supplemental Material, Table S1 ( On day 3 after dosing, animals were euthanized by CO2/O2 asphyxiation, and blood was immediately collected from the abdominal aorta. The liver, thymus, spleen, and adipose tissue were removed, weighed (liver, thymus, and spleen), snap frozen, and stored at –80°C. All animal treatments were performed with permission of the Animal Ethical Committee (DEC Utrecht) and performed according to the Law on Animal Experiments (1977). Animals were treated humanely and with regard for alleviation of suffering.

Compound analysis. Adipose and liver tissues samples were homogenized in Na2SO4, followed by extraction and clean-up in one step, and then eluted with 200 mL 1:1 hexane:dichloromethane on an open column packed with 40% wt/wt H2SO4-impregnated silica and KOH-silica. Blood plasma samples were extracted on an open column using Chem-Elut (Agilent Technologies, Santa Clara, CA, USA) and then NaCl eluted with 75 mL 3:2 hexane:2-propanol. Clean-up was performed using a miniaturized silica column (as described above), and samples were eluted using 30 mL hexane. Because the samples typically contained high levels of the analytes, only a small fraction was evaporated and analyzed. Prior to evaporation, we spiked a fraction of the samples with 13C-labeled standards. We checked potential loss of analytes during extraction and clean-up by reextracting the samples using the identical protocol used for the samples. This procedure indicated that the losses that occurred during this first step were minor, and thus most likely do not significantly contribute to the measured outcomes. Tetradecane was added prior to evaporation. Sample analysis followed the U.S. Environmental Protection Agency Method 1613 (U.S. Environmental Protection Agency 1994) using single ion monitoring mode on an Agilent 6809N gas chromatograph (Agilent Technologies) coupled with a Micromass Ultima Autospec Ultra high-resolution mass spectrometer (HRMS; Waters Corp., Milford, MA, USA). Compounds were separated on a 60 m × 0.25 mm DB5-MS column (0.25 μM; J&W Scientific, Folsom, CA, USA). The HRMS was operated with electron impact ionization with electron energy of 35 eV and an ion source temperature of 250°C. To reduce the number of analyses, samples were pooled before clean-up. To retain unique individual results, liver, adipose, and plasma samples were not pooled within the same treatment group of one congener, but between similar exposure levels of TCDD, PeCDD, 4-PeCDF, and PCB-126 or PCB-118, PCB-156, and PCB-153. This method was used because full congener–specific separation could be achieved on the high-resolution GC–HRMS. For lipid determination, samples were evaporated to dryness after the extraction step, and the amount of lipids was determined gravimetrically. Concentrations were calculated based on lipid weight and wet weight. The analysis of samples for the PCB-118 5,000 μg/kg BW dose failed during the procedure; thus analysis for this group could not be completed.

Plasma and PBL isolation. Blood from two mice was pooled (total volume of approximately 1.4 mL); plasma and PBLs were then isolated using Ficoll Paque gradient (GE Healthcare Europe, Diegem, Belgium). Plasma samples were stored at –80°C until compound analysis. Isolated lymphocytes were lysed with RLT buffer (QIAGEN, Venlo, the Netherlands) as described in the QIAGEN RNAeasy kit protocol and stored at –80°C until use.

EROD activity. We determined hepatic CYP1A1 activity using ethoxyresorufin-O-deethylase (EROD) activity in hepatic microsomal fractions as described by Schulz et al. (2012).

RNA isolation and quantitative real-time polymerase chain reaction (PCR). Total RNA was isolated from liver and PBLs using a QIAGEN RNeasy kit (QIAGEN). Purity and concentration of the isolated RNA was determined by measuring the absorbance ratio at 260/280 nm and 230/260 nm with a Nanodrop 2000 spectrophotometer (Thermo Scientific, Asheville, NC, USA). RNA was reverse transcribed to complementary DNA (cDNA) using the iScript cDNA synthesis Kit (Bio-Rad, Veenendaal, the Netherlands). Quantitative real-time PCR analyses were performed using the iQ Real-Time PCR Detection System with SYBR green (Bio-Rad). Amplification reactions were set up with 15 μL mastermix containing 12.5 μL iQ SYBR Green Supermix (Bio-rad), 0.5 μL distilled H2O, 1 μL (10 μM) forward primer, 1 μL (10 μM) reverse primer, and 10 μL first strand cDNA (10X diluted). Primer sequences were as follows: Cyp1a1: forward-5´-GGTT​AACC​ATGA​CCGG​GAAC​T-3´ and reverse-5´-TGCC​CAAA​CCAA​AGAG​AGTG​A-3´ (Schulz et al. 2012); Cyp1a2: forward-5´-ACATT​CCCA​AGGA​GCGC​TGTA​TCT​-3´and reverse-5´-GTCG​ATGG​CCGA​GTTG​TTAT​TGGT​-3´(Flaveny et al. 2010); Cyp1b1: forward-5´-GTGG​CTGC​TCAT​CCTC​TTTA​CC-3​´and reverse-5´-CCCA​CAAC​CTGG​TCCA​ACTC-3´ (Berge et al. 2004); β-actin: forward-5´-ATGC​TCCC​CGGG​CTGT​AT-3´ and reverse-5´-CATA​GGAG​TCCT​TCTG​ACCC​ATTC​-3´(Schulz et al. 2012). All primers were run through the National Center for Biotechnology Information Primer-BLAST database (​-blast/) to confirm specificity and validate for optimal annealing temperature (60°C for all primers) and efficiency. The efficiency of all primer pairs was 98–102% (tested at 60°C). The following program was used for denaturation and amplification of cDNA: 3 min at 95°C, followed by 40 cycles of 15 sec at 95°C and 45 sec at 60°C. Gene expression for each sample was expressed as threshold cycle (Ct), normalized to the reference gene β-actinCt). We calculated fold induction relative to the control group.

Data analysis. We obtained concentration–response curves using a sigmoidal dose–response nonlinear regression curve fit with variable slope (GraphPad Prism 6.01; GraphPad Software Inc., San Diego, CA, USA):

y equals E sub 0, plus open left-parenthesis fraction open left-parenthesis E sub maximum times X super n, right-parenthesis, over open left-parenthesis b super n plus X super n, right-parenthesis, right-parenthesis [1]


In this Hill equation, y is the dependent variable (EROD activity or fold induction of mRNA levels), x the independent variable (administered or systemic dose), E0 is the estimated background response level, Emax is the maximum response, b is the estimated median effective concentration (EC50), and n is the shaping parameter of the Hill curve.

We calculated the potency of a congener relative to TCDD using the dose or concentration [benchmark response (BMR)] needed for a congener to reach 20% of the TCDD response (BMR20TCDD). Using the congener-specific BMR20TCDD concentration, REPs were calculated relatively to TCDD:

REP congener X equals fraction BMR sub 20 TCDD of TCDD over BMR sub 20 TCDD of congener X[2]



Statistical analysis. Statistical significant differences of the means and variances were determined using one-way analysis of variance (ANOVA) followed by Tukey-Kramer multiple comparisons. Differences were considered statistically significant at p < 0.05. Statistical calculations were performed using GraphPad 6.01 (GraphPad Software Inc.).


Effect on body and organ weight. To evaluate the possible toxic effects of the congeners tested, we examined body and organ weights. We observed no changes in body weight in congener-treated mice compared with vehicle controls. Relative thymus weights showed a decreasing trend for all compounds except PCB-126; however, this decrease was statistically significantly different from the vehicle controls only in mice treated with TCDD (≥ 2.5 μg/kg BW), PeCDD (0.5, 10, and 100 μg/kg BW), and PCB-153 (500,000 µg/kg BW). We also observed a dose-dependent increasing trend in liver weight for all compounds, but this increase was significantly different from vehicle controls only at doses of ≥ 10 µg/kg BW (TCDD), ≥ 100 µg/kg BW (PeCDD), ≥ 100 µg/kg BW (4-PeCDF), ≥ 1,000 µg/kg BW (PCB-126), ≥ 150,000 µg/kg BW (PCB-118), ≥ 50,000 µg/kg BW (PCB-156), and ≥ 500,000 μg/kg BW (PCB-153). In addition, we observed a dose-dependent increasing trend in hepatic lipid content of mice treated with all compounds except PCB-153, compared with vehicle controls. No statistically significant changes in spleen weight were observed for any of the compounds tested. Additional information is provided in Supplemental Material, Table S2 (

Distribution of the compounds. To calculate systemicREPs, we analyzed liver, adipose, and plasma concentrations of the test compounds [see Supplemental Material, Table S3 (]. Within the 3-day period between dosing and sacrifice, concentrations of all congeners increased linearly with the administered dose (Figure 1), which indicates an absence of autoinduction of metabolism for the different dose levels within this time period.

Figure 1. Line graph of the relation between oral dose in µg/kg bw (x-axis) and systemic concentration in ng/g tissue (y-axis) in mouse liver (black line) and adipose tissue (black dotted line). Relation between oral dose and systemic concentration are shown for TCDD (black circle), PeCDD (white square), 4-PeCDF (black triangle), PCB-126 (grey triangle upside down), PCB-118 (black square), PCB-156 (white circle) and PCB-153 (grey diamond). Details in figure legend and Results under “Distribution of the compounds”.Figure 1 – Relationship between oral dose and mean systemic concentration in liver or adipose tissue of female C57BL/6 mice 3 days after administration of a single dose of TCDD, PeCDD, 4-PeCDF, PCB‑126, PCB‑118, PCB‑156, or PCB‑153. Data represent mean ± SD of 6 mice.

View larger image (TIF File)

On a wet weight basis (nanograms per gram of tissue), concentrations of TCDD, PeCDD, 4-PeCDF, and PCB-126 were higher in the liver than in adipose tissue [see Supplemental Material, Table S3 (]. In contrast, concentrations of the mono-ortho PCBs 156 and 118, and the non-dioxin-like PCB-153 were lower in liver than in adipose tissue. These differences were even more pronounced when concentrations were expressed as percent of dose per gram of tissue. Thus, the more potent DLCs had a higher liver affinity than the less potent PCBs 118 and 156. Therefore, we determined the ratio between liver and adipose tissue concentrations to study congener-specific hepatic sequestration. Diliberto et al. (1997) previously suggested that a liver:adipose ratio > 0.3 reflects congener-specific hepatic sequestration. In our study, we observed liver:adipose ratios > 0.3 for TCDD, PeCDD, 4-PeCDF, and PCB-126 but liver:adipose ratios < 0.3 for PCBs 118, 156, and 153 (Table 1). Hepatic sequestration was dose dependent for TCDD and PCB-126 (as shown by increasing liver:adipose ratios at higher dose levels) but not for PeCDD and 4-PeCDF.

View HTML fileTable 1 – Liver:adipose concentration ratios.

View Table (HTML Version)
View larger image (TIF File)

Dose–response curves. We used tissue and plasma concentrations to determine dose–response relationships of hepatic EROD activity and gene expression of Cyp1a1, 1b1, and 1a2 in liver and PBLs [see Supplemental Material, Figure S1 (]. All compounds except PCB-153 caused a statistically significant, dose-dependent increase in hepatic EROD activity and in Cyp1a1 and 1a2 mRNA levels. Hepatic Cyp1b1 mRNA expression was dose-dependently increased by TCDD, PCDD, 4-PeCDF, and PCB-156. We observed a dose-dependent trend for PCB-118; however, the maximum induction for PCB-118 was < 0.3% of the maximal response of TCDD. PCB-126 did not induce Cyp1b1 mRNA levels in the liver. In PBLs, Cyp1a1 mRNA levels were dose-dependently induced by all compounds except PCB-118 and PCB-153. Cyp1b1 mRNA was statically significantly and dose-dependently induced by TCDD, PeCDD, and 4-PeCDF. PCB-126 induced Cyp1b1 mRNA only at the highest dose tested, with 3.5% of the maximal induction of TCDD. PCB-118, PCB-156, and PCB-153 did not induce Cyp1b1 mRNA levels in PBLs, and Cyp1a2 mRNA was not expressed in PBLs.

For all DLCs, a maximum induction (Ymax) was reached only for hepatic EROD activity but not for Cyp1a1, 1b1, or 1a2 mRNA in the liver and PBLs, even at the highest doses tested. Furthermore, we observed differences in curve Hill slopes between congeners for all end points tested [see Supplemental Material, Figure S1 (]. Dose–response curves of Cyp1a1 mRNA in liver and PBLs based on administered dose or on liver or plasma concentration are provided in Supplemental Material, Figure S2. Congener-specific differences in Ymax and Hill slopes can add a significant uncertainty in calculating EC50 values that generally form the basis of REP determination. To reduce this uncertainty, we focused on the lower part of the dose–response curves (BMR20TCDD) as a comparative end point (see Supplemental Material, Figures S1 and S2).

See HTML fileTable 2 – Mean BMR20TCDD concentrations for TCDD, PeCDD, 4-PeCDF, PCB-126, PCB-118, and PCB-156 and corresponding REPs for various end points in liver and PBLs.

View Table (HTML version)
View larger image (TIF File)

BMR20TCDD concentrations and REPs. BMR20TCDD values for hepatic end points were calculated based on administered dose and on hepatic, adipose, or plasma concentration, whereas BMR20TCDD for PBL end points were calculated using only the administered dose or plasma concentration. The administered dose or systemic levels needed for a congener to reach the BMR20TCDD varied strongly between end points, but also between the liver and PBLs (Table 2). Compared with liver, a higher concentration was usually needed in PBLs to reach a BMR20TCDD for the same end point. In the liver, EROD activity was the most sensitive biomarker for TCDD, PeCDD, 4-PeCDF, and PCB-126 exposure, followed by Cyp1a1 and Cyp1a2 mRNA induction. In contrast, hepatic Cyp1a2 mRNA induction appeared to be the most sensitive biomarker for PCB-118 and PCB-156, followed by EROD activity and Cyp1a1 gene expression. In PBLs in the TCDD group, the BMR20TCDD for Cyp1a1 and Cyp1b1 were similar. In contrast, for PeCDD and 4-PeCDF, the BMR20TCDD of Cyp1b1 expression was at least twice that of Cyp1a1 gene expression. In Figure 2, we present an overview of the REP differences based on liver, adipose, and plasma concentrations. A BMR20TCDD was not reached for all congeners or end points studied; thus, these data were excluded from the REP calculations.

Figure 2. Bar graphs of fold change (x-axis) in <sup>systemic</sup>REP compared to <sup>intake</sup>REP for PeCDD, 4-PeCDF, PCB-126, PCB-118 and PCB-156. Bars (y-axis) represents different biological endpoints determined. Hepatic EROD activity (black), <italic>Cyp1a1</italic> (dark grey), <italic>Cyp1b1</italic> (light grey) and <italic>Cyp1a2</italic> (<italic>white</italic>) gene expression in liver and PBLs. Details in figure legend and Results under “BMR<sub>20TCDD</sub> concentrations and Relative Effect Potencies (REPs)”.Figure 2 – Fold change in systemicREP compared with intakeREP (set to 1) for PeCDD (A), 4-PeCDF (B), PCB-126 (C), PCB-118 (D), and PCB-156 (E). Changes in REPs were calculated for hepatic EROD activity, and Cyp1a1, Cyp1b1, and Cyp1a2 gene expression in liver and PBLs. Abbreviations: ND, not determined; Syst, systemic.

View larger image (TIF File)

For comparison of congener-specific REPs across exposure matrices (intake, liver, adipose, or plasma), the intakeREP was set to 1 and deviations were calculated for various systemicREPs with the same end point (Figure 2). We observed two different types of deviations between systemicREPs and intakeREPs. Based on liver concentrations (wet weight or lipid weight), systemicREPs of PeCDD, 4-PCDF, and PCB-126 were at most one-third of the intakeREPs. In contrast, systemicREPs of PCB-118 and PCB-156 are up to one order of magnitude higher than their intakeREPs. When systemicREPs for hepatic effects of PeCDD, 4-PeCDF, and PCB126 were calculated using adipose tissue and plasma concentrations, systemicREPs were up to one order of magnitude higher than intakeREPs, depending on the end point studied. We found the opposite for the systemicREPs of PCB-118 and PCB-156, which were at most one-third of the intakeREPs. In PBLs, systemicREPs based on plasma concentrations also deviated from intakeREPs, in a manner similar to that of systemicREPs of hepatic end points based on plasma concentration.

These two different types of deviations from intakeREPs that we found for systemicREPs differentiate the more potent AhR agonists (PeCDD, 4-PeCDF, and PCB-126) from the less potent mono-ortho PCBs (PCB-118 and PCB-156). In both groups, systemicREPs can differ as much as one order of magnitude from the intakeREPs (Figure 2).


The TEF approach is the most commonly used method of assessing the risk of complex mixtures of dioxins and DLCs. Current TEF values are derived mainly from a range of intakeREPs, preferably from (sub)​chronic in vivo studies. These intakeREPs link the administered dose to a toxic or biological effect, subsequently leading to the derivation of intakeTEFs (Van den Berg et al. 1998, 2006).

At present, available data are insufficient to establish whether or not intakeTEFs are valid for risk assessment based on plasma or adipose tissue concentrations. Thus far, the limited experimental evidence available suggests that systemicREPs of DLCs may differ from intakeREPs (Budinsky et al. 2006; DeVito et al. 1997, 2000). This discrepancy originates most likely from toxicokinetic differences between various DLCs. Several studies have shown that many DLCs bind strongly to CYP1A2 protein and, as a result, strongly sequester in the rodent liver (DeVito et al. 1998; Diliberto et al. 1995, 1997, 1999). This binding affinity toward CYP1A2 influences the hepatic, plasma, and adipose tissue disposition of DLCs. This was confirmed using CYP1A2 knockout mice in which the liver:adipose ratio decreased to < 0.3 for TCDD and 4-PeCDF, which is indicative of no hepatic sequestration (Diliberto et al. 1997). These ratios are significantly lower than those we observed in the present study for both congeners (Table 1). It is worth nothing that the dose dependency and hepatic sequestration we observed in our single dose, 3-day study are similar for all tested compounds—except for 4-PeCDF at the two highest concentrations tested—to those observed by DeVito et al. (1998) in a multiple dose, subchronic 13-week study of female B6C3F1 mice. In addition, the TCDD EC50 systemic liver concentrations for hepatic EROD activity were similar. Comparable findings can also be expected for the other DLCs tested because metabolism and elimination of these compounds are very similar. In light of the similarities between results of the two studies, we assume that intakeREPs and systemicREPs do not deviate over time, even when they have not reached a steady state. In the present study, intakeREPs and systemicREPs for Cyp1a1, 1a2, and 1b1 induction were determined 3 days after a single oral dose. Previous studies have shown that hepatic CYP1A1, 1A2, and 1B1 protein levels are already maximal in rats 3 days after a single dose of TCDD (Santostefano et al. 1997). Although induction of CYP1A1, 1A2, and 1B1 enzymes is not a measure of toxicity, this is considered to be the most sensitive biomarker for AHR activation (Abel and Haarmann-Stemmann 2010; Denison and Heath-Pagliuso 1998). Moreover, studies have shown a high correlation in REPs between induction of these enzymes and toxic responses inflicted by DLCs, such as wasting syndrome, thymic atrophy, or hepatic porphyrin accumulation (Safe 1990; Van Birgelen et al. 1996).

Similar to earlier studies, we observed distinct deviations between intakeREPs and systemicREPs based on liver, plasma, or adipose tissue concentrations (Budinsky et al. 2006; DeVito and Birnbaum 1995; DeVito et al. 1997, 2000). We observed congener-specific differences between the potent PeCDD, 4-PeCDF, and PCB-126 versus the less potent mono-ortho PCBs, PCB-118 and PCB-156 (Figure 2). On the basis of the liver:adipose ratios established in our study (Table 1), it appears that these congener-specific differences have a toxicokinetic basis, in which hepatic sequestration due to CYP1A2 binding plays a significant role. It is unclear whether a CYP1A2-sequestered compound is bioavailable to activate the AhR and cause dioxin-like responses. For this reason, REPs calculated on total hepatic tissue concentration, instead of the “free” available concentrations, may lead to either an over- or underestimation of the potency of a congener, depending on the relative degree of hepatic sequestration compared with that of TCDD. The systemicREPs based on plasma concentrations for Cyp1a1 and 1b1 gene expression in PBLs and liver show similar deviations from intakeREPs for all DLCs tested. The systemicREPs are sometimes more than half a log unit different from the intakeREPs, which is more than the assumed uncertainty range applied to the WHO-TEF values (Van den Berg et al. 2006). To further address this issue, we compared intakeREPs and systemicREPs from the present study with existing WHO-TEF values and the half log uncertainty around that value (Figure 3). On the basis of this comparison, we observed that

Figure 3. Two graphs showing Relative Effect Potencies (y-axis) in relation to the WHO-TEF  ± half log uncertainty range.  REPs were determined for hepatic EROD activity, hepatic gene expression of <italic>Cyp1a1</italic>, <italic>Cyp1b1</italic>, <italic>Cyp1a2</italic>, and gene expression of <italic>Cyp1a1</italic> and <italic>Cyp1b1</italic> in peripheral blood lymphocytes of PeCDD, 4-PeCDF, PCB-126 (left graph) and PCB-118 and PCB-156 (right graph). REPs were calculated based on administered dose (Intake), lipid-based liver concentration (Liver) or lipid-based plasma concentration (Plasma) displayed on the x-axis. Details in figure legend and in Discussion.Figure 3 – REPs determined in the present study in relation to the WHO-TEF ± half log uncertainty range. REPs were determined for EROD activity, and gene expression of Cyp1a1, Cyp1b1, and Cyp1a2 in the liver, and gene expression of Cyp1a1 and Cyp1b1 in PBLs of mice administered a single dose of PeCDD, 4-PeCDF, or PCB‑126 (A) or of PCB‑118 and PCB‑156 (B). REPs for hepatic end points were calculated based on administered dose (Intake), lipid-based liver concentration (Liver), or lipid-based plasma concentration (Plasma); for PBLs, REPs were calculated using the administered dose or plasma concentration. Black lines represent the mean of the REPs; the dotted line and the blue shaded area represent the WHO-TEF ± half log uncertainty range.

View larger image (TIF File)

  • REPs of PeCDD fall mostly within the uncertainty range of the WHO-TEF of 1, with no large difference between systemicREPs and intakeREPs.
  • Based on the intake dose and hepatic concentrations, deviations from the half log unit uncertainty are observed for 4-PeCDF, but systemicREPs based on plasma concentrations are close to the WHO-TEF of 0.3.
  • For PCB-126, intakeREPs and systemicREPs are up to two orders of magnitude below the WHO-TEF value of 0.1. Of all end points studied, only Cyp1a1 mRNA expression in PBLs falls within the half log unit uncertainty.
  • REPs based on intake dose and plasma concentrations for mono-ortho PCBs 118 and 156 are consistently lower than the WHO-TEFs of 0.00003. In contrast, REPs based on liver effects and concentrations are significantly higher than the WHO-TEFs for both PCBs. However, because of differences in Cyp1a2 sequestration between the mono-ortho PCBs and the reference compound TCDD, caution should be taken not to overinterpret these liver-based systemicREPs.

Most REPs determined in the present study are significantly lower than those established by the WHO (Van den Berg et al. 2006). However, the WHO-TEFs were derived from a range of intakeREPs often involving (semi)chronic studies and different species, whereas our study involves a single-dose exposure with relatively acute effects after 3 days only in mice. In the present study, we did not aim to recalculate or debate the current WHO-TEFs or methodology. However, the current WHO-TEF concept is based on the assumption that intakeREPs represent systemicREPs, but a full data set to reject or accept this assumption is lacking. In our study, we compared intakeREPs with systemicREPs obtained from a mouse model to provide more knowledge about possible deviations between both types of REPs. More data, for example, additional in vivo rat data and human in vitro data from our EU-SYSTEQ project studies, may provide additional information with respect to deviation of the intakeREPs and systemicREPs from our studies with current WHO-TEF values. With these additional data, it can then be discussed whether systemicREPs would better reflect a risk than intake (WHO-)TEFs.


There are significant differences between intakeREPs and systemicREPs for hepatic EROD activity and Cyp1a1, 1a2, and 1b1 gene expression in the liver and PBLs. To avoid flawed calculations due to, for example, congener-specific hepatic sequestration, it may be more appropriate to use blood or adipose tissue as a matrix to calculate systemicREPs. The systemicREPs based on plasma/adipose concentration in our study are sometimes more than half a log unit different from the intakeREPs. This suggests that using intakeREPs or intakeTEFs to calculate TEQs in blood for PeCDD, 4-PeCDF, and PCB-126 result in an underestimation of the risk. In contrast, using intakeREPs or intakeTEFs for the mono-ortho PCBs 118 and 156 to calculate blood TEQs in blood may lead to an overestimation of the risk. Overall, our comparison of intakeREPs and systemicREPs in mice reveals significant congener-specific differences that warrants the development of systemicTEFs to calculate TEQs in blood and body tissues.


Abel J, Haarmann-Stemmann T. 2010. An introduction to the molecular basics of aryl hydrocarbon receptor biology. Biol Chem 391(11):1235–1248.

Berge G, Øvrebø S, Eilertsen E, Haugen A, Mollerup S. 2004. Analysis of resveratrol as a lung cancer chemopreventive agent in A/J mice exposed to benzo[a]pyrene. Br J Cancer 91(7):1380–1383.

Bilau M, Matthys C, Baeyens W, Bruckers L, De Backer G, Den Hond E, et al. 2008. Dietary exposure to dioxin-like compounds in three age groups: results from the Flemish environment and health study. Chemosphere 70(4):584–592.

Birnbaum LS. 1994. The mechanism of dioxin toxicity: relationship to risk assessment. Environ Health Perspect 102(suppl 9):157–167.

Birnbaum LS, Tuomisto J. 2000. Non-carcinogenic effects of TCDD in animals. Food Addit Contam 17(4):275–288.

Budinsky RA, Paustenbach D, Fontaine D, Landenberger B, Starr TB. 2006. Recommended relative potency factors for 2,3,4,7,8-pentachlorodibenzofuran: the impact of different dose metrics. Toxicol Sci 91(1):275–285.

Chen CY, Hamm JT, Hass JR, Birnbaum LS. 2001. Disposition of polychlorinated dibenzo-p-dioxins, dibenzofurans, and non-ortho polychlorinated biphenyls in pregnant Long Evans rats and the transfer to offspring. Toxicol Appl Pharmacol 173(2):65–88.

De Mul A, Bakker MI, Zeilmaker MJ, Traag WA, van Leeuwen SPJ, Hoogenboom RLAP, et al. 2008. Dietary exposure to dioxins and dioxin-like PCBs in the Netherlands anno 2004. Regul Toxicol Pharmacol 51(3):278–287.

Denison MS, Heath-Pagliuso S. 1998. The Ah receptor: a regulator of the biochemical and toxicological actions of structurally diverse chemicals. Bull Environ Contam Toxicol 61(5):557–568.

DeVito MJ, Birnbaum LS. 1995. The importance of pharmacokinetics in determining the relative potency of 2,3,7,8-tetrachlorodibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzofuran. Toxicol Sci 24(1):145–148.

DeVito MJ, Diliberto JJ, Ross DG, Menache MG, Birnbaum LS. 1997. Dose-response relationships for polyhalogenated dioxins and dibenzofurans following subchronic treatment in mice: I. CYP1A1 and CYP1A2 enzyme activity in liver, lung, and skin. Toxicol Appl Pharmacol 147(2):267–280.

DeVito MJ, Menache MG, Diliberto JJ, Ross DG, Birnbaum LS. 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(3):157–172.

DeVito MJ, Ross DG, Dupuy AE Jr, Ferrario J, McDaniel D, Birnbaum LS. 1998. Dose-response relationships for disposition and hepatic sequestration of polyhalogenated dibenzo-p-dioxins, dibenzofurans, and biphenyls following subchronic treatment in mice. Toxicol Sci 46(2):223–234.

Diliberto JJ, Akubue PI, Luebke RW, Birnbaum LS. 1995. Dose-response relationships of tissue distribution and induction of CYP1A1 and CYP1A2 enzymatic activities following acute exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in mice. Toxicol Appl Pharmacol 130(2):197–208.

Diliberto JJ, Burgin D, Birnbaum LS. 1997. Role of CYP1A2 in hepatic sequestration of dioxin: studies using CYP1A2 knock-out mice. Biochem Biophys Res Commun 236(2):431–433.

Diliberto JJ, Burgin DE, Birnbaum LS. 1999. Effects of CYP1A2 on disposition of 2,3,7,8-tetrachlorodibenzo-p-dioxin, 2,3,4,7,8-pentachlorodibenzofuran, and 2,2´,4,4´,5,5´-hexachlorobiphenyl in CYP1A2 knockout and parental (C57BL/6N and 129/Sv) strains of mice. Toxicol Appl Pharmacol 159(1):52–64.

Flaveny CA, Murray IA, Perdew GH. 2010. Differential gene regulation by the human and mouse aryl hydrocarbon receptor. Toxicol Sci 114(2):217–225.

Fürst P. 2006. Dioxins, polychlorinated biphenyls and other organohalogen compounds in human milk. Mol Nutr Food Res 50(10):922–933.

Hamm JT, Chen CY, Birnbaum LS. 2003. A mixture of dioxins, furans, and non-ortho PCBs based upon consensus toxic equivalency factors produces dioxin-like reproductive effects. Toxicol Sci 74(1):182–191.

Law on Animal Experiments. 1977. Law of January 12, 1977, Laying Down Rules on the Conduct of Tests on Animals [in Dutch]. Available: [accessed 7 June 2013].

Liem AKD, Fürst P, Rappe C. 2000. Exposure of populations to dioxins and related compounds. Food Addit Contam 17(4):241–259.

Llobet JM, Martí-Cid R, Castell V, Domingo JL. 2008. Significant decreasing trend in human dietary exposure to PCDD/PCDFs and PCBs in Catalonia, Spain. Toxicol Lett 178(2):117–126.

Loutfy N, Fuerhacker M, Tundo P, Raccanelli S, El Dien AG, Ahmed MT. 2006. Dietary intake of dioxins and dioxin-like PCBs, due to the consumption of dairy products, fish/seafood and meat from Ismailia City, Egypt. Sci Total Environ 370(1):1–8.

Safe S. 1990. Polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs), and related compounds: environmental and mechanistic considerations which support the development of toxic equivalency factors (TEFs). Crit Rev Toxicol 21(1):51–88.

Safe SH. 1994. Polychlorinated biphenyls (PCBs): environmental impact, biochemical and toxic responses, and implications for risk assessment. Crit Rev Toxicol 24(2):87–149.

Santostefano MJ, Ross DG, Savas U, Jefcoate CR, Birnbaum LS. 1997. Differential time-course and dose-response relationships of TCDD-induced CYP1B1, CYP1A1, and CYP1A2 proteins in rats. Biochem Biophys Res Commun 233(1):20–24.

Schulz VJ, Smit JJ, Huijgen V, Bol-Schoenmakers M, Van Roest M, Kruijssen LJW, et al. 2012. Non-dioxin-like AhR ligands in a mouse peanut allergy model. Toxicol Sci 128(1):92–102.

Tard A, Gallotti S, Leblanc JC, Volatier JL. 2007. Dioxins, furans and dioxin-like PCBs: occurrence in food and dietary intake in France. Food Addit Contam 24(9):1007–1017.

U.S. Environmental Protection Agency. 1994. Method 1613: Tetra- through Octa-Chlorinated Dioxins and Furans by Isotope Dilution HRGC/HRMS. Available:​/organics/dioxins/upload/2007_07_10_meth​ods_method_dioxins_1613.pdf [accessed 7 June 2013].

van Birgelen AP, DeVito MJ, Akins JM, Ross DG, Diliberto JJ, Birnbaum LS. 1996. Relative potencies of polychlorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls derived from hepatic porphyrin accumulation in mice. Toxicol Appl Pharmacol 138(1)98–109; doi:10.1006/taap.1996.0103.

Van den Berg M, Birnbaum L, Bosveld AT, Brunström B, Cook P, Feeley M, et al. 1998. Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ Health Perspect 106:775–792.

Van den Berg M, Birnbaum LS, Denison M, DeVito M, Farland W, Feeley M, et al. 2006. The 2005 World Health Organization reevaluation of human and mammalian toxic equivalency factors for dioxins and dioxin-like compounds. Toxicol Sci 93(2):223–241.

White SS, Birnbaum LS. 2009. An overview of the effects of dioxins and dioxin-like compounds on vertebrates, as documented in human and ecological epidemiology. J Environ Sci Health C Environ Carcinog Ecotoxicol Rev 27(4):197–211.

WP-Backgrounds Lite by InoPlugs Web Design and Juwelier Schönmann 1010 Wien