Embed Size (px)
Citation preview
IA
SMa
b
c
a
ARRAA
K2SAANF
1
icwrtThc2
mbt
of
(((
0d
Toxicology Letters 203 (2011) 252–257
Contents lists available at ScienceDirect
Toxicology Letters
journa l homepage: www.e lsev ier .com/ locate / tox le t
mmediate and highly sensitive aversion response to a novel food item linked toH receptor stimulation�
anna Lensua,b,∗, Jouni T. Tuomistoa, Jouko Tuomistoa, Matti Vilukselaa,arjo Niittynena, Raimo Pohjanvirtac
Department of Environmental Health, National Institute for Health and Welfare (THL), P.O.B. 95, FI-70701 Kuopio, FinlandFaculty of Health Sciences, The School of Pharmacy, University of Eastern Finland, P.O.B. 1611, FI-70211 Kuopio, FinlandDepartment of Food Hygiene and Environmental Health, Faculty of Veterinary Medicine, University of Helsinki, P.O.B. 66, FI-00014 University of Helsinki, Kuopio, Finland
r t i c l e i n f o
rticle history:eceived 25 January 2011eceived in revised form 17 March 2011ccepted 21 March 2011vailable online 31 March 2011
eywords:,3,7,8-Tetrachlorodibenzo-p-dioxin
a b s t r a c t
Aversion to novel food items was studied in male rats and mice after 2,3,7,8-tetrachlorodibenzo-p-dioxin(TCDD) exposure using chocolate consumption as an indicator. The correlation of this phenomenon withsusceptibility to acute toxicity and CYP1A1 induction was examined by determining the dose–responseof chocolate aversion in differently dioxin-sensitive rat lines after TCDD (0.01–10 �g/kg). Furthermore,the dependence of this behavioral alteration on the AH receptor (AHR) was studied employing AHR-deficient and wild-type mice. We offered chocolate for both species as a novel food item immediatelyafter the exposure, and it was available with standard rodent chow for 3 days. The ED50 value for the
ensitivity differencesH receptorHR-knockout miceeophobiaood aversion
extremely resistant rat line A (LD50 value > 10,000 �g/kg) was 0.36 �g/kg, for the semi-resistant line B(LD50 value 830 �g/kg) 1.07 �g/kg and for the TCDD-sensitive line C (LD50 value 40 �g/kg) 0.34 �g/kg.Interestingly, the ED50 values for chocolate aversion were very similar to those for CYP1A1 induction inthese rat lines. Findings on AHR-deficient and wild-type mice implied the involvement of the AHR inthis intriguing response, which may thus represent a mechanism to restrict exposure to potentially toxicdietary substances causing hepatic induction of drug-metabolizing enzymes.
. Introduction
Polychlorinated dioxins and dioxin-like chemicals are contam-nants found practically everywhere in the environment. Theseompounds belong to the group of halogenated aryl hydrocarbonshich share structural similarity and similar patterns of biological
esponses. Dioxins are extremely resistant to biodegradation andhey concentrate in the food chain by accumulating in body fat.hese chemicals induce various biological and toxic effects which
ave been shown to be mediated via binding to cytosolic aryl hydro-arbon receptor (AH receptor, AHR) [reviewed e.g. in Lindén et al.,010; Pohjanvirta and Tuomisto, 1994; White and Birnbaum, 2009].� Preliminary results of some of the data were presented in the followingeetings: the 7th Valamo conference on Enviroment and Health-approaches to
enefit-risk analysis, 3–5.12.07 Valamo, Heinävesi, Finland and the 45th Congress ofhe European Societies of Toxicology (Eurotox 2008), Rhodes, Greece, 5–8.10.2008.∗ Corresponding author at: Department of Biology of Physical Activity, University
f Jyväskylä, P.O.Box 35, FI-40014, Jyväskylä, Finland. Tel.: +358 50 3011 272;ax: +358 14 260 2071.
E-mail addresses: [email protected] (S. Lensu), [email protected]. Tuomisto), [email protected] (J. Tuomisto), [email protected]. Viluksela), [email protected] (M. Niittynen), [email protected]. Pohjanvirta).
378-4274/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.toxlet.2011.03.025
© 2011 Elsevier Ireland Ltd. All rights reserved.
AHR protein is a highly conserved transcription factor. In theabsence of a ligand, the AHR is located in the cytosol in associa-tion with the chaperone proteins heat shock protein 90, XAP2 andp23. After ligand binding the AHR translocates into the nucleus. Thechaperone proteins dissociate from it and the AHR heterodimerizeswith a closely related protein, AHR nuclear translocator (ARNT). Thecomplex formed then binds to xenobiotic response elements (XREs)in the DNA and thereby leads to alterations in the transcriptionalactivity of AHR responsive genes (Kazlauskas et al., 2000; Mimuraet al., 1999; Whitlock, 1990).
Following acute exposure, TCDD is the most toxic synthetic com-pound known. However, sensitivity to its toxicity varies amonganimal species ranging from the most sensitive lake trout sack fry[LD50 value 0.074 �g/kg, injected into egg (Walker et al., 1996)]to the most resistant rat strain, Han/Wistar [Kuopio; H/W; LD50value more than 9600 �g/kg (Unkila et al., 1994)]. The most sensi-tive rat substrain is Long-Evans (Turku/AB; L-E; LD50 ∼10–20 �g/kg)(Pohjanvirta et al., 1993). In other words, there is a thousand-folddifference within a single species in sensitivity to TCDD lethality.This interstrain difference originates from a polymorphism in the
AHR. H/W rats were shown to have a deletion/insertion type muta-tion in the AHR mRNA and thereby two smaller receptor proteins(∼98 kDa) than the wildtype AHR present in L-E rats (∼106 kDa)(Pohjanvirta et al., 1998, 1999). In addition to the AHR, there is
y Lett
aroat
erTrai(pafcTon
ditmcca
2
2
NWl(rwatmGE
cGlacdCi(
2
uew1Tmtw
2
2s
s
S. Lensu et al. / Toxicolog
currently unidentified gene B contributing to the resistance, asevealed by crossbreeding experiments with H/W and L-E rats inur laboratory. The outcome was three different lines of rats (A, Bnd C) having different alleles of the AHR and B genes in their geno-ypes and different sensitivities to TCDD (Tuomisto et al., 1999).
Alterations in feeding behavior are characteristic of TCDDxposure in laboratory animals. In addition to their persistentlyeduced feed intake, lost body weight and (at lethal doses) wasting,CDD-treated rats display modified responses to various feedingegulatory challenges [reviewed in Lindén et al., 2010; Pohjanvirtand Tuomisto, 1994]. The impact of TCDD on ingestive behav-or appears to be specific and not e.g. due to general malaisePohjanvirta et al., 1994). Some years ago, our group made a sur-rising observation: soon after TCDD treatment rats exhibitedvoidance of such novel food items which they normally craveor, chocolate and cheese (Tuomisto et al., 2000). This behavioralhange was found to emerge during the first night post-exposure.hus, it appeared to represent one of the most rapidly developingvert signs of TCDD exposure in adult animals. Furthermore, it didot seem to correlate with sensitivity to TCDD lethality.
In the present study, we sought to characterize theose–response relationship of this phenomenon. We also exam-
ned its generality by testing mice in addition to rats. Moreover,he role of the AHR was explored by including AHR-knockout
ice in the experimental setting. As the novel food item we usedhocolate, because normally rodents find it highly tasty and readilyonsume it, but in our previous study TCDD treatment resulted inversion to it (Tuomisto et al., 2000).
. Material and methods
.1. General conditions in animal experiments, TCDD exposure
Animals for the experiments were obtained from the breeding colonies of theational Public Health Institute (presently the National Institute for Health andelfare), Kuopio, Finland. In the study, male rats were individually housed in stain-
ess steel wire-mesh cages. Male mice were housed individually in polycarbonateMacrolon) cages and had aspen chips (Tapvei, Kaavi, Finland) as bedding mate-ial. All animals were adapted to the conditions and daily handling for at least aeek before starting the experiments. In the animal room, temperature (21 ± 1 ◦C)
nd humidity (50 ± 10%) were controlled, and artificial lights were on from 7 a.m.o 7 p.m. All the animal experiments were reviewed and approved by the Ani-
al Experiment Committee of the University of Kuopio and the Kuopio Provincialovernment, and they were conducted in accordance with the Guidelines of theuropean Community Council directives 86/609/EEC.
The rats were fed on powdered feed (R36, Lactamin, Södertälje, Sweden; energyontent 12.6 kJ/g). Mice were fed on pelleted chow (Altromin 1314F, Seelenkamp,ermany; energy content 12.5 kJ/g). These feeds and tap water were available ad
ibitum throughout the studies. TCDD (UFA-Oil institute, Ufa, Russia) was >99% pures determined by gas chromatography–mass spectrometry and it was dissolved inorn oil (Sigma, C8267, St. Louis, MO, USA) (Pohjanvirta et al., 1987). Animals wereosed orally, the administration volume for rats was 4 ml/kg and for mice 10 ml/kg.ontrol animals were treated with an equal volume of corn oil. The chocolate used
n our experiments was regular milk chocolate commercially available in FinlandPanda, Vaajakoski, Finland; energy content 23 kJ/g).
.2. Data analysis and statistics
The data were analyzed with MS-Excel and SPSS 17.0. Linear mixed model wassed to estimate dose–response effects among different TCDD doses and differ-nt animals. Because of continuous data of rats and mice, daily feeding measuresere analyzed by accounting for their repeated nature individually (Zeger and Liang,
986). Linear mixed model enabled testing of continuous responses (the effect ofCDD dose on consumptions by time) and interactions (dose, measuring day andouse or rat genotype). An autoregressive correlation structure turned out to model
he dependence of the observations adequately. The level of statistical significanceas set at the p-value equal to or less than 0.05.
.3. Experimental designs
.3.1. Dose–response of aversion to a novel food item in rats with differentensitivities to acute lethality of TCDD
To find out whether the sensitivity to the aversive effect of TCDD depends onensitivity to acute toxicity, rat lines generated by crossbreeding H/W and L-E rats
ers 203 (2011) 252–257 253
were used. The genotype and sensitivity to TCDD lethality of the lines A, B and C areknown (Table 1). Line A rats are the most resistant of these lines being homozygousfor the Han-Wistar –type resistance alleles of the AHR gene (Ahrhw/hw). Line B ratsare semi-resistant being homozygous for the Han-Wistar –type resistance allelesof gene B (Bhw/hw). Line C rats are the most sensitive being homozygous for thewildtype alleles of both of these TCDD-resistance genes (Ahrwt/wt, Bwt/wt) (Tuomistoet al., 1999).
At the start of the habituation to single-rat wire-mesh cages and powdered feed,male rats of all three lines were 7–12 weeks old. The mean number of rats per groupwas seven (range 3–10). Rats were exposed to a wide range (0.01–10.0 �g/kg) ofTCDD doses orally (po) and controls received an equal volume (4 ml/kg) of corn oil(Fig. 1). On the day of exposure, the mean weight (±SD) was 251 ± 43 g (n = 50) forline A rats, 276 ± 45 g (n = 49) for line B rats, and 280 ± 21 g (n = 53) for line C rats.Due to the interline variation in body weight, we verified the results by calculatingthe metabolic masses of the rats [body weight (kg)]0.67 (Donhoffer, 1986; Feldmanand McMahon, 1983). Thereby the different sizes of the animals were taken intoaccount and the caloric intakes of foods were related to the metabolic needs of therats.
Water was available ad libitum throughout the study. TCDD was administeredbetween 9 a.m. and 12 noon, and immediately after the exposure chocolate wasoffered for all the rats as a novel food item. Chow and chocolate consumptions wererecorded once a day for three consecutive days after exposure. The mean value of 3days’ consumptions was used in the dose–response assessment.
2.3.1.1. Assessment of ED50 value for avoidance of a novel food item, chocolate.ED50 (effective dose; causing a 50% response in exposed animals) values and theirconfidence limits for inducing chocolate avoidance in three rat lines were estimatedby benchmark dose analysis. Analysis was done using Benchmark Dose Software2.1.2. (BMDS) by US EPA (http://www.epa.gov/ncea/bmds/index.html). For the esti-mation, continuous Hill model was used. The form of the response function in Hillmodel is:
Y[dose] = � +[
� × dosen
kn + dosen
]
where Y[dose] is the observed effect at [dose]; � = intercept (control effect, Ymin);k = slope (ED50); n = power factor, Hill coefficient; � = Sign (Ymax – Ymin).
In the model, variance was assumed to change as a power function of meanvalue. Non-homogenous variance for the ith dose group is as follows:
ı2i
= ˛[�(dosei)]�
where ˛ = power parameter, �(dosei) = observed mean (from the model) for the ithdose group and � = coefficient of variation.
In the Hill model, � was set to 1 because in our data variances were propor-tional to the mean. Furthermore, in the model ˛ was restricted to be ≥1 and n > 1.If ˛ < 1, then the slope of the dose–response curve becomes infinite at the controldose.
2.3.2. Generality of the novel food item avoidance and involvement of the AHR inthe response
To test whether chocolate avoidance after TCDD exposure also occurs in otherspecies than rats, and to find out the role of the AHR in the response, we offeredchocolate as a novel food item to mice. We used two genotypes of C57BL/6 mice:wildtype C57BL/6Kuo (Pohjanvirta et al., submitted for publication) and AHR knock-out (Ahr−/− , originating from The Jackson Laboratories, Bar Harbor, ME, USA). TheC57BL/6Kuo mice are homozygous for the AHRb−1 allele, hence they carry a high-affinity form of the AHR and are responsive to TCDD. The LD50 value for male miceis ∼350 �g/kg (Pohjanvirta et al., submitted for publication). Ahr−/− mice are unableto express the AHR due to replacement of exon 2 of the Ahr gene with a neomycin-resistance gene (Schmidt et al., 1996) and they are resistant to AHR-mediated effectsof TCDD, for example acute lethality (Fernandez-Salguero et al., 1996; Mimuraet al., 1997). The genotype of these mice for the unidentified resistance gene B isunknown.
The mice (n = 25) were 14–26-week-old at the exposure and weighed 28.1 ± 2.2 g(mean ± SD). Groups did not differ in this respect as the mice were age- and weight-matched among the groups, five mice per group. TCDD or corn oil (10 ml/kg) wereadministered in a random order to C57BL/6Kuo (corn oil, 5.0 �g/kg or 100.0 �g/kg,po) and Ahr−/− mice (corn oil or 100 �g/kg, po). Mice were exposed between12 noon and 1 p.m. and chocolate was given to all animals immediately after
administration. Body weights of the animals and their consumptions of regularchow, chocolate and water were recorded daily for three consecutive days afterexposure.All original data are available at Opasnet website (http://en.opasnet.org/w/TCDDprovoked aversion to foods in rats).
254 S. Lensu et al. / Toxicology Letters 203 (2011) 252–257
Table 1ED50 values and 95% confidence intervals for aversion to a novel food item, chocolate (male rats) along with ED50 values for EROD activity (measure of CYP1A1 induction) andLD50 values for the rat lines used in the study. The genotype of rats affects TCDD-sensitivity: Han-Wistar-type allele (hw) of the AHR and B genes is related with resistance toacute toxicity and found in resistant rats whereas wild-type (wt) allele is found in most rats sensitive to TCDD toxicity.
Rat line Genotype Lethality, LD50a (�g/kg) EROD activity, ED50
b (�g/kg) Aversion to novel fooditem, ED50 (�g/kg)
Aversion to novel fooditem, 95% CI (�g/kg)
A Ahrhw/hw , Bwt/wt >10,000 0.15 0.36 0.13–0.58B Ahrwt/wt , Bhw/hw 830 0.28 1.07 0.86–1.28C Ahrwt/wt , Bwt/wt 40 0.14 0.34 0.26–0.42
3
3r
tlwr1(taf
cwist3l0irlm
ottsotccnewmm
wb(Fmt
3
s
a For male rats, data from Tuomisto et al. (1999).b For female rats, data from Simanainen et al. (2003).
. Results
.1. Dose–response of aversion to a novel food item in differentat lines
TCDD dose-dependently decreased the daily mean consump-ion of chocolate measured over 3 days post-exposure in all ratines resulting in a total aversion at 3 �g/kg (Fig. 1A). The decrease
as significant at 0.30, 3.0 and 1.0 �g/kg TCDD in line A, B and Cats, respectively. ED50 values were 0.36 �g/kg for line A rats and.07 �g/kg for line B rats (Table 1). The 95% confidence intervalsCIs) for ED50 values were 0.13–0.58 and 0.86–1.28 �g/kg, respec-ively. Line C turned out to be the most sensitive rat line as totalversion was already recorded at a dose of 1.0 �g/kg. The ED50 valueor line C rats was 0.34 �g/kg (95% CI 0.26–0.42 �g/kg).
On the day of exposure, during the first measuring day, chocolateonsumption in line A rats diminished significantly as comparedith controls at doses equal to or higher than 3.0 �g/kg TCDD and
n line C rats at doses equal to or higher than 1.0 �g/kg (data nothown). In line B rats the decrease reached significance only onhe second day post exposure, at doses equal to or higher than.0 �g/kg. During the first day post-exposure an increase in choco-
ate consumption was observed in line B and C rats exposed to doses.3 �g/kg and below (as compared with their own control level dur-
ng the exposure day). Nevertheless, their chocolate consumptionemained at the control level or below it. By the second day choco-ate consumption had stabilized in all exposed groups (linear mixed
odel, data not shown).Low doses of TCDD did not affect chow consumption (Fig. 1B)
r body weights of the rats (data not shown). However, becausehe weights already differed among and within the lines beforehe exposure, the results were normalized by calculating the con-umptions of chocolate and chow in relation to the metabolic massf each rat [BW (kg)]0.67. The result of chocolate aversion remainedhe same, in both continuous data and the aggregated data of 3 days’onsumption (data not shown). The highest doses caused signifi-ant effects in total energy intake (as assessed relative to metaboliceeds, Fig. 1C). In general, there was a strain difference in the totalnergy intake [F (2, 118.64) = 27.202, p < 0.001, linear mixed model]ith line C rats consuming significantly less energy per metabolicass than rats of lines A or B (p < 0.001 for both, linear mixedodel).To verify the results of novel chocolate aversion, ED50 values
ere also estimated by relating the intake of chocolate to metabolicody weight. By that means the ED50 value for line A was 0.38 �g/kg95% CI: 0.23–0.54) and for line C 0.35 �g/kg (95% CI: 0.27 – 0.43).or the estimation of ED50 value for line B only the constant varianceodel (� = 0) proved to model the data adequately, and as a result
he ED50 value for line B was 1.10 �g/kg (95% CI: 0.04–2.17).
.2. Role of the AHR in aversion to a novel food item in mice
AHR status affected chocolate preference (Fig. 2A). TCDD expo-ure caused neophobia-like aversion in wildtype C57BL/6Kuo mice,
but the effect only lasted for 2 days. Wildtype C57BL/6Kuo controlmice preferred chocolate despite its novelty. In contrast, controlAhr−/− mice appeared to be intrinsically neophobic eating onlyabout half as much chocolate as their wildtype partners did onday 0. Thereafter, control Ahr−/− mice accepted chocolate in thesame fashion as did TCDD-treated wildtype mice. Surprisingly, theimmediate reaction of TCDD-treated Ahr−/− mice did not differfrom that of wildtype C57BL/6Kuo controls; their chocolate con-sumption on day 0 was thus twice that of control Ahr−/− mice.During the next three days, TCDD-treated Ahr−/− mice increasedtheir chocolate intake in parallel with control Ahr−/− mice. Chowconsumption diminished already on day 0 in all groups from theirbefore-treatment level (Fig. 2B), remaining then at that low level.However, the difference between TCDD-exposed and control miceagain displayed a mirror image in wildtype and Ahr−/− mice (inwildtype mice, controls ate less chow than TCDD-exposed animalswhereas the converse was true of Ahr−/− mice). TCDD and mousegenotype had a significant interactive effect on both consump-tions [chocolate: F(1, 20) = 5.269, p < 0.05 and chow: F(1, 20) = 6.034,p < 0.05, linear mixed model].
The increased chocolate consumption resulted in an increase intotal energy intake during the observation period (Fig. 2C), inde-pendently of genotype or TCDD treatment. Only for C57BL/6Kuocontrols the increase failed to reach significance. The weights of theanimals did not differ in the course of the study, and TCDD did notaffect water consumption of the mice (data not shown). However,the genotypes had a difference in basal water consumption as con-trol C57BL/6Kuo mice drank more than Ahr−/− controls from day0 onwards [F(1, 60.995) > 4.35, p < 0.05 for each day, linear mixedmodel] (data not shown).
4. Discussion
We have previously reported a surprisingly rapid alteration infood selection in rats treated with a high dose of TCDD (Tuomistoet al., 2000). These rats exhibited avoidance of such novel fooditems which they normally crave for, chocolate and cheese. Thisbehavioral change was found to emerge during the first night post-exposure. Thus, it appeared to represent one of the most immediateresponses ever reported in adult animals following a single dose ofTCDD. However, in those studies the doses of TCDD were ratherhigh, with the lowest doses for resistant rats being 50 �g/kg ormore and some of the doses tested in sensitive animals being lethal.Here we demonstrate that aversion to a novel food item, chocolate,is an extremely sensitive response with no relationship with sus-ceptibility to acute lethality of TCDD. In spite of the fact that thethree rat lines differ widely in their sensitivities to acute lethalityof TCDD (Tuomisto et al., 1999), none of them ate chocolate follow-ing a dose as low as 3.0 �g/kg (po). Furthermore, the ED50 valuesfor this aversion were extremely low (0.36 �g/kg, 1.07 �g/kg and
0.34 �g/kg for line A, B and C rats, respectively) compared withtheir LD50 values (>10,000 �g/kg, 830 �g/kg and 40 �g/kg, respec-tively) (Table 1). The result remained practically the same even ifthe intakes were adjusted to metabolic needs.
S. Lensu et al. / Toxicology Letters 203 (2011) 252–257 255
Fig. 1. Dose–responses for chocolate consumption (A), chow consumption (B) andenergy intake (C) in male rats of lines A, B and C. Group mean values of averagedaily consumption for three consecutive days after exposure to TCDD (±SE; n = 7,range 3–10) of daily consumptions are depicted for three consecutive days afterexposure to TCDD. For clarity, error bars are slightly shifted. In statistical analysesthe repetitive nature of measures for each rat was taken into account and statisticallysignificant differences (p ≤ 0.05, linear mixed model) are shown as follows: A fromcontrol in line A, B from control in line B, and C from control in line C. Significantdifferences (p ≤ 0.05, linear mixed model) among lines after an identical TCDD doseare shown as follows: d between lines A and B, e between lines A and C, and f betweenlines B and C.
Fig. 2. Time-courses for chocolate consumption (A), chow consumption (B) andenergy intake (C) in Ahr−/− mice and wildtype male mice (CB57BL/6Kuo) exposed todifferent doses of TCDD. Group mean values (±SE; n = 5) of average daily consump-tions are depicted for four consecutive days after exposure to TCDD. Statisticallysignificant (p ≤ 0.05) differences are indicated with letters as follows: a, from con-trol on the corresponding day; b, between genotypes after an identical treatment; c,within group CB57BL/6Kuo controls, from their own initial measure; d, within groupCB57BL/6Kuo 5 �g/kg, from their own initial measure; e, within group CB57BL/6Kuo100 �g/kg, from their own initial measure; f, within group Ahr−/− controls, from theirown initial measure, and g, within group Ahr−/− 100 �g/kg, from their own initialmeasure.
2 y Lett
aifWd(dtdtscEo2
rrlaicdttbtnca
a(oir2iliaixti
atamptpteTc2
ebftto
the Centre of Environmental Health Risk Analysis [Grant no. 53307]and from the Academy of Finland [Grant no. 123345 to R.P.]. Addi-
56 S. Lensu et al. / Toxicolog
Other sensitive, major acute responses (excluding molecularnd intracellular events) to TCDD in adult animals are scarce. Its known that the most sensitive endpoints of toxicity reported soar are certain developmental defects in rodents (Bell et al., 2010;
hite and Birnbaum, 2009). These include changes in male repro-uctive organs following a single maternal dose of 0.05–1 �g/kgGray et al., 1997; Mably et al., 1992a, 1992b, 1992c) or in toothevelopment commencing at 0.03 �g/kg (Kattainen et al., 2001). Inhe present study, novelty avoidance emerged at a similarly lowose level, but importantly, after acute exposure in adult rats. Ofhe intracellular events the induction of CYP1A1 is one of the mostensitive effects in adult animals. Interestingly, the sensitivity tohocolate aversion correlates quite well with sensitivity to liverROD activity (ethoxyresorufin-O-deethylase activity, a measuref CYP1A1 induction) in these strains (Table 1) (Simanainen et al.,003).
The neophobia-like behavior was also discernible in TCDD-esponsive mice, although it was more transient in mice than inats. Untreated wildtype C57BL/6Kuo mice preferred novel choco-ate to chow whereas untreated Ahr−/− mice exhibited neophobiand selected their familiar chow. However, a surprising outcomen Ahr−/− mice was recorded upon exposure to TCDD. Chocolateonsumption was significantly higher than in their controls onay 0 and then increased further in parallel with the increase inheir corn oil-treated controls. These observations support the viewhe TCDD-induced neophobia-like behavior is AHR-dependent,ecause no sign of it was detectable in Ahr−/− mice after TCDDreatment. Actually, TCDD-exposed Ahr−/− mice tended to eat sig-ificantly more chocolate compared with their corn oil-treatedounterparts, a phenomenon for which we have no explanationt present.
Some of the responses mediated by AHR could be considereddaptive, such as the rapid induction of metabolizing enzymesreviewed in Lindén et al., 2010). More generally, many membersf basic helix-loop-helix/PAS transcription family proteins, includ-ng AHR, act as environmental sensors in the organism and theyegulate downstream responses to environmental cues (Gu et al.,000; McIntosh et al., 2010). Hence AHR may play a protective role
n an organism, acting in the protective organs including skin andiver (reviewed in Bock and Köhle, 2009). Our data of novel foodtem aversion supports this; the aversion could be regarded as andaptive and beneficial response, which is a protective mechanismnduced by AHR activation. In addition to induced metabolism ofenobiotics, AHR activation by small doses of a ligand could leado avoidance of such food which has a chronologic linkage to thenduction of metabolism.
The novelty avoidance behavior can generally be regardeds beneficial to the animal. Taste memory has a role in gus-atory neophobia, which is an organized and adaptive behaviorgainst various toxic, malaise-generating compounds and whichay increase survival. When a novel food item is present, neo-
hobia prevents the animal from eating substantial amounts ofhe food until it appears to be safe. However, attenuation of neo-hobia is adaptive and evolutionarily beneficial because abilityo eat new foods may ensure dietary variability and temporaryxcess energy intake (Hart, 1988; Reilly and Bornovalova, 2005).his adaptation was manifest in control rats in the considerableonsumption of chocolate already in the first night (Tuomisto et al.,000).
The target of TCDD for novelty avoidance seems to be differ-nt from the pathway leading to lowered body weight. It haseen shown that there are wide species- and strain-specific dif-
erences in some (e.g. weight loss and diminished feeding, liveroxicity, lethality; type II responses), but not all TCDD-inducedoxic responses (e.g. enzyme induction, thymus atrophy or devel-pmental dental defects; type I responses) (Kattainen et al., 2001;ers 203 (2011) 252–257
Simanainen et al., 2002). It appears that novelty avoidance belongsto type I endpoints being independent of genotype variationof resistance alleles. However, similarly to induction of drug-metabolizing enzymes in these rat lines (Simanainen et al., 2003),the ED50 value for novel food item aversion was highest in line Brats. The slightly different response of line B rats in comparisonwith lines A and C may suggest a modulating role for the uniden-tified resistance gene B in the aversion response. Because of theexcellent correlation of aversion with enzyme induction, an estab-lished AHR-mediated response to TCDD, these findings reinforcethe notion of AHR involvement in the aversion response. More-over, novelty avoidance emerges at such low doses that they donot negatively affect energy intake or body weight. On the con-trary, in TCDD-treated mice the availability of chocolate increasedtotal energy intake.
As discussed in the previous paragraph, novel food item avoid-ance cannot be secondary to reduced feeding. On the other hand,the existing data do not support a major role in it for avoid-ance of food energy, but the role of conditioned taste aversion(CTA) is a tricky question (Tuomisto et al., 2000). In a previousstudy with saccharine, a typical CTA response was manifest onday 3 in TCDD-resistant H/W rats at 1000 �g/kg TCDD, but was notseen at 50 �g/kg either in H/W rats or in TCDD-sensitive L-E rats(Pohjanvirta et al., 1994). However, in more recent experimentswith chocolate and powdered chow we found strong evidenceof CTA on day 7 in line A rats treated with a much lower doseof TCDD, 3 �g/kg (Lensu et al., 2011). On the other hand, whilenovelty was the avoided property in the case of cheese, choco-late was avoided even if it was provided for the rats for 16 daysprior to exposure, and the avoidance persisted for five weeks(Tuomisto et al., 2000). Hence it outlasted the period of acutedecrease in feed intake of exposed H/W rats, which resume eat-ing at near-control levels in 1–3 weeks (Pohjanvirta and Tuomisto,1987; Pohjanvirta et al., 1987). Furthermore, although the intakeof fat-rich diets seems to be more affected by TCDD exposure thanthat of protein-rich diets (Pohjanvirta et al., 1994; Tuomisto et al.,2000), the diminished energy intake following TCDD exposure hasnot proved to result from selective avoidance of any macronu-trient (Pohjanvirta and Tuomisto, 1990). The avoided property ofchocolate after TCDD is unknown. Whether the avoidance is trig-gered by the taste or texture of the novel food item remains to beestablished.
In conclusion, we have shown here that aversion to a novelfood item is probably an AHR-dependent response emergingboth in rats and mice. Furthermore, it is a rapid and extremelysensitive response being, similarly to CYP1A1 induction, indepen-dent of sensitivity to acute lethality, or of diminished feeding orlost body weight. It may represent an attempt of the body torestrict exposure to diets whose components result in inductionof drug-metabolizing enzymes and thus contain potentially toxicsubstances. The regulatory pathway(s) underlying this intriguingphenomenon remain to be elucidated.
Funding
This work was supported by the Graduate School in Environ-mental Health (Ministry of Education, Finland) [S.L.] and by grantsfrom the Centre of Excellence Program of the Academy of Finland to
tional grants from the following Finnish foundations are warmlyacknowledged: Societas Biologica Fennica Vanamo, Foundation ofEmil Aaltonen and the Finnish Cultural Foundation, Regional fundof Northern Savo [S.L.].
y Lett
C
aa
A
NfsHft
R
B
B
D
F
F
G
G
H
K
K
L
L
M
M
M
197–211.
S. Lensu et al. / Toxicolog
onflict of interest statement
Study sponsors had no role in study design; in the collection,nalysis, and interpretation of data; in the writing of the report;nd in the decision to submit the paper for publication.
cknowledgements
Arja Moilanen (previously Tamminen), Minna Voutilainen, Ullaaukkarinen, Janne Korkalainen, and Markku Forsman are thanked
or excellent technical assistance with the studies and the wholetaff of Laboratory Animal Unit of the Department of Environmentalealth, National Institute for Health and Welfare is acknowledged
or their collaboration. M.Sc. Pekka Tiittanen is thanked for statis-ical advice.
eferences
ell, D.R., Clode, S., Fan, M.Q., Fernandes, A., Foster, P.M.D., Jiang, T., Loizou, G., Mac-Nicoll, A., Miller, B.G., Rose, M., Tran, L., White, S., 2010. Interpretation of studieson the developmental reproductive toxicology of 2,3,7,8-tetrachlorodibenzo-p-dioxin in male offspring. Food Chem. Toxicol. 48, 1439–1447.
ock, K.W., Köhle, C., 2009. The mammalian aryl hydrocarbon (Ah) receptor: frommediator of dioxin toxicity toward physiological functions in skin and liver. Biol.Chem. 390, 1225–1235.
onhoffer, S., 1986. Body size and metabolic rate: exponent and coefficient of theallometric equation. The role of units. J. Theor. Biol. 119, 125–137.
eldman, H.A., McMahon, T.A., 1983. The mass exponent for energy metabolism isnot a statistical artifact. Respir. Physiol. 52, 149–163.
ernandez-Salguero, P.M., Hilbert, D.M., Rudikoff, S., Ward, J.M., Gonzalez, F.J.,1996. Aryl-hydrocarbon receptor-deficient mice are resistant to 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced toxicity. Toxicol. Appl. Pharmacol. 140,173–179.
ray, L.E., Ostby, J.S., Kelce, W.R., 1997. A dose–response analysis of the reproduc-tive effects of a single gestational dose of 2,3,7,8-tetrachlorodibenzo-p-dioxinin male Long Evans Hooded rat offspring. Toxicol. Appl. Pharmacol. 146, 11–20.
u, Y.Z., Hogenesch, J.B., Bradfield, C.A., 2000. The PAS superfamily: sensors ofenvironmental and developmental signals. Annu. Rev. Pharmacol. Toxicol. 40,519–561.
art, B.L., 1988. Biological basis of the behavior of sick animals. Neurosci. Biobehav.Rev. 12, 123–137.
attainen, H., Tuukkanen, J., Simanainen, U., Tuomisto, J.T., Kovero, O., Lukinmaa,P.L., Alaluusua, S., Tuomisto, J., Viluksela, M., 2001. In utero/lactational 2,3,7,8-tetrachlorodibenzo-p-dioxin exposure impairs molar tooth development in rats.Toxicol. Appl. Pharmacol. 174, 216–224.
azlauskas, A., Poellinger, L., Pongratz, I., 2000. The immunophilin-like protein XAP2regulates ubiquitination and subcellular localization of the dioxin receptor. J.Biol. Chem. 275, 41317–41324.
indén, J., Lensu, S., Tuomisto, J., Pohjanvirta, R., 2010. Dioxins, the aryl hydrocarbonreceptor and the central regulation of energy balance. Front. Neuroendocrinol.31, 452–478.
ensu, S., Tuomisto, J.T., Tuomisto, J., Pohjanvirta, R., 2011. Characterization of the2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)-provoked strong and rapid aver-sion to unfamiliar foodstuffs in rats. Toxicol., doi:10.1016/j.tox.2011.03.007.
ably, T.A., Bjerke, D.L., Moore, R.W., Gendron-Fitzpatrick, A., Peterson, R.E., 1992a.In utero and lactational exposure of male rats to 2,3,7,8-tetrachlorodibenzo-p-dioxin. 3. Effects on spermatogenesis and reproductive capability. Toxicol. Appl.Pharmacol. 114, 118–126.
ably, T.A., Moore, R.W., Goy, R.W., Peterson, R.E., 1992b. In utero and lactationalexposure of male rats to 2,3,7,8-tetrachlorodibenzo-p-dioxin. 2. Effects on sex-
ual behavior and the regulation of luteinizing hormone secretion in adulthood.Toxicol. Appl. Pharmacol. 114, 108–117.ably, T.A., Moore, R.W., Peterson, R.E., 1992c. In utero and lactational exposure ofmale rats to 2,3,7,8-tetrachlorodibenzo-p-dioxin. 1. Effects on androgenic status.Toxicol. Appl. Pharmacol. 114, 97–107.
ers 203 (2011) 252–257 257
McIntosh, B.E., Hogenesch, J.B., Bradfield, C.A., 2010. Mammalian Per-Arnt-Sim pro-teins in environmental adaptation. Annu. Rev. Physiol. 72, 625–645.
Mimura, J., Ema, M., Sogawa, K., Fujii-Kuriyama, Y., 1999. Identification of a novelmechanism of regulation of Ah (dioxin) receptor function. Genes Dev. 13, 20–25.
Mimura, J., Yamashita, K., Nakamura, K., Morita, M., Takagi, T.N., Nakao, K., Ema, M.,Sogawa, K., Yasuda, M., Katsuki, M., Fujii-Kuriyama, Y., 1997. Loss of teratogenicresponse to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in mice lacking the Ah(dioxin) receptor. Genes Cells 2, 645–654.
Pohjanvirta, R., Miettinen, H.M., Sankari, S., Lindén, J. Unexpected gender differencein sensitivity to the acute toxicity of dioxin in mice, submitted for publication.
Pohjanvirta, R., Tuomisto, J., 1987. Han/Wistar rats are exceptionally resistant toTCDD. II. Arch. Toxicol. Suppl. 11, 344–347.
Pohjanvirta, R., Tuomisto, J., 1990. 2,3,7,8-Tetrachlorodibenzo-p-dioxin enhancesresponsiveness to post-ingestive satiety signals. Toxicology 63, 285–299.
Pohjanvirta, R., Tuomisto, J., 1994. Short-term toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in laboratory animals: effects, mechanisms, and animal models.Pharmacol. Rev. 46, 483–549.
Pohjanvirta, R., Tuomisto, J., Vartiainen, T., Rozman, K., 1987. Han/Wistar rats areexceptionally resistant to TCDD. I. Pharmacol. Toxicol. 60, 145–150.
Pohjanvirta, R., Unkila, M., Tuomisto, J., 1993. Comparative acute lethality of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), 1,2,3,7,8-pentachlorodibenzo-p-dioxinand 1,2,3,4,7,8-hexachlorodibenzo-p-dioxin in the most TCDD-susceptible andthe most TCDD-resistant rat strain. Pharmacol. Toxicol. 73, 52–56.
Pohjanvirta, R., Unkila, M., Tuomisto, J., 1994. TCDD-induced hypophagia is notexplained by nausea. Pharmacol. Biochem. Behav. 47, 273–282.
Pohjanvirta, R., Viluksela, M., Tuomisto, J.T., Unkila, M., Karasinska, J., Franc, M.A.,Holowenko, M., Giannone, J.V., Harper, P.A., Tuomisto, J., Okey, A.B., 1999. Physic-ochemical differences in the AH receptors of the most TCDD-susceptible and themost TCDD-resistant rat strains. Toxicol. Appl. Pharmacol. 155, 82–95.
Pohjanvirta, R., Wong, J.M., Li, W., Harper, P.A., Tuomisto, J., Okey, A.B., 1998. Pointmutation in intron sequence causes altered carboxyl-terminal structure in thearyl hydrocarbon receptor of the most 2,3,7,8-tetrachlorodibenzo-p-dioxin-resistant rat strain. Mol. Pharmacol. 54, 86–93.
Reilly, S., Bornovalova, M.A., 2005. Conditioned taste aversion and amygdala lesionsin the rat: a critical review. Neurosci. Biobehav. Rev. 29, 1067–1088.
Schmidt, J.V., Su, G.H., Reddy, J.K., Simon, M.C., Bradfield, C.A., 1996. Characterizationof a murine Ahr null allele: involvement of the Ah receptor in hepatic growthand development. Proc. Natl. Acad. Sci. U.S.A. 93, 6731–6736.
Simanainen, U., Tuomisto, J.T., Tuomisto, J., Viluksela, M., 2002. Structure-activity relationships and dose responses of polychlorinated dibenzo-p-dioxinsfor short-term effects in 2,3,7,8-tetrachlorodibenzo-p-dioxin-resistant and -sensitive rat strains. Toxicol. Appl. Pharmacol. 181, 38–47.
Simanainen, U., Tuomisto, J.T., Tuomisto, J., Viluksela, M., 2003. Dose–responseanalysis of short-term effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin inthree differentially susceptible rat lines. Toxicol. Appl. Pharmacol. 187,128–136.
Tuomisto, J.T., Viluksela, M., Pohjanvirta, R., Tuomisto, J., 1999. The AH receptorand a novel gene determine acute toxic responses to TCDD: segregation of theresistant alleles to different rat lines. Toxicol. Appl. Pharmacol. 155, 71–81.
Tuomisto, J.T., Viluksela, M., Pohjanvirta, R., Tuomisto, J., 2000. Changes in foodintake and food selection in rats after 2,3,7, 8-tetrachlorodibenzo-p-dioxin(TCDD) exposure. Pharmacol. Biochem. Behav. 65, 381–387.
Unkila, M., Pohjanvirta, R., MacDonald, E., Tuomisto, J.T., Tuomisto, J., 1994. Doseresponse and time course of alterations in tryptophan metabolism by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the most TCDD-susceptible and the mostTCDD-resistant rat strain: relationship with TCDD lethality. Toxicol. Appl. Phar-macol. 128, 280–292.
Walker, M.K., Cook, P.M., Butterworth, B.C., Zabel, E.W., Peterson, R.E., 1996. Potencyof a complex mixture of polychlorinated dibenzo-p-dioxin, dibenzofuran, andbiphenyl congeners compared to 2,3,7,8-tetrachlorodibenzo-p-dioxin in caus-ing fish early life stage mortality. Fundam. Appl. Toxicol. 30, 178–186.
White, S.S., Birnbaum, L.S., 2009. An overview of the effects of dioxins and dioxin-likecompounds on vertebrates, as documented in human and ecological epi-demiology. J. Environ. Sci. Health C: Environ. Carcinog. Ecotoxicol. Rev. 27,
Whitlock Jr., J.P., 1990. Genetic and molecular aspects of 2,3,7,8-tetrachlorodibenzo-p-dioxin action. Annu. Rev. Pharmacol. Toxicol. 30, 251–277.
Zeger, S.L., Liang, K.Y., 1986. Longitudinal data-analysis for discrete and continuousoutcomes. Biometrics 42, 121–130.
