Environmental Toxicology and Chemistry, Vol. 31, No. 5, pp. 1149–1157, 2012
#
2012 SETAC
Printed in the USA
DOI: 10.1002/etc.1776
M
ARIA
L
ARSSON
,* D
AN
O
RBE
, and M
AGNUS
E
NGWALL
( Submitted 19 August 2011; Returned for Revision 15 September 2011; Accepted 23 December 2011)
Abstract
—
In the present study, relative potency factors (REPs) of 16 individual polycyclic aromatic hydrocarbons (PAHs) were investigated using the H4IIE-luc bioassay. Exposure time-dependent effects on the REPs were examined using 24, 48, and 72 h of exposure. Seven different mixtures of PAHs were tested for additivity at an exposure time of 24 h. Three of the PAH mixtures were also studied at 48 and 72 h of exposure. The mixture toxicities were predicted using the REP concept and the concentration addition (CA) model. Relative potency factor values investigated in the present study were similar to those reported in earlier studies. Declining REPs with an increasing exposure time were shown for all PAHs, indicating that this bioassay approach could be developed to assess the persistency of aryl hydrocarbon receptor (AhR) agonistic PAHs and in the risk assessment of complex PAH mixtures. The results from the mixture studies indicated that additive interactions of PAHs are time dependent. Generally, 48- and 72-h exposures resulted in biological effects that were similar to the CA and REP model predictions, while these models tended to underestimate the effect, to some extent, in the 24-h exposure, at least for the mixtures containing two to four PAHs. Thus, it cannot be ruled out that in the 24-h exposures, the tested PAH mixtures had slight synergistic effects. Further research is needed to identify and test additional AhR activating PAHs and investigate whether the effects in the H4IIE-luc bioassay are additive for more complex samples containing both PAHs and other
AhR-activating contaminants. Also, the observed superinduction of luciferase by PAH-mixes warrants studies of whether this also can occur for relevant AhR-mediated endpoints in vivo. Environ. Toxicol. Chem. 2012;31:1149–1157.
#
2012 SETAC
Keywords —PAHs AhR agonists Relative potencies Mixture toxicity Concentration-addition
INTRODUCTION
Polycyclic aromatic hydrocarbons (PAHs) form a large group of widespread pollutants of high environmental concern.
They are found in elevated levels in the environment, mainly as a consequence of human activity. Many of the PAHs are toxic, and exposure can result in mutagenesis and carcinogenesis in humans and animals [1,2]. Risk assessments of PAHs are complicated because these compounds mostly exist in the environment as complex mixtures. Generally, the composition of the mixtures is rarely known, and the mechanisms of toxicity and interactions between the chemicals are not completely understood [3–6].
One strategy to assess the toxicity of a mixture is to use relative potency factors (REPs). The toxic potency of a compound can be quantified as the relative potency compared to a model toxicant such as 2,3,7,8-tetrachlorodibenzop -dioxin
(TCDD), or another potent compound. The toxicity equivalent
(TEQ) of a mixture of toxicants acting through a common mechanism of action is calculated as the concentration of each compound multiplied by its specific REP. The REP approach is useful in studies combining instrumental chemical analysis and biological analysis. With mass balance analysis, that is, comparing bioassay-derived TEQs with TEQ based on REPs, it is possible to estimate the contribution of instrumentally identified and quantified compounds to the observed response in the bioanalysis.
The H4IIE-luc bioassay is a useful tool to study aryl hydrocarbon receptor (AhR)-activating potency of single compounds and mixtures of compounds and to estimate toxicological risks
* To whom correspondence may be addressed
(maria.larsson@oru.se).
Published online 10 February 2012 in Wiley Online Library
(wileyonlinelibrary.com).
1149 of mixtures in environmental samples [7,8]. Certain PAHs act through the AhR, which is a ligand-activated transcription factor that mediates many of the biological effects of these compounds [9,10]. Earlier studies have reported relative potencies for a number of PAHs [8,11–16], but they differ in methods, cell lines, and time of exposure. Relative potency factors are species- and assay-specific, and the use of nonspecific REPs can result in misleading mass balance analysis.
Based on concentration–response curves of individual compounds, the toxicity of a mixture can be predicted using the concentration addition (CA) model. The CA model is based on the assumption that all chemicals in the mixture have a common mechanism of action and act at the same molecular target site.
One chemical is supposed to behave as a dilution of the other, meaning that any compound can be substituted by an equally potent concentration of another compound without altering the overall effect. Unlike the REP concept, concentration–response curves do not have to be parallel. The CA model has been used to predict the toxicities of many mixtures and has shown to be an accurate reference model of mixtures of chemicals known to have similar modes of action [17–20]. Most studies employ the fixed ratio design, that is, the ratio between the individual compounds in the mixtures is constant, and the total concentration of the mixtures is varied. Often, the compounds are mixed in an equivalent-effect concentration, for example, an effective concentration for 50% (EC50). The advantage of such mixtures is that no discrimination of low-potency compounds occurs [18,21,22]. Both the REP approach and the CA model are applications of the concept of dose-addition and are based on additive behavior of the compounds and noninteractions in the mixture.
The objective of the present study was to develop H4IIE-luc assay-specific REP values for 16 individual PAHs that could be used in a mass balance analysis of PAH-contaminated samples.

1150 Environ. Toxicol. Chem.
31, 2012
Mass balance analysis is based on the assumption of additive behavior of compounds; thus, mixtures with a known composition of PAHs were tested. The predicted toxicity of the mixtures was assessed using the CA model and the REP theory.
The predicted concentration–response curves were compared with the observed concentration–response curves of mixtures containing these PAHs, looking for additivity. An important aim of the present study was to study the exposure-time– dependent effects on the REPs and the PAH mixtures. Extended exposure duration can provide answers on the metabolic degradability of PAHs [8,12], which is important from a toxicological point of view and for the risk assessment of environmental mixtures. For these reasons, all individual PAHs and three of the mixtures were tested during additional exposure times, 48 and 72 h, in addition to 24 h. In the past, long-term exposures, such as 48 and 72 h, have only been conducted for a few PAHs.
Polycyclic aromatic hydrocarbons investigated in the present study are of environmental relevance because they are found in elevated levels in contaminated areas [14,23–
25], and many of them are known to be potent AhR agonists.
Studies of individual PAHs have shown that PAHs are metabolized readily in H4IIE-luc cells, but knowledge of PAH mixtures is lacking [8,12,15]. Aryl hydrocarbon receptor-mediated toxicity of real environmental samples has been shown in several studies, but the assumption that PAH mixtures exhibit additive effects through the AhR pathway has not been thoroughly studied. In addition, the possibility to use an in vitro hepatic dioxin bioassay to estimate the degradability of a complex mixture of PAHs and similar compounds has not been studied. If successful, such an approach could prove useful as an alternative to the selective cleanup of complex environmental extracts and thus provide a new tool in the risk assessment of complex mixtures of PAHs and similar compounds.
MATERIALS AND METHODS
Test chemicals
Individual standards of 16 PAHs were obtained for the present study. Fluoranthene (Fla), pyrene (Pyr), benzo[g,h,i]perylene (B[ghi]P), chrysene (Chr), benzo[a]pyrene (B[a]P), indeno[1,2,3-cd]pyrene (I[123cd]P), benzo[j]fluoranthene (B[j]F), benzo[b]fluoranthene (B[b]F), benzo[k]fluoranthene (B[k]F), dibenzo[a,h]anthracene (DB[ah]A), all with a purity of 99%, and naphto[2,3-a]pyrene (NPyr), benz[a]anthracene (B[a]A), both with a purity of 98%, were purchased from Ultra Scientific
Analytical Solutions. Dibenzo[a,c]anthracene (DB[ac]A) with a purity of 97.5%, dibenzo[a,j]anthracene (DB[aj]A) with a purity of 99.8%, perylene (Per) and benzo[e]pyrene (B[e]P), both with a purity of 99%, were purchased from Labor
Dr. Ehrenstorfer–Scha¨rfers. The TCDD standard, with a purity of 99.1%, was purchased from AccuStandard.
Stock solutions of the individual PAHs were made in dimethyl sulfoxide (DMSO) (2.5 mM). Seven different mixtures of PAHs were obtained by mixing various numbers and combinations of the individual PAHs in equal molar concentrations at a 1:1 ratio
(Table 1). Final concentration in all mixture stock solutions was
2.5 mM. The mixture ratio was one-half in the binary mixture, one-third in the ternary mixtures, and so on.
H4IIE-luc assay
The AhR-mediated activities for the single PAHs and the
PAH mixtures were measured using the H4IIE-luc assay, a rat hepatoma cell line stably transfected with a luciferase reporter
M. Larsson et al.
Table 1. List of the polycyclic aromatic hydrocarbons (PAHs) and the seven PAH mixtures tested in the present study a
Compounds Mixtures
Name MW 15-mix 8-mix 4A-mix 4B-mix 3A-mix 3B-mix 2-mix
Pyr
Fla
Chr
B[a]A
B[k]F
B[j]F
B[b]F
202 EC22
202 EC2 EC3
228 EC60 EC71
228 EC46 EC61 EC76
252 EC94
252 EC81
252 EC74
EC83
EC107
B[e]P
B[a]P
252 EC4
252 EC64 EC77 EC91 EC91 EC96
Per 252 EC3 EC6
I[123-cd]P 276 EC70 EC77 EC88
B[ghi]P 276
DB[ac]A 278 EC82
DB[aj]A 278 EC96
DB[ah]A 278 EC93 EC98
NPyr
EC30
302 EC77 EC85 EC92
EC103
EC94
EC5
EC81
EC34
EC104
EC97 a
Polycyclic aromatic hydrocarbons included in each mixture are marked with the effective concentration (EC) level of the PAH in the mixture.
Effective concentration values are calculated from the individual PAH dose–response curves, 24 h of exposure.
MW
¼ molecular weight; Pyr
¼ pyrene; Fla
¼ fluoranthene; Chr
¼ chrysene; B[a]A
¼ benzo[a]anthracene; B[k]F
¼ benzo[k]fluoranthene;
B[j]F
¼ benzo[j]fluoranthene; B[b]F
¼ benzo[b]fluoranthene; B[e]P
¼ benzo[e]pyrene; B[a]P
¼ benzo[a]pyrene; Per
¼ perylene; I[123-cd]P
¼ indeno[1,2,3-cd]pyrene; B[ghi]P
¼ benzo[g,h,i]perylene; DB[ac]A
¼ dibenz
[a,c]anthracene; DB[aj]A
¼ dibenz[a,j]anthracene; DB[ah]A
¼ dibenz[a,h] anthracene; NPyr
¼ naphto[2,3-a]pyrene.
gene [26]. Cells were cultivated in alpha-minimal essential medium (Sigma) supplemented with 10% fetal calf serum
(Labkemi) and incubated at 37
8
C, 5% CO
2
, and 100% humidity.
To prepare for analysis, the cells were seeded into 96-well plates in aliquots of 100 m l (density, 3–6 10
4 cells/well) and incubated in 24 h until 80 to 90% confluence was reached.
Single PAH and PAH-mixture dilutions (10 concentrations) were prepared in culture medium and added to the cells in triplicate wells. The maximum concentration was 10 m M. In each assay, a standard curve of TCDD (0.4–300 pM) and a solvent control (DMSO) were tested in triplicate wells. The final
DMSO concentration in all wells was 0.4%, and the total volume was 200 m l/well.
Analysis of luciferase activity
After 24, 48, or 72 h of exposure, the medium was removed and the cells were washed twice by adding 100 m l phosphate buffered saline solution and lysed in 25 m l phosphate buffered saline solution at 20
8
C in 3 to 24 h. Afterward, 25 m l of Steady
Lite substrate mix (PerkinElmer) was added to the frozen cell lysates and placed in darkness at room temperature for 1 h for enzymatic reaction. The cell lysates (30 m l) were transferred to a white 96-well microtiter plate, and the luciferase activity was measured in a luminometer (Wallac 1420, Victor
2
). Concentration–response curves were constructed for the individual
PAHs and the mixtures using a sigmoidal concentration– response (variable slope) equation. The curve fitting was done with the GraphPad Prism
1
4.0 software.
Relative potency factors
CALCULATIONS
Each PAH experiment was repeated two to four times and individual concentration–response curves were performed for

AhR agonistic potencies and additivity of PAHs each experiment. Relative potency factors for the single PAHs were obtained from the concentration–response curves by relating the luciferase induction potency of the PAH to that of the positive reference TCDD using the equation
REP i
¼ TCDD EC x
= PAH EC x
(1)
The REPs based on EC50 and EC25 were calculated by dividing the ECx for TCDD by the ECx for the PAH compound where x is 25 or 50% of TCDD maximum induction (TMI). A mean value for each relative potency factor was calculated from the independent experiments.
As the concentrations of the compounds in the mixture are known, the TEQ of the mixture can be calculated by their REP values, using the equation
X
TEQ mix
¼ ð conc i
REP i
Þ
(2)
When the compounds in the mixture are mixed in a 1:1 molar ratio and the concentration of the mixture is the same as the sum of the individual concentrations, the equation can be rewritten as
1
= n i
¼
1
REP i
¼
REP mix
(3) where n is the numbers of mixture components. The predicted
REP i in a mixture can thereby be compared with the observed
REP in the mixture, looking for additivity. The observed REP values for the mixtures were calculated using the same equation as for the individual REP calculations.
Predicted toxicity
The mixture toxicity was predicted using the CA model on the basis of the concentration–response curves of the single
PAHs. Because the composition of the mixtures is known, the concentrations of each PAH can be expressed as a fraction of the total concentration. Concentration addition is expressed mathematically as
!
1
EC x ; mix
¼ i
¼
1 p i
EC x
; i
(4) where n is the number of mixture components, EC x ,mix is the effect concentration of the mixture provoking x percent effect, EC x,i is the concentration of the i th mixture component provoking x percent effect when applied singly, and p i is the fraction of the i th component in the mixture [27]. Data from two to four experiments were pooled and reanalyzed before the predicted toxicity calculations were performed. To make the response curves comparable and enable data pooling, the PAH responses were normalized against the mean of the maximum response observed for their corresponding TCDD standard. The mean solvent control response was subtracted from both TCDD standard and PAH responses prior to conversion to get the responses scaled from 0 to 100% of TMI. Concentrations giving
10 to100% mixture effects were calculated in steps of 5%, and the concentration/effect pairs were plotted and analyzed using the sigmoidal concentration–response (variable slope) curve fitting (GraphPad Prism) to give a predicted concentration– response curve.
Observed toxicity
Concentration–response curves were calculated for each
PAH mixture using nonlinear curve regression (GraphPad
Environ. Toxicol. Chem.
31, 2012 1151
Prism) with a sigmoidal curve as a model equation. To make the response curves comparable, the mixture responses were normalized against the mean of the maximum response observed by the TCDD standard. The mean solvent control response was subtracted from both TCDD standard and mixture responses prior to conversion to get responses scaled from 0 to
100% TMI.
RESULTS AND DISCUSSION
Relative potencies of PAHs
Relative potencies of 10 U.S. Environmental Protection Agency priority PAHs and six other PAHs were determined in the
H4IIE-luc bioassay. Three different exposure times (24, 48, and 72 h) were chosen to examine the time-dependent effects on the REPs. At the concentrations tested, 4 of the 16 PAHs tested elicited full agonist AhR-mediated response at all three exposure times (Table 2, Fig. 1).
Based on the 24-h assay, the following order of decreasing
REPs (EC25) was found: DB[ah]A
>
B[k]F
>
B[j]F
>
NPyr
>
I[123cd]P
>
B[b]F
>
DB[ac]A
>
DB[aj]A, with values ranging from 1 10
3 to 6 10
4
. The B[a]P, Chr, B[j]F, and B[a]A reached 100% of TMI in the 24-h assay, but their maximum induction levels declined with increased exposure time, and Chr and B[a]A did not reach 50% of TMI after 72 h of exposure.
Thus, their responses were fairly close to 50% of TMI; REPs based on EC50 were calculated by linear extrapolation of the concentration–response curves. Extrapolation will result in some uncertainty in the REP estimations, however. Benzo[e]pyrene only induced significant responses in the 24-h assay, probably because of a high metabolic rate. The maximum response observed was 60% of TMI. Benzo[g,h,i]perylene induced significant responses at all exposure times, but only reached 50% of TMI in the 24-h assay. Pyrene induced above
50% of TMI in the 24-h assay, but only 40% of TMI in the
48- and 72-h assays. Relative potency factors based on EC50
(48 h and 72 h) were calculated by linear extrapolation of the concentration–response curves. Perylene and Fla elicited weak responses in the H4IIE-luc assay and did not reach 25% of TMI at any exposure time.
Each PAH was tested independently at least two times, and the variation in REP values was generally 1.1- to twofold for most of the individual PAHs at all three exposure durations. The highest variation in the present study was shown for DB[ah]A in the 72-h assay, and the variation was eightfold for REPs based on EC25, and fivefold for REPs based on EC50. Mean REP values for each PAH are listed in Table 2.
Although many of the PAHs had considerable luciferaseinducing potencies, all the PAHs exhibited declining REPs with increased exposure time. Concentration–response curves shifted to the right on the x-axis with increasing exposure time
(Fig. 1) and resulted in reduced REP values. Relative potency factors (EC50) obtained in the 24-h assay were approximately a factor of two to six higher than the REPs (EC50) obtained in the 48-h assay, and the REPs obtained in the 48-h assay were approximately a factor of two to seven higher than REPs obtained from the 72-h assay. Declining REP values with increased duration of exposure are assumed to be a consequence of the metabolism of PAHs in the cells. Polycyclic aromatic hydrocarbons induce the CYP1A1, a member of the xenobiotic metabolizing enzyme system, through the Ah-receptor, an important pathway for the metabolism and removal of these compounds from the cells [9]. Different declining rates of the
REP values indicate varying metabolic rates of individual PAHs

1152 Environ. Toxicol. Chem.
31, 2012 M. Larsson et al.
Table 2. Relative potency factors based on effective concentration (EC25 or EC50) for polycyclic aromatic hydrocarbons (PAHs) derived from 24 h, 48 h and
72 h exposure of H4IIE-luc cells and maximum response levels for the PAHs relative to 2,3,7,8-tetrachlorodibenzop -dioxin
24-h exposure 48-h exposure 72-h exposure
PAH Max % a
EC25 EC50 Max % a
EC25 EC50 Max % a
EC25 EC50
Pyr
Fla
Chr
B[a]A
B[k]F
B[j]F
B[b]F
B[e]P
B[a]P
Per
I[123-cd]P
B[ghi]P
DB[ac]A
DB[aj]A
DB[ah]A
NPyr
71
11
98
99
108
100
91
60
115
20
105
50
94
113
114
99
4.97E-06
No
3.95E-05
1.29E-05
2.32E-03
4.16E-04
5.37E-04
7.05E-07
5.39E-05
No c
4.06E-04
2.62E-06
5.75E-04
5.82E-04
1.45E-03
1.82E-04
2.95E-06
No
2.02E-05
1.25E-05
7.11E-04
1.37E-04
1.44E-04
6.42E-07
3.08E-05
No
1.45E-04
WR d
2.47E-04
2.54E-04
4.72E-04
1.01E-04
44
5
73
62
96
99
102
3
95
11
117
29
91
121
88
117
1.83E-06
No
5.73E-06
2.68E-06
1.56E-04
9.30E-05
6.59E-05
No
8.93E-06
No
6.85E-05
1.06E-06
5.47E-05
1.28E-04
2.37E-04
7.71E-05
1.16E-06 b
No
4.75E-06
2.32E-06
1.12E-04
6.09E-05
4.05E-05
No
8.85E-06
No
3.87E-05
WR
3.71E-05
7.81E-05
1.07E-04
4.43E-05
34
1
48
27
108
75
98
2
70
10
104
21
84
101
101
120
2.44E-06
No
2.93E-06
6.24E-07
4.68E-05
1.49E-05
1.48E-05
No
2.64E-06
No
1.14E-05
4.10E-07
1.60E-05
2.28E-05
6.90E-05
2.09E-05
1.30E-06 b
No
2.30E-06 b
6.00E-07 b
2.78E-05
9.03E-06
1.25E-05
No
2.41E-06
No
9.91E-06
WR
1.06E-05
1.36E-05
4.71E-05
1.60E-05 a c
Max %
¼ mean maximum response observed for the PAH expressed as a percentage of the mean of the maximum response observed for TCDD.
b
Calculated by linear extrapolation.
No
¼ no quantifiable response.
d
WR
¼ weak response
<
50% of TCDD maximum induction.
REPs
¼ relative potency factors; TCDD
¼
2,3,7,8-tetrachlorodibenzop -dioxin; Pyr
¼ pyrene; Fla
¼ fluoranthene; Chr
¼ chrysene; B[a]A
¼ benzo[a]anthracene;
B[k]F
¼ benzo[k]fluoranthene; B[j]F
¼ benzo[j]fluoranthene; B[b]F
¼ benzo[b]fluoranthene; B[e]P
¼ benzo[e]pyrene; B[a]P
¼ benzo[a]pyrene; Per
¼ perylene; I[123cd]P
¼ indeno[1,2,3-cd]pyrene; B[ghi]P
¼ benzo[g,h,i]perylene; DB[ac]A
¼ dibenz[a,c]anthracene; DB[aj]A
¼ dibenz[a,j]anthracene; DB[ah]A
¼ dibenz[a,h]anthracene;
NPyr
¼ naphto[2,3-a]pyrene.
[12]. Generally, REPs based on EC25 declined faster than REPs based on EC50. This was probably because of the complete metabolism of the PAHs in the low-dose wells during longer exposure time. In the higher doses, the metabolism was not complete and an activity still remained after 72 h. This resulted in steeper slopes of the concentration–response curves (Fig. 1), leading to parallelism between the PAH and TCDD curves and similar REP
25 and REP
50 values (Table 2).
It has been suggested that REP ranges (EC20–EC80) should be used instead of fixed effect levels when the REP value is not constant over all concentrations [28]. Nonparallel concentration–response curves will provide variations depending on which effect level is being used for REP calculations. For this reason, it is important to use the same effect level in mass balance calculations. In the present study, we introduce REP values based on EC25 and EC50, commonly used in the literature. We think that because of smaller slopes of the concentration–response curves relative TCDD EC25, and improved parallelism between the PAH and TCDD curves in the range close to EC50, it is more appropriate to use the REP values based on EC50 in mass balance analysis when it is possible, that is, when the maximum induction levels of the samples reach 50% or more of TMI. In the 72-h assay, REPs based on EC25 and EC50 are similar and can be used with similar results in TEQ calculations.
The REP values determined in the present study are comparable to those reported in earlier studies (Table 3). Of the
PAHs tested, B[k]F and DB[ah]A belonged to the three most potent PAHs in all five studies compared. Variation in REP values was generally one order of magnitude or less. If only
REPs of 24 h were compared, the present study’s values varied
1.0- to 4.3-fold compared with those of two other studies [8,12] except for Chr (23-fold) and were 4.8 to 29 times lower than
REPs estimated in a study by Behnisch et al. [11]. Comparison of REPs of 48 h, including the present study and a study by
Masunaga et al. [12], varied 1.1- to 2.8-fold, and REPs of 72 h, including values from three studies, varied with a factor of 1.4 to
28 (Table 3). This variation is reasonable for H4IIE-luc assay
REPs based on data from several studies with differences in methods [29].
Several PAHs elicited superinduction in the H4IIE-luc bioassay, that is, maximum induction levels above the maximum effect induced by TCDD (Fig. 1). At the concentrations tested,
DB[ah]A, B[k]F, DB[aj]A, and B[a]P elicited significant superinduction levels in the 24-h assay. Naphto[2,3-a]pyrene,
I[123cd]P, and DB[aj]A showed significant superinduction levels in the 48-h assay and after 72 h of exposure; only NPyr and B[k]F showed significant superinduction. This phenomenon was observed for B(a)P, B[k]F, Db(ah)A, and I(123-cd)P in a previous study by Behnisch et al. [11], using the H4IIE-luc assay and 24 h of exposure.
Mixture toxicity
PAHs tested were mixed in a 1:1 ratio based on equal molar concentrations, and all mixtures tested had the same total concentration. Each PAH concentration (ECx) in the mixtures corresponded to an effect level, x percent that each compound provoked when applied singly (Table 1). Additive interactions were studied in the mixtures by comparing the observed toxicity in the mixtures with the toxicity predicted using the CA model.
The CA model calculates concentrations of mixtures of chemicals that produce a predetermined effect, and these predictions are based on the effect concentration of each chemical that will singly produce the same effect. Thus, for mathematical reasons, mixture responses higher than the lowest maximal effect by one mixture chemical alone cannot be predicted [30]. Some of the concentration–response curves of the single PAHs did not reach maximum induction, that is, maximum efficacy within the tested concentrations. Linear extrapolation of these concentration–response curves did not affect the results and generated the same predicted mixture curve shapes as without extrapolation. All predicted curves plateaued at effect levels

AhR agonistic potencies and additivity of PAHs
Pyr Fla
Chr
Environ. Toxicol. Chem.
31, 2012
B[a]A
24 h
48 h
72 h
100
1153
100
50
100
50
100
50 50
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
B[k]F B[j]F B[b]F B[e]P
100
50
100
50
100
50
100
50
0
10 -6 10 -4 10 -2 10 0 10 2
0
10 -6 10 -4 10 -2 10 0
0
10 2 10 -6 10 -4 10 -2 10 0 10 2
0
10 -6 10 -4 10 -2 10 0 10 2
B[a]P Per
I[123-cd]pyr B[ghi]P
100
50
100
50
100
50
100
50
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -6 10 -4 10 -2 10 0 10 2
DB[ac]A DB[aj]A DB[ah]A NPyr
100
50
100
50
100
50
100
50
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -6 10 -4 10 -2 10 0 10 2
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
Concentration (
µ
M)
Fig. 1. Concentration–response curves for induction of luciferase activity by 16 polycyclic aromatic hydrocarbons (PAHs) at 24, 48, and 72 h of exposure in the H4IIE-luc bioassay. Data represent the result of two to four independent experiments with three replicates each. Error bars represent standard deviations. Pyr
¼ pyrene; Fla
¼ fluoranthene; Chr
¼ chrysene; B[a]A
¼ benz[a]anthracene; B[k]F
¼ benzo[k]fluoranthene; B[j]F
¼ benzo[j]fluoranthene;
B[b]F
¼ benzo[b]fluoranthene; B[e]P
¼ benzo[e]pyrene; B[a]P
¼ benzo[a]pyrene; Per
¼ perylene; I[123cd]P
¼ indeno[1,2,3-cd]pyrene; B[ghi]P
¼ benzo[g,h,i]perylene;
DB[ac]A
¼ dibenzo[a,c]anthracene; DB[aj]A
¼ dibenzo[a,j]anthracene; DB[ah]A
¼ dibenzo[a,h]anthracene; NPyr
¼ naphto[2,3-a]pyrene.
close to 100%, contrary to the observed mixture curves where superinduction was observed in most mixtures.
The predicted toxicity underestimated the observed toxicity in the mixtures composed of two to four PAHs in the 24-h assay
(Figs. 2 and 3). The EC50 values predicted for those mixtures were approximately a factor of 1.6 to 8 higher than the observed
EC50 values. This demonstrates a nonadditive behavior of the compounds and indicates synergistic effects in the mixtures.
The difference between the observed EC50 and predicted EC50 was largest in the 3B-mix and the 2-mix. In the 4A-mix and 3Amix, the difference was fairly low (factor 1.6 and 1.7, respectively) and the toxicity was almost additive in some range of the curves. No correlation could be seen between the mixture compositions, that is, ECx values and the degree of additivity.
The relative potencies of the mixtures declined by the shift of the concentration–response curves to the right on the x-axis with increasing exposure time (Fig. 3). For the 3A-mix, the observed potency was slightly greater than predicted after 24 and 48 h of exposure. Additivity, however, was demonstrated in the 72-h assay. The predicted toxicity in the mixture containing
15 PAHs (15-mix) was slightly overestimated in the 24-h assay in some range of the curve, but additivity was observed in both the 48- and 72-h assay. The 8-mix containing eight PAHs demonstrated additivity in all exposure times (Fig. 3). Generally, the difference between predicted and observed effect decreased with increasing exposure times and all three mixtures demonstrated additive interactions after 72 h of exposure. The results from the additivity predictions by the REP-concept showed almost the same result as the CA model (Table 4).
Comparison of the predicted REP with the observed REP in the
24-h assay indicated slight antagonistic effects in the 15-mix and 8-mix, and synergistic effects in the other five mixtures accurately as the CA model. The additivity tendency seems to increase with increased duration of exposure for all mixtures except for 8-mix.
The CA model has proven to be a precise reference model for several mixtures of chemicals known to have similar modes of action [17–20], but researchers disagree regarding the use of

1154 Environ. Toxicol. Chem.
31, 2012 M. Larsson et al.
Table 3. Comparison of relative potency factors reported for individual polycyclic aromatic hydrocarbons relative to 2,3,7,8-tetrachlorodibenzop -dioxin a
24-h exposure 48-h exposure 72-h exposure
PAH
Present study
Machala
[8]
Masunaga
[12]
Behnisch
[11]
Present study
Masunaga
[12]
Present study
Villeneuve
[15]
Masunaga
[12]
Pyr
Fla
Chr
B[a]A
B[k]F
B[j]F
B[b]F
B[e]P
B[a]P
Per
I[123-cd]P
B[ghi]P
DB[ac]A
DB[aj]A
DB[ah]A
NPyr
3.0E-06
No
2.0E-05
1.3E-05
7.1E-04
1.4E-04
1.4E-04
6.4E-07
3.1E-05
No
1.5E-04
WR
2.5E-04
2.5E-04
4.7E-04
1.0E-04
1.8E-06
2.3E-08
1.0E-04
7.0E-06
1.6E-03
3.7E-04
3.4E-05
5.2E-07
9.0E-05
NR
3.0E-04
No
1.9E-04
3.7E-04
1.2E-03
2.1E-04
NR
NR
4.7E-04
1.2E-05
4.2E-04
NR
1.2E-04
NR
1.6E-05
NR
1.2E-04
No
NR
NR
1.6E-04
NR
1.4E-05
1.7E-04
4.1E-03
4.2E-03
5.8E-04
4.1E-03
4.9E-03
1.2E-06
No b
4.8E-06
2.3E-06
1.1E-04
6.1E-05
4.1E-05
No
8.9E-06
No
3.9E-05
WR
3.7E-05
7.8E-05
1.1E-04
4.4E-05
NR
NR
1.8E-06
WR
7.6E-05
NR
2.3E-05
NR
3.5E-06
NR
3.5E-05
No
NR
NR
3.8E-05
NR
1.3E-06
No
2.3E-06
6.0E-07
2.8E-05
9.0E-06
1.3E-05
No
2.4E-06
No
9.9E-06
WR
1.1E-05
1.4E-05
4.7E-05
1.6E-05
NR
NR
1.6E-06
1.4E-06
1.1E-04
NR
4.0E-06
NR
1.3E-06
NR
1.3E-05
NR
NR
NR
4.0E-06
NR a b
Relative potency factors shown are derived from EC50 at 24 h, 48 h and 72 h exposure of H4IIE-luc cells.
No
¼ no quantifiable response.
c
WR
¼ weak response
<
50% of TCDD maximum induction.
NR
¼ not reported; REPs
¼ relative potency factors; PAHs
¼ polycyclic aromatic hydrocarbons; TCDD
¼
2,3,7,8-tetrachlorodibenzo-p-dioxin; Pyr
¼ pyrene;
Fla
¼ fluoranthene; Chr
¼ chrysene; B[a]A
¼ benzo[a]anthracene; B[k]F
¼ benzo[k]fluoranthene; B[j]F
¼ benzo[j]fluoranthene; B[b]F
¼ benzo[b]fluoranthene;
B[e]P
¼ benzo[e]pyrene; B[a]P
¼ benzo[a]pyrene; Per
¼ perylene; I[123-cd]P
¼ indeno[1,2,3-cd]pyrene; B[ghi]P
¼ benzo[g,h,i]perylene; DB[ac]A
¼ dibenz[a,c]anthracene; DB[aj]A
¼ dibenz[a,j]anthracene; DB[ah]A
¼ dibenz[a,h]anthracene; NPyr
¼ naphto[2,3-a]pyrene.
NR
NR
WR c
WR
5.3E-05
NR
1.4E-05
NR
WR
NR
4.8E-05
No
NR
NR
1.1E-04
NR dose addition models to predict the toxicity of PAH mixtures.
Additivity models are supposed to underestimate the risks because of PAHs’ multiple cellular mechanisms. For example, the TEQ model cannot predict the interactive developmental effects of PAHs [3,4]. Olmstead and LeBlanc [31] reported the joint effects of four PAHs on the growth rate on the crustacean
Daphnia magna during sub-chronic exposure. The CA model tended to overpredict the toxicity of the PAH mixtures, and their data was represented better by an alternative model based on the concept of independent joint action. The CYP1A induction of
19 PAHs in a fish hepatoma cell line was investigated in a study from Fent and Ba¨tscher [32]. They observed additivity interactions in different PAH mixtures using the TEQ concept based on EC50-derived induction equivalency factors. In the present study, the AhR-mediated activity of PAHs was examined using a rat hepatoma cell line. The CA model and the REP concept were used to investigate the additivity interactions of PAHs.
Generally, increased numbers of PAHs in the mixtures resulted in more additive behavior of the mixture. After 72 h of exposure, all three mixtures tested showed additivity, and for shorter
150
2-Mix
100
50
150
100
50
3B-Mix
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2 observed predicted
4A-Mix 4B-Mix
150
100
50
150
100
50
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
Concentration (
µ
M)
Fig. 2. The concentration–response relationship observed and predicted for four polycyclic aromatic hydrocarbon (PAH) mixtures at 24 h of exposure in the
H4IIE-luc bioassay. The mixtures are mixed in equal molar concentration, 1:1 ratio. The observed data is based on a single experiment consisting of triplicates.
Predicted data is based on two to four experiments for each of the individual PAHs, with triplicate exposures for each concentration. The predicted curves were calculated using the concentration addition (CA) model as described in Materials and Methods . Error bars represent standard deviations.

AhR agonistic potencies and additivity of PAHs Environ. Toxicol. Chem.
31, 2012
24 h 48 h 72 h
150
3A-Mix
150
100
150
100 100
50 50 50
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2 observed predicted
150
100
8-Mix
150
100
150
100
50 50 50
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
1155
150
15-Mix
100
50
150
100
50
150
100
50
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 2
Concentration ( µ M)
Fig. 3. The concentration–response relationships observed and predicted for three polycyclic aromatic hydrocarbon (PAH) mixtures at 24, 48, and 72 h of exposure in the H4IIE-luc bioassay. The mixtures are mixed in equal molar concentration, 1:1 ratio. Observed data are based on two experiments consisting of triplicates. Predicted data are based on two to four independent experiments for each of the individual PAHs and the mixtures with triplicate exposures for each concentration. The predicted curves were calculated using the concentration addition (CA) model as described in Materials and Methods . Error bars represent standard deviations.
exposures, the observed effects were slightly greater than predicted from the CA model, suggesting synergistic interactions in the binary to quarternary mixes. The reasons for the slight synergistic effects observed are unknown but could be related to the superinduction seen by several of the PAHs. It is also possible that the differences between observed and predicted effects are within the precision of the bioassay, taking into account the uncertainties of the single PAH REPs.
Environmental samples contain a complex cocktail of chemicals; the levels vary depending on the source of contamination and the availability and degradability of the chemicals. Mass balance analysis of PAH-contaminated samples has shown that chemically derived TEQs often account for only a portion of the
AhR-mediated activities found in samples from contaminated sites [23,33], and the results from the present study suggest additive behavior of PAHs through the Ah receptor. It would be interesting to further examine additional PAHs and their relative contribution to the total AhR-inducing potency of environmental extracts. Further research can investigate the additivity in more complex mixtures containing numerous PAHs, 20 or more, in a variety of combinations. It would also be valuable to investigate whether the effects in the H4IIE-luc bioassay are
Table 4. Comparison of relative potency factors (REPs) observed in the polycyclic aromatic hydrocarbon (PAH) mixtures and predicted mixture REPs calculated by the relative potencies of the single PAHs at effective concentration (EC25 and EC50)
24-h exposure 48-h exposure 72-h exposure
Predicted Observed Predicted Observed Predicted Observed
EC25
15-mix
8-mix
3A-mix
2-mix
3B-mix
4A-mix
4B-mix
EC50
15-mix
8-mix
3A-mix
2-mix
3B-mix
4A-mix
4B-mix
4.39E-04
2.68E-04
2.14E-04
1.18E-04
9.66E-04
5.01E-06
6.27E-05
1.52E-04
9.77E-05
9.23E-05
6.59E-05
3.09E-04
4.17E-06
3.61E-05
3.79E-04
1.38E-04
2.90E-04
1.44E-03
6.09E-03
7.45E-05
2.63E-04
5.43E-05
5.59E-05
1.42E-04
3.44E-04
8.34E-04
2.55E-05
8.44E-05
6.00E-05
5.00E-05
5.15E-05
3.57E-05
2.57E-05
3.06E-05
7.49E-05
3.68E-05
9.06E-05
4.59E-05
2.48E-05
4.63E-05
1.42E-05
1.34E-05
1.16E-05
1.06E-05
1.05E-05
9.44E-06
1.59E-05
5.62E-06
1.40E-05
1.15E-05
4.89E-06
9.44E-06

1156 Environ. Toxicol. Chem.
31, 2012 additive for more complex samples containing both PAHs and other AhR-activating contaminants. In vivo studies on superinduction by PAH mixtures would also advance knowledge in this area.
CONCLUSIONS
The present study has contributed to a growing number of H4IIE-luc specific REPs of PAHs that can be used in the mass balance analysis of PAH-contaminated samples. Relative potency factors were similar to those reported in other studies.
Declining potencies with an increasing exposure duration were shown for all PAHs tested and are assumed to be a consequence of metabolism. Based on our results, we propose that the AhRinducing potency of more persistent contaminants in a sample can be calculated using 72-h-based REPs instead of using selective cleanup methods.
Additive behavior of PAHs is important to apply the REP concept in the risk assessment of PAH-contaminated samples successfully. The additivity of PAHs in seven mixtures was studied using the CA model and the REP concept. Our results indicated that the additivity of PAHs in a mixture depends on exposure time. Generally, the correspondence with the additivity model inherent in the CA and TEF concept was best at long exposure times, and all three mixtures demonstrated additive effects on luciferase induction after 72 h of exposure.
Shorter exposure times tended to yield higher potencies than predicted in the CA or REP models for the less complex mixtures, suggesting synergistic interactions to some extent.
Our results emphasize the use of method-specific REPs in mass balance calculations and mixture toxicity studies. Further research is needed to identify other AhR-activating substances and test additional mixtures, to improve the reliability of REPs in risk assessments and the knowledge of mixture interactions through the Ah receptor. In addition, in vivo studies should be done to see if superinduction of PAH toxicity is a real risk to consider.
Acknowledgement— This project was supported financially by Sparbanksstiftelsen Nya, the Swedish Environmental Protection Agency, and the KK
Foundation (The Knowledge Foundation).
REFERENCES
1. Balch GC, Metcalfe CD, Huestis SY. 1995. Identification of potential fish carcinogens in sediment from Hamilton Harbour, Ontario, Canada.
Environ Toxicol Chem 14:79–91.
2. Spink DC, Wu SJ, Spink BC, Hussain MM, Vakharia DD, Pentecost
BT, Kaminsky LS. 2008. Induction of CYP1A1 and CYP1B1 by benzo(k)fluoranthene and benzo(a)pyrene in T-47D human breast cancer cells: Roles of PAH interactions and PAH metabolites.
Toxicol Appl
Pharm 226:213–224.
3. Incardona JP, Day HL, Collier TK, Scholz NL. 2006. Developmental toxicity of 4-ring polycyclic aromatic hydrocarbons in zebrafish is differentially dependent on AH receptor isoforms and hepatic cytochrome P4501A metabolism.
Toxicol Appl Pharm 217:308–321.
4. Billiard SM, Meyer JN, Wassenberg DM, Hodson PV, Di Giulio RT.
2008. Nonadditive effects of PAHs on early vertebrate development:
Mechanisms and implications for risk assessment.
Toxicol Sci 105:
5–23.
5. Ma Q, Lu AYH. 2007. CYP1A induction and human risk assessment: An evolving tale of in vitro and in vivo studies.
Drug Metab Dispos 35:1009–
1016.
6. Lemieux CL, Lambert IB, Lundstedt S, Tysklind M, White PA. 2008.
Mutagenic hazards of complex polycyclic aromatic hydrocarbon mixtures in contaminated soil.
Environ Toxicol Chem 27:978–990.
7. Behnisch PA, Hosoe K, Sakai S. 2001. Bioanalytical screening methods for dioxins and dioxin-like compounds—A review of bioassay/ biomarker technology.
Environ Int 27:413–439.
M. Larsson et al.
8. Machala M, Vondracek J, Blaha L, Ciganek M, Neca J. 2001. Aryl hydrocarbon receptor-mediated activity of mutagenic polycyclic aromatic hydrocarbons determined using in vitro reporter gene assay.
Mutat Res-Gen Tox En 497:49–62.
9. Denison MS, Pandini A, Nagy SR, Baldwin EP, Bonati L. 2002.
Ligand binding and activation of the Ah receptor.
Chem Biol Interact
141:3–24.
10. Marlowe JL, Puga A. 2005. Aryl hydrocarbon receptor, cell cycle regulation, toxicity, and tumorigenesis.
J Cell Biochem 96:1174–
1184.
11. Behnisch PA, Hosoe K, Sakai S. 2003. Brominated dioxin-like compounds: In vitro assessment in comparison to classical dioxin-like compounds and other polyaromatic compounds.
Environ Int 29:861–
877.
12. Masunaga S, Sakashita R, Furuichi T, Shirai J, Kannan K, Giesy J. 2004.
Effect of exposure duration on the aryl hydrocarbon receptor-mediated activity of polycyclic aromatic hydrocarbons measured by in vitro reporter gene assay.
Organohalogen Compd 66:623–629.
13. Till M, Riebniger D, Schmitz H-J, Schrenk D. 1999. Potency of various polycyclic aromatic hydrocarbons as inducers of CYP1A1 in rat hepatocyte cultures.
Chem Biol Interact 117:135–150.
14. Ziccardi MH, Gardner IA, Denison MS. 2002. Application of the luciferase recombinant cell culture bioassay system for the analysis of polycyclic aromatic hydrocarbons.
Environ Toxicol Chem 21:2027–
2033.
15. Villeneuve DL, Khim JS, Kannan K, Giesy JP. 2002. Relative potencies of individual polycyclic aromatic hydrocarbons to induce dioxinlike and estrogenic responses in three cell lines.
Environ Toxicol Chem 17:128–
137.
16. Engwall M, Brunstro¨m B, Na¨f C, Hjelm K. 1999. Levels of dioxin-like compounds in sewage sludge determined with a bioassay based on
EROD induction in chicken embryo liver cultures.
Chemosphere
38:2327–2343.
17. Payne J, Rajapakse N, Wilkins M, Kortenkamp A. 2000. Prediction and assessment of the effects of mixtures of four xenoestrogens.
Environ
Health Perspect 108:983–987.
18. Faust M, Altenburger R, Backhaus T, Blanck H, Boedeker W, Gramatica
P, Hamer V, Scholze M, Vighi M, Grimme LH. 2001. Predicting the joint algal toxicity of multi-component s-triazine mixtures at low-effect concentrations of individual toxicants.
Aquat Toxicol 56:13–32.
19. Arrhenius A
Predictability of the mixture toxicity of 12 similarly acting congeneric inhibitors of photosystem II in marine periphyton and epipsammon communities.
Aquat Toxicol 68:351–367.
20. Zhang Y-H, Liu S-S, Song X-Q, Ge H-L. 2008. Prediction for the mixture toxicity of six organophosphorus pesticides to the luminescent bacterium
Q67.
Ecotoxicol Environ Saf 71:880–888.
21. Altenburger R, Backhaus T, Boedeker W, Faust M, Scholze M, Grimme
LH. 2000. Predictability of the toxicity of multiple chemical mixtures to
Vibrio fischeri : Mixtures composed of similarly acting chemicals.
Environ Toxicol Chem 19:2341–2347.
22. Bellas J, Saco-A evaluation of polycyclic aromatic hydrocarbons using marine invertebrate embryo-larval bioassays.
Mar Pollut Bull 57:493–502.
23. Andersson E, Rotander A, von Kronhelm T, Berggren A, Ivarsson P,
Hollert H, Engwall M. 2009. AhR agonist and genotoxicant bioavailability in a PAH-contaminated soil undergoing biological treatment.
Environ Sci Pollut Res 16:521–530.
24. Lundstedt S, Haglund P, Oberg L. 2003. Degradation and formation of polycyclic aromatic compounds during bioslurry treatment of an aged gasworks soil.
Environ Toxicol Chem 22:1413–1420.
25. Bergknut M, Kitti A, Lundstedt S, Tysklind M, Haglund P. 2004.
Assessment of the availability of polycyclic aromatic hydrocarbons from gasworks soil using different extraction solvents and techniques.
Environ Toxicol Chem 23:1861–1866.
26. Murk AJ, Legler J, Denison MS, Giesy JP, Van De Guchte C, Brouwer A.
1996. Chemical-activated luciferase gene expression (CALUX): A novel in vitro bioassay for Ah receptor active compounds in sediments and pore water.
Fund Appl Toxicol 33:149–160.
27. Berenbaum MC. 1985. The expected effect of a combination of agents:
The general solution.
J Theor Biol 114:413–431.
28. Villeneuve DL, Blankenship AL, Giesy JP. 2000. Derivation and application of relative potency estimates based on in vitro bioassay results.
Environ Toxicol Chem 19:2835–2843.
29. Windal I, Denison MS, Birnbaum LS, Van Wouwe N, Baeyens W,
Goeyens L. 2005. Chemically activated luciferase gene expression
(CALUX) cell bioassay analysis for the estimation of dioxin-like

AhR agonistic potencies and additivity of PAHs activity: Critical parameters of the CALUX procedure that impact assay results.
Environ Sci Technol 39:7357–7364.
30. Rajapakse N, Ong D, Kortenkamp A. 2001. Defining the impact of weakly estrogenic chemicals on the action of steroidal estrogens.
Toxicol
Sci 60:296–304.
31. Olmstead AW, LeBlanc GA. 2005. Joint action of polycyclic aromatic hydrocarbons: Predictive modeling of sublethal toxicity.
Aquat Toxicol
75:253–262.
Environ. Toxicol. Chem.
31, 2012 1157
32. Fent K, Ba¨tscher R. 2000. Cytochrome P4501A induction potencies of polycyclic aromatic hydrocarbons in a fish hepatoma cell line:
Demonstration of additive interactions.
Environ Toxicol Chem 19:
2047–2058.
33. Keiter S, Grund S, van Bavel B, Hagberg J, Engwall M, Kammann U,
Klempt M, Manz W, Olsman H, Braunbeck T, Hollert H. 2008. Activities and identification of aryl hydrocarbon receptor agonists in sediments from the Danube river.
Anal Bioanal Chem 390:2009–2019.