Food and Chemical Toxicology 59 (2013) 534–540
Contents lists available at SciVerse ScienceDirect
Food and Chemical Toxicology
journal homepage: www.elsevier.com/locate/foodchemtox
Triclosan exposure reduces thyroxine levels in pregnant and lactating rat dams and in directly exposed offspring ⇑ Marta Axelstad , Julie Boberg, Anne Marie Vinggaard, Sofie Christiansen, Ulla Hass
National Food Institute, Technical University of Denmark, Division of Toxicology and Risk Assessment, Mørkhøj Bygade 19, DK-2860 Søborg, Denmark article info abstract
Article history: Thyroid disrupting chemicals can potentially disrupt brain development. Two studies investigating the Received 18 April 2013 effect of the antibacterial compound triclosan on thyroxine (T4) levels in rats are reported. In the first, Accepted 25 June 2013 Wistar rat dams were gavaged with 75, 150 or 300 mg triclosan/kg bw/day throughout gestation and lac- Available online 4 July 2013 tation. Total T4 serum levels were measured in dams and offspring, and all doses of triclosan significantly
lowered T4 in dams, but no significant effects on T4 levels were seen in the offspring at the end of the lac- Keywords: tation period. Since this lack of effect could be due to minimal exposure through maternal milk, a second Triclosan study using direct per oral pup exposure from postnatal day 3–16 to 50 or 150 mg triclosan/kg bw/day Rat was performed. This exposure pointed to significant T4 reductions in 16 day old offspring in both dose Thyroxine (T4) Developmental groups. These results corroborate previous studies showing that in rats lactational transfer of triclosan Thyroid disrupting chemical (TDC) seems limited. Since an optimal study design for testing potential developmental neurotoxicants in rats, should include exposure during both the pre- and postnatal periods of brain development, we suggest that in the case of triclosan, direct dosing of pups may be the best way to obtain that goal. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction cognitive and motor function have been reported in numerous epi- demiological studies (Ghassabian et al., 2011; Haddow et al., 1999; Triclosan is a compound that has been used in consumer prod- Henrichs et al., 2010; Kooistra et al., 2006; Li et al., 2010; Pop et al., ucts for many years due to its high efficiency as an antibacterial 1999, 2003). In addition, an ample number of animal studies have agent and its low acute toxicity (SCCS, 2009). In the EU, about shown that prenatal hypothyroxinemia may significantly affect 85% of the total volume of triclosan is used in personal care prod- nerve cell migration and other molecular aspects of brain develop- ucts (toothpaste, soap, shampoo and cosmetics), 5% is used for tex- ment in rats (Auso et al., 2004; Berbel et al., 2010; Cuevas et al., tiles (antibacterial clothing) and 10% for plastics and food contact 2005; Gilbert and Sui, 2008, 2006; Lavado-Autric et al., 2003; Opa- materials (SCCS, 2009). In recent years it has become evident that zo et al., 2008; Sharlin et al., 2008). Our previous research has indi- triclosan acts as a thyroid hormone disrupting chemical. This effect cated that in rats, maternal hypothyroxinemia is only significantly has been seen in metamorphosis studies in bullfrogs and Xenopus correlated to altered behaviour and hearing in the offspring, if laevis (Helbing et al., 2011; Veldhoen et al., 2006), and the com- marked postnatal T4 decreases in the offspring are also present. pound has also been shown to reduce thyroxine (T4) levels in a ser- The fact that postnatal thyroid hormone insufficiency has been ies of rat studies. In these studies, young adult rats (22–60 days of present in almost all studies showing altered behaviour after age) have been exposed to triclosan for varying periods of time developmental hypothyroidism in rats (Akaike et al., 1991; Brosvic (Crofton et al., 2007; Paul et al., 2010a; Stoker et al., 2010; Zorrilla et al., 2002; Noda et al., 2005; Provost et al., 1999) further corrob- et al., 2009), or rat dams have been exposed during gestation and orate the hypothesis that in rat dams, T4 reductions during gesta- lactation (Paul et al., 2012, 2010b; Rodriguez and Sanchez, 2010). tion alone are not enough to induce adverse behavioural effects
The doses needed to observe T4 reductions in these studies have in the offspring (Axelstad et al., 2011a,b). Such an important spe- been varying between 10 and 300 mg/kg bw/day, probably reflect- cies difference between humans and rats may beexplained by the ing variations in rat strain and study design. fact that while both rodents and humans need thyroid hormones
In humans, even mild reductions in T4 levels in pregnant for differentiation and maturation of the central nervous system, women can have severe consequences for the neurological devel- important differences in timing of these events exit. Since a larger opment of children, as associations with delayed or impaired part of the brain maturation takes place prenatally in humans com-
pared to rats (Howdeshell, 2002), prenatal T4 deficits might have ⇑ Corresponding author. Tel.: +45 35 88 75 41. more severe consequences for foetal brain development in humans E-mail address: [email protected] (M. Axelstad). than in rats.
0278-6915/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.fct.2013.06.050 M. Axelstad et al. / Food and Chemical Toxicology 59 (2013) 534–540 535
The aim of the present work was to develop a study design in On each day of dosing, the entire litter (8 pups) was weighed, and the average pup weight was calculated. All pups in the litter were dosed according to the aver- which postnatal T4 reductions would be induced in rat offspring. age weight receiving 2 ll of test solution per gram body weight. Thus they received For this purpose two studies were performed. In the first, pregnant from approximately 12 llto90ll test solution per day in the dosing period. On rat dams were exposed to triclosan during gestation and lactation, PND 16 the offspring were sacrificed by decapitation and trunk blood was collected and T4 levels in dams and offspring were examined. Due to a sus- for total T4 analyses. The trunk blood was collected and analyzed individually for pected influence of triclosan on the sex hormone balance (Kumar each pup. et al., 2009; Stoker et al., 2010), also prostate glands were exam- ined in the offspring. In spite of marked reductions in maternal 2.3. Thyroxine immunoassay T levels during gestation and lactation, offspring T levels were 4 4 Blood samples from dams (GD 15 and PND 16) and offspring (PND 16) in the not significantly reduced when measured at the end of the lacta- first study, and from offspring (PND 16) in the second study, were analyzed for total tion period. As this result indicated limited triclosan excretion to thyroxine (T4) concentration in plasma. Trunk blood was used for the analysis. A the maternal milk, and consequently limited exposure in the Delfia time-resolved fluoroimmunoassay kit from Perkin Elmer (cat. No. 1244- neonatal period, a second study was performed. Here, neonatal 030, Wallac Oy, Turku, Finland) was modified and developed specifically for analy- sis of T in rat samples. Instead of human T standards and T antibody supplied in rat offspring were exposed to triclosan through direct peroral 4 4 4 the Delfia kits, T4 standards in T4-free rat serum (cat. Nos. 30042 and 30041, respec- dosing during lactation, with subsequent measurements of serum tively) as well as rat specific biotinylated T4 (30,039) antibody from Biovian Ltd. T4 levels. (Finland) were used. Otherwise the assay was run as outlined in the manufacturer’s protocol and described in Axelstad et al. (2008). The analytical sensitivity is around
10 nM. Historic control values for T4 levels in PND 16 males and females were re- corded being 40 ± 13 nM and 38 ± 8 nM (n = 5), respectively. Thus, the CV% between 2. Materials and methods in vivo studies (n = 5) were 32% and 22% for males and females, respectively.
2.1. Chemicals 2.4. Statistical analysis The vehicle used was corn oil (Sigma–Aldrich, Denmark). Triclosan (purity Statistical analysis of data with normal distribution and homogeneity of vari- >99.0%, CAS No. 3380-34-5, Alfa Aesar No. L18655) was from VWR-Bie & Berntsen, ance were analysed using analysis of variance (ANOVA), followed by Dunnett’s post Herlev, Denmark. The triclosan solutions were kept dark, at room temperature, and hoc test. In study 1, when more than one pup from each litter was examined, sta- continuously stirred during the dosing period. New solutions were prepared for tistical analyses were adjusted using litter as an independent, random and nested each of the two studies, but no verification of dose concentrations was performed. factor in ANOVA, or analysis were done using litter means. Furthermore, since study 1 was performed in two blocks, block was included as an independent random and nested factor in the analysis however no significant effects of block were seen. In 2.2. Animals and treatment cases where normal distribution and homogeneity of variance could not be ob- tained by data transformation, a non-parametric Kruskal–Wallis test was used. Both studies were performed under conditions approved by the Danish Animal Trend analysis on dose–response relations for hormone levels, body- and organ Experiments Inspectorate and by the in-house Animal Welfare Committee. The ani- weights were performed using Spearman’s test. In study 2, litter was not used as mals received a complete rodent diet (Altromin Standard Diet 1314) and acidified the statistical unit, as triclosan exposure was direct, and not through the dam. tap water ad libitum, and were housed under standard conditions as described in Christiansen et al. (2012). In the first study (study 1), 40 time-mated Wistar rats (HanTac:WH, Taconic 3. Results Europe, Ejby, Denmark) were supplied at gestation day (GD) 3 of pregnancy. The study was performed using 2 blocks with one week in between, and all dose groups In study 1, maternal body weight gain was not significantly were equally represented in the blocks. The dams were distributed into four dose groups (0; 75; 150 or 300 mg/kg/day; n = 10 per group), and gavaged once daily affected by triclosan exposure when weight gain was calculated from gestation day (GD) 7 to postnatal day (PND) 16 (day of delivery excluded), from the beginning of the dosing period to the day before birth at a constant volume of 2 ml/kg bw/day. The individual doses were based on the (GD7 to 21). However, when maternal body weight gain was mea- body weight of the animal on the day of dosing. The dams were inspected twice sured from GD7 to the day after birth (PND 1), a significant dose- a day for general toxicity including changes in clinical appearance. Body weights dependent downward trend was seen (p = 0.001) and a statistically were recorded on GD 4 and daily during the dosing period. On GD 15 dams were anesthetized with HypnormÒ (fentanyl citrate/flunisone)/DormicumÒ (midazolam), significant decrease was seen in the highest dose group compared and blood was drawn from the tail vein. The day after delivery, the pups were to control (p = 0.007), indicating that 300 mg triclosan/kg/day counted, sexed, weighed, checked for anomalies and anogenital distance (AGD). caused a moderate degree of maternal toxicity during gestation Of the 10 time-mated dams in each group, 7–9 were pregnant and gave birth to via- (Table 1). Gestation length, gender distribution, postimplantation ble litters. Body weights were measured on PND 6 and 13, and offspring were exam- loss and litter size were unaffected by the exposure and so were ined for the presence of nipples/areolas on PND 13. On PND 16 all dams and pups were sacrificed. Dams were weighed, anaesthetized in CO2/O2, decapitated, and maternal body weight gains, neonatal deaths and offspring body trunk blood was collected. The number of implantation scars in the uterus was reg- weights in the postnatal period. Furthermore, no effect was seen istered and thyroid glands were excised and weighed. All offspring were weighed, on male or female anogenital distance or on nipple retention decapitated and trunk blood collected. Blood samples from the offspring were (Table 1). pooled within litter in a male and a female sample. Prostates from 1 male per litter were excised, weighed and prepared for histopathological examinations. Thyroids Triclosan exposure had a marked effect on total T4 levels in dam from 1 to 2 males per litter were excised and weighed. Thyroids from 1 male per serum. On GD 15, a significant dose-dependent downwards trend litter were excised together with the thyroid cartilage and prepared for histological was seen (p < 0.001) and the T4 levels were decreased by 59%, examination. Thyroids and prostates were fixed in formalin, embedded in paraffin, 72%, and 72% in the three triclosan groups respectively and stained with haematoxylin and eosin, and histological evaluations of one sec- (p = 0.028, p = 0.0001, p = 0.0005) (Fig. 1A and Table 2). The effect tion per organ from animals from the control group and the highest dose group were performed by a pathologist blinded to treatment groups. of triclosan on maternal T4 levels was also present on PND 16, In the direct postnatal exposure study (study 2), 6 time-mated pregnant Wistar where the dose-dependent downwards trend was also significant rats were supplied at GD 16, one week before expected delivery. Each dam was indi- (p < 0.001) and the T4 levels were decreased by 38%, 55% and 58% vidually housed, and each litter remained housed with the respective dam until sac- in the three triclosan groups respectively (p = 0.032, p = 0.001, rifice. Two days after delivery (PND 3) the litters were culled to 8 offspring, with an equal representation of males and females when possible, and litters were assigned p = 0.0006) (Fig. 1B and Table 2). Offspring total T4 levels on PND into one of three dosage groups (0, 50 or 150 mg/kg/day). The maximum dose was 16 (study 1) showed no statistically significant dose-dependent set at 150 mg/kg/day to avoid general toxic effects in the young pups. All offspring trends and group means did not differ significantly from controls were dosed orally from PND 3 to PND 16 using a micropipette. The dams were not in any dose group (Table 2). dosed, but remained in the cage to allow the pups to feed normally. Each pup was Both absolute and relative thyroid gland weights were unaf- held by one hand while the other slowly allowed test solution to drip directly into the mouth, but only as fast as the pup would allow, ensuring precise and reliable fected by triclosan exposure in dams and offspring, as no dose- dosing. dependent trends and no significant differences between groups 536 M. Axelstad et al. / Food and Chemical Toxicology 59 (2013) 534–540
Table 1 Reproductive and developmental data from study 1 and 2. The table shows pregnancy and litter data, including body weights (bw) from dams and offspring exposed indirectly to 0, 75, 150 or 300 mg triclosan/kg bw/day from GD 7 to PND 16 (study 1), and offspring body weights in pups exposed directly to triclosan at doses of 0, 50 or 150 mg/kg bw/day from PND 3–16 (study 2). Data represent group means based on litter means ± SD.
Study 1 (indirect dosing of pups) Control 75 mg/kg/day 150 mg/kg/day 300 mg/kg/day Dams and litters n =9 n =7 n =8 n =8 Dam bw gain, GD 7-GD 21 80.8 ± 8.3 78.6 ± 14 77.6 ± 16 72 .8 ± 16 Dam bw gain, GD 7-PND 1 12.9 ± 5.5 15.1 ± 7.0 8.7 ± 7.1 2.9 ± 5.2** Dam bw gain, PND 1–16 39.3 ± 10.0 36.4 ± 5.2 33.8 ± 10.8 38.3 ± 12.8 Gestation length (days) 22.89 ± 0.33 23.14 ± 0.38 23.00 ± 0.0 23.00 ± 0.0 % Postimplantation loss 8.19 ± 6.7 28.4 ± 41.7 19.5 ± 32.2 22.4 ± 41.2 % Perinatal loss 12.5 ± 7.2 35.5 ± 40.6 19.5 ± 32.2 22.4 ± 41.2 Litter size 10.67 ± 2.7 9.00 ± 3.8 11.00 ± 3.9 9.67 ± 4.8 % Perinatal deaths 4.67 ± 4.5 6.59 ± 17.4 0.00 ± 0.0 0.00 ± 0.0 % Males 52.6 ± 21.2 45.6 ± 8.9 48.8 ± 12.3 59.5 ± 22.7 Offspring Mean birth weight 6.23 ± 0.5 6.12 ± 0.2 6.08 ± 0.4 6.16 ± 0.4 AGD males (mm) 4.03 ± 0.1 3.98 ± 0.1 4.00 ± 0.1 3.97 ± 0.1 AGD females (mm) 2.21 ± 0.2 2.14 ± 0.1 2.19 ± 0.1 2.19 ± 0.1 Nipples males 0.00 ± 0.0 0.00 ± 0.0 0.12 ± 0.2 0.06 ± 0.1 Nipples females 12.4 ± 0.3 12.3 ± 0.2 12.2 ± 0.2 12.4 ± 0.2 Mean pup bw PND 6 13.11 ± 1.5 12.62 ± 0.6 11.83 ± 1.5 12.19 ± 1.6 Mean pup bw PND13 26.31 ± 4.0 26.06 ± 1.5 23.93 ± 3.9 24.42 ± 3.7 Mean pup bw PND 16 31.72 ± 5.3 31.84 ± 2.4 28.15 ± 4.8 29.65 ± 4.9 Study 2 (direct dosing of pups) Control 50 mg/kg/day 150 mg/kg/day No. of litters n = 2 (1 after day 7) n =2 n =2 Litter size (culled to 8 at PND 3) 8 8 8 and 6 Mean pup bw PND 6 12.1 ± 0.5 13.3 ± 0.8 13.8 ± 0.5 Mean pup bw PND13 26.29 ± 0.0 29.95 ± 2.4 32.44 ± 2.2 Mean pup bw PND 16 32.58 ± 0.0 37.16 ± 2.9 39.91 ± 3.4
** p < 0.01.
were seen (Table 3). Also no histopathological effects on the off- spring thyroids on PND 16 were seen in the high dose group (data not shown). Statistical analysis of the both relative and absolute prostate weights (analysed with body weight as covari- ate) indicated no differences between treatment groups, and no dose-dependent trends were seen (Table 3). Histopathology of the prostates showed no differences between controls and the high dose group at PND 16 (data not shown). In the direct exposure study (study 2) one of the two litters in the control group had to be sacrificed on PND 7, because the dam did not take care of her offspring. Furthermore, two pups from one of the litters in the high dose group died on day PND 6. This is sometimes seen in rat litters, and was not attributed to triclosan exposure. The triclosan dosing did not cause any general toxicity effects in the exposed offspring, and no significant effects on pup body weights or body weight gains were seen during the exposure period (PND 3–16), compared to controls (Table 1). On days 13 and 16, the pups in the direct exposure study (study 2) weighed more than offspring of the same age in the indirect exposure study, and the differences were statistically significant (p = 0.0051 and p = 0.0037 on PND 13 and 16 respectively). This difference was probably due to a combination of the pups receiving oil gavage and the fact that there were fewer offspring in this study in each litter, because of the culling performed on day 3.
T4 levels in the pups that had been directly exposed to triclosan at doses of 50 and 150 mg/kg/day were significantly decreased by 16% (p = 0.029) and 39% (p < 0.001) in the two dose groups respec- tively, when measured on PND 16 (Table 2), and a significant dose dependent trend was also seen (p < 0.001). However, an important limitation in study 2 was that all the control pups were from the same litter. Therefore there was a possible risk that due to genetic
similarities, the offspring may have had high T4 levels, which could have caused the significant T4 reductions in both dose groups. To Fig. 1. Total thyroxine (T ) levels (nM) in dams on gestation day (GD) 15 (A) and 4 further complicate interpretation of the data, the T values from postnatal day (PND) 16 (B), after exposure to 0, 75, 150 or 300 mg triclosan/kg bw/ 4 day from GD 7 – PND 16 (study 1). Data represent group means based on litter the pups in this litter were in the high end of our previous control means + SEM, n = 7–9. *p < 0.05; ***p < 0.001. values (Axelstad et al., 2008, 2011a,b). M. Axelstad et al. / Food and Chemical Toxicology 59 (2013) 534–540 537
Table 2
T4 levels in dams and offspring from study 1 and 2. The table shows total T4 levels (nM) in dams and offspring, after exposure to 0, 75, 150 or 300 mg triclosan/kg bw/day from GD7 to PND 16 in study 1, and in pups exposed directly to triclosan at doses of 0, 50 or 150 mg/kg/day from PND 3 to 16 in study 2. Data represent group means ± SD.
Study 1 (indirect dosing of pups) Control 75 mg/kg/day 150 mg/kg/day 300 mg/kg/day No. of litters 9 7 8 8 * *** *** T4 in dams GD 15 38.7 ± 20.4 15.8 ± 5.4 10.7 ± 7.7 10.8 ± 4.5 * *** *** T4 in dams PND 16 17.5 ± 6.9 10.9 ± 4.7 7.9 ± 3.7 7.4 ± 2.7
T4 in offs. PND 16 (males) 34.9 ± 9.2 26.0 ± 4.0 34.9 ± 9.9 27.4 ± 8.1
T4 in offs. PND 16 (females) 40.9 ± 6.3 35.4 ± 5.2 38.5 ± 13.8 33.6 ± 5.2 Study 2 (direct dosing of pups) Control 50 mg/kg/day 150 mg/kg/day No. of samples 8 (1 l) 16 (2 l) 14 (2 l) * *** T4 in offs. PND 16 (males) 45.5 ± 6.8 38.0 ± 5.0 27.7 ± 2.7 * *** T4 in offs. PND 16 (females) 46.6 ± 5.4 39.2 ± 4.9 28.3 ± 3.7
* p < 0.05. *** p < 0.001.
Table 3 Dam and offspring organ weights from study 1. The table shows absolute and relative thyroid and prostate weights in dams and male pups exposed indirectly to 0, 75, 150 or 300 mg triclosan/kg bw/day from GD7 to PND 16 (study 1). Data represent group means ± SD. No significant differences between dose groups were observed.
Study 1 Control 75 mg/kg/day 150 mg/kg/day 300 mg/kg/day Dams Dam BW, PND 16 282 ± 25 282 ± 11 266 ± 13 272 ± 12 Thyroid weight dams PND 16 0.017 ± 0.003 0.021 ± 0.011 0.020 ± 0.007 0.016 ± 0.003 Rel. thyroid weight/100 g bw 0.591 ± 0.09 0.746 ± 0.37 0.741 ± 0.29 0.576 ± 0.12 Offspring Mean male pup bw PND 16 31.92 ± 5.1 31.32 ± 3.3 30.70 ± 5.0 30.03 ± 4.7 Thyroid weight PND 16 0.47 ± 0.05 0.48 ± 0.06 0.40 ± 0.06 0.45 ± 0.06 Rel. thyroid weight/100 g bw 1.50 ± 0.16 1.52 ± 0.11 1.30 ± 0.15 1.54 ± 0.26 Prostate weight PND 16 1.19 ± 0.2 1.31 ± 0.4 1.19 ± 0.3 1.01 ± 0.3 Rel. prostate weight/100 g bw 0.038 ± 0.005 0.041 ± 0.01 0.039 ± 0.007 0.034 ± 0.009
4. Discussion on GD 20 and PND 4, while no significant reductions were seen on PND 14 or PND 21. The authors suggested that the T4 reductions The present studies confirm that triclosan can disrupt thyroid observed on PND 4 pups could have resulted from transplacental hormone levels, as total T4 levels were markedly decreased in rat exposure to triclosan and that toxicokinetic factors probably af- dams on both GD15 and PND 16 after exposure to 75; 150 and fected maternal transfer of triclosan into milk and thereby limited 300 mg triclosan/kg bw/day during gestation and lactation. After lactation exposure to the pups. The present T4 results (study 1) fit indirect exposure through placenta and maternal milk (study 1), well with results from Paul et al. (2010b, 2012), showing no signif- the offspring T4 levels on PND 16 were not significantly decreased, icant effects on offspring T4 levels after approximately two weeks whereas direct pup exposure to 50 and 150 mg/kg bw/day from of lactational exposure. To further test whether the lack of T4
PND 3–16 (study 2) appeared to decrease offspring T4 levels. Since reduction in the 16-day old offspring was due to limited triclosan previous studies have indicated that triclosan does not pass to the excretion in the milk, the direct exposure study (study 2) was per- milk in sufficient amounts to significantly reduce T4 levels during formed. Due to the quite time consuming dosing method, this the late lactation period (Paul et al., 2012), and the results from study was only performed on a limited number of litters. However, study 1 corroborate this finding, the present results taken together the blood samples from each of the eight dosed pups in each litter indicate that triclosan can probably only significantly reduce T4 were not pooled, yielding between 8 and 16 samples in each dose levels in the offspring during the entire lactation period, if the group. The conclusion in study 2 was that since T4 levels in the di- chemical is directly dosed. rectly exposed pups seemed decreased compared to controls the
T4 levels have previously been measured in a number of triclo- lack of significant effect seen in any dose group in the offspring san studies in rats, and the present results corroborate and add to from study 1, was probably due to limited exposure through these findings. In female Long-Evans (LE) rats, dosed daily with tri- maternal milk. In future developmental studies of triclosan, includ- closan either from PND 28–31 or during gestation and lactation, ing direct measurements of triclosan levels in the dam’s milk at dif- doses of 100 mg/kg bw/day and above significantly lowered serum ferent post-natal times, as well as performing a larger study with levels of T4 in the dams (Crofton et al., 2007; Paul et al., 2010a, direct postnatal dosing of the offspring, would more definitively 2012). A number of studies performed in Wistar rats have further- determine if the different sensitivities observed between neonatal more shown that even lower doses of triclosan can induce T4 and 14–16 day old pups, are caused by changes in exposure to tri- reductions in this strain. Here doses of 30 mg/kg bw/day and above closan or by changes in triclosan catabolism. caused T4 levels to decrease markedly in young Wistar males dosed It seemed quite clear that the T4 reductions seen in dams dosed daily from PND 23–58 (Zorrilla et al., 2009), whereas doses of with the 75 and 150 mg/kg from GD7-15 (59 and 72% reduction in 37.5 mg/kg bw/day and above resulted in decreased total serum T4, respectively) were more marked than seen in pups dosed di-
T4 levels in young Wistar females dosed from PND 22–43 (Stoker rectly from PND 3–16 with 50 and 150 mg/kg (16% and 39% reduc- et al., 2010). tion in T4 respectively), indicating that triclosan also may not have In developmental studies performed by Paul et al. (2010b, triggered the same degree of toxicodynamic effects in the offspring
2012), 300 mg/kg caused serum T4 levels in offspring to decrease as seen in more mature animals. This could be explained by the 538 M. Axelstad et al. / Food and Chemical Toxicology 59 (2013) 534–540 mode of action for triclosan, as the compound is suspected of et al., 2008), and taken together the in vivo and in vitro data indi- affecting the thyroid hormone system by causing induction of cate that triclosan could affect the reproductive hormone axis. phase II liver enzymes (sulfonation or glucuronidation), and there- The lack of adverse effects in the few reproductive endpoints tested by upregulating thyroid hormone catabolism (Crofton et al., 2007; in the present study (AGD, nipple retention, prostate weight and Paul et al., 2010a, 2012). This mode of action is indicated by the histology) could reflect that these endpoints are more sensitive observation of increased liver weights (Crofton et al., 2007; Paul to anti-androgenic chemicals, and it is possible that triclosan et al., 2010a; Zorrilla et al., 2009), increased PROD activity in the is too weak an androgen receptor antagonist to cause anyanti- liver (a marker of Cyp2b activity) (Paul et al., 2010b, 2012; Zorrilla androgenic effects in vivo. et al., 2009) and upregulated mRNA expression and activity of Based on what is presently known on the effects of triclosan, an some phase I and phase II hepatic enzymes (Paul et al., 2010a, adverse outcome pathway for the effects of triclosan on the thyroid 2012). Since hepatic excretory function in rats develops postnatally hormone system has been proposed (US EPA, 2011). Here activa- to reach maximum capacity at an age of approximately 30 days tion of the pregnane X receptor (PXR) and/or the constitutive (Klinger, 2005), and due to this reduced activity of many phase II androstane receptor (CAR) in rat liver by triclosan is an initiating enzymes, neonatal rats have a decreased capacity to metabolize event, leading to up regulation of hepatic phase I and phase II en- and excrete (Suchy et al., 2007). A possible explanation for the zymes, which could increase catabolism of thyroid hormones and smaller T4 reductions seen in the pups compared to adults might consequently lower serum levels of these – effects that could be that triclosan exposure did not lead to the same degree of induc- potentially lead to altered neurodevelopment. Furthermore, triclo- tion of liver enzymes as seen in adult animals, and consequently san has recently been shown also to affect the thyroid system by only induced smaller T4 reductions. However, this is only one inhibiting deoidinase activity (Butt et al., 2011). Uncertainty exists, explanation for the observed results and more studies are needed as to whether this adverse outcome pathway is also relevant for to confirm the hypothesis that thyroid disrupting compounds act- humans, because species differences with regard to activation ing by increasing liver metabolism may not affect thyroid hormone and amino acid sequence of the PXR exist (Jones et al., 2000), which levels in neonatal rats as much as seen in older animals. make it difficult to extrapolate results obtained with rodent PXR In the present study no thyroid weight increases were observed directly to humans (Dybdahl et al., 2012). However in vitro data in dams, despite a considerable drop in T4 levels. Normally thyroid show that triclosan is a moderate inducer of human PXR activity weight increases are considered mediated through an increase of in human hepatoma cells (Jacobs et al., 2005), indicating that this TSH, in response to decreased thyroid hormone levels. TSH levels could also be an initiating event in an adverse outcome pathway were however not measured, because previous triclosan studies in humans. have shown that TSH levels were not affected at dose levels where significant T4 decreases were seen (Zorrilla et al., 2009; Paul et al., 2010a). Histological examination of thyroids was only performed 5. Conclusions in pre- and postnatally exposed male offspring showing no changes, whereas no examination was performed in dams or in In conclusion, triclosan markedly lowered maternal T4 levels in pups exposed directly to triclosan. rat dams during gestation and lactation, and nine days of exposure In the present developmental toxicity study (study 1) neither resulted in a LOAEL of 75 mg/kg bw/day, corroborating effects seen anogenital distance, nipple retention, prostate weight nor prostate in previous rat studies. In humans, correct maternal T4 levels dur- histology were affected by exposure to triclosan. These endpoints ing pregnancy are crucial for fetal brain development, and even are typically affected by perinatal exposure to anti-androgenic slight maternal hypothyroxinemia can result in adverse effects chemicals (Christiansen et al., 2010; McIntyre et al., 2001), indicat- on the cognitive and motor function of children (Ghassabian ing that triclosan exposure at the tested dose levels did not affect et al., 2011; Henrichs et al., 2010; Kooistra et al., 2006; Li et al., male reproductive development in an anti-androgenic manner. Ad- 2010; Pop et al., 1999, 2003). Based on its thyroid disrupting prop- verse reproductive effects have previously been reported in males erties there might be a need for further assessment of triclosan as a rats exposed during adulthood (Kumar et al., 2009). These included potential developmental neurotoxicant. Since the first ten postna- decreased weights of several reproductive organs, histopatholo- tal days in the rat approximate the last trimester of human preg- gical changes in these, and decreased levels of FSH, LH and testos- nancy with regard to the development of the central nervous terone level after two months of exposure to 20 mg/kg bw/day. It is system (Howdeshell, 2002), the results presented here imply that however possible that the effects seen by Kumar et al., could be in a developmental toxicity study of triclosan, direct postnatal due to impurities in the triclosan used for the studies, as dioxin exposure would be the optimal study design to use, for successfully and furan contamination has been seen in several different triclo- covering the entire period of human brain development during san samples produced in India and China (Menoutis and Parisi, pregnancy. It is furthermore important to bear in mind that hu- 2002). In a study using post-weaning male rats exposed to up to mans are exposed to a variety of thyroid disrupters, all probably 300 mg/kg/day from PND 23–58, no significant effects on timing acting in a dose-additive manner (Flippin et al., 2009). Since triclo- of sexual maturation or reproductive organ weight were seen (Zor- san may be a potential contributor to thyroid disruption in hu- rilla et al., 2009). In studies focusing on reproductive effects in fe- mans, exposure should be carefully regulated in order to protect male offspring, a triclosan doses of 150 mg/kg/day lowered the age pregnant women and their children from excessive exposure to of vaginal opening and caused increased uterine weights (Stoker thyroid disrupting chemicals. et al., 2010), while much lower doses of triclosan potentiated the effects of estradiol treatment on uterine weight (Louis et al., Funding 2013; Stoker et al., 2010). These effects were compatible with an estrogenic mode of action, which was not tested in the present This work was supported financially by the Danish Environ- study. mental Protection Agency. A number of in vitro studies examining the endocrine disrupting modes of action of triclosan have also been performed during re- cent years. Antagonistic activity in both ER- and AR-responsive Conflict of Interest bioassays have been shown in several studies (Ahn et al., 2008; Chen et al., 2007; Gee et al., 2008; Christen et al., 2010; Vinggaard The authors declare that there are no conflicts of interest. M. Axelstad et al. / Food and Chemical Toxicology 59 (2013) 534–540 539
Acknowledgements 1999. Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N. Engl. J. Med. 19, 549–555. Helbing, C.C., van Aggelen, G., Veldhoen, N., 2011. Triclosan affects thyroid Dorte Hansen, Lillian Sztuk, Bo Herbst, Sarah Simonsen, hormone-dependent metamorphosis in anurans. Toxicol. Sci. 119, 417–418. Kenneth Worm, Heidi Letting, Birgitte Møller Plesning, Vibeke Henrichs, J., Bongers-Schokking, J.J., Schenk, J.J., Ghassabian, A., Schmidt, H.G., Kjær and Ulla El-Baroudy are thanked for their excellent technical Visser, T.J., Hooijkaas, H., de Muinck Keizer-Schrama, S.M., Hofman, A., Jaddoe, V.V., Visser, W., Steegers, E.A., Verhulst, F.C., de Rijke, Y.B., Tiemeier, H., 2010. assistance. Maternal thyroid function during early pregnancy and cognitive functioning in early childhood: the generation R study. J. Clin. Endocrinol. Metab. 95, 4227– 4234. Howdeshell, K., 2002. A model of the development of the brain as a construct of the References thyroid system. Environ. Health Perspect. 110, 337–348. Jacobs, M.N., Nolan, G.T., Hood, S.R., 2005. Lignans, bacteriocides and organochlorine compounds activate the human pregnane X receptor (PXR). Akaike, M., Kato, N., Ohno, H., Kobayashi, T., 1991. Hyperactivity and spatial maze Toxicol. Appl. Pharmacol. 209, 123–133. learning impairment of adult rats with temporary neonatal hypothyroidism. Jones, S.A., Moore, L.B., Shenk, J.L., Wisely, G.B., Hamilton, G.A., McKee, D.D., Neurotoxicol. Teratol. 13, 317–322. Tomkinson, N.C.O., LeCluyse, E.L., Lambert, M.H., Willson, T.M., Kliewer, S.A., Auso, E., Lavado-Autric, R., Cuevas, E., Escobar Del Rey, F., Morreale De Escobar, G., Moore, J.T., 2000. The pregnane X receptor: a promiscuous xenobiotic receptor Berbel, P., 2004. A moderate and transient deficiency of maternal thyroid that has diverged during evolution. Mol. Endocrinol. 14, 27–39. function at the beginning of fetal neocorticogenesis alters neuronal migration. Klinger, W., 2005. Developmental pharmacology and toxicology: Biotransformation Endocrinology 145, 4037–4047. of drugs and other xenobiotics during postnatal development. Eur. J. Drug Axelstad, M., Hansen, P.R., Boberg, J., Bonnichsen, M., Nellemann, C., Lund, S.P., Metab. Pharmacokinet 30, 3–17. Hougaard, K.S., Hass, U., 2008. Developmental neurotoxicity of propylthiouracil Kooistra, L., Crawford, S., van Baar, A.L., Brouwers, E.P., Pop, V.J., 2006. Neonatal (PTU) in rats: relationship between transient hypothyroxinemia during effects of maternal hypothyroxinemia during early pregnancy. Pediatrics 117, development and long-lasting behavioural and functional changes. Toxicol. 161–167. Appl. Pharmacol. 232, 1–13. Kumar, V., Chakraborty, A., Kural, M.R., Roy, P., 2009. Alteration of testicular Axelstad, M., Boberg, J., Hougaard, K.S., Christiansen, S., Jacobsen, P.R., Mandrup, steroidogenesis and histopathology of reproductive system in male rats treated K.R., Nellemann, C., Lund, S.P., Hass, U., 2011a. Effects of pre- and postnatal with triclosan. Reprod. Toxicol. 27, 177–185. exposure to the UV-filter octyl methoxycinnamate (OMC) on the reproductive, Lavado-Autric, R., Ausó, E., García-Velasco, J.V., Arufe Mdel, C., Escobar del Rey, F., auditory and neurological development of rat offspring. Toxicol. Appl. Berbel, P., Morreale de Escobar, G., 2003. Early maternal hypothyroxinemia Pharmacol. 250, 278–290. alters histogenesis and cerebral cortex cytoarchitecture of the progeny. J. Clin. Axelstad, M., Boberg, J., Nellemann, C., Kiersgaard, M., Jacobsen, P.R., Christiansen, S., Invest. 111, 1073–1082. Hougaard, K.S., Hass, U., 2011b. Exposure to the widely used fungicide Li, Y., Shan, Z., Teng, W., Yu, X., Li, Y., Fan, C., Teng, X., Guo, R., Wang, H., Li, J., Chen, mancozeb causes thyroid hormone disruption in rat dams but no Y., Wang, W., Chawinga, M., Zhang, L., Yang, L., Zhao, Y., Hua, T., 2010. developmental neurotoxicity in the offspring. Toxicol. Sci. 120, 439–446. Abnormalities of maternal thyroid function during pregnancy affect Berbel, P., Navarro, D., Ausó, E., Varea, E., Rodríguez, A.E., Ballesta, J.J., Salinas, M., neuropsychological development of their children at 25–30 months. Clin. Flores, E., Faura, C.C., de Escobar, G.M., 2010. Role of late maternal thyroid Endocrinol. (Oxf.) 72, 825–829. hormones in cerebral cortex development: an experimental model for human Louis, G.W., Hallinger, D.R., Stoker, T.E., 2013. The effect of triclosan on the prematurity. Cereb. Cortex 20, 1462–1475. uterotrophic response to extended doses of ethinyl estradiol in the weanling Brosvic, G.M., Taylor, J.N., Dihoff, R.E., 2002. Influences of early thyroid hormone rat. Reprod. Toxicol. 36, 71–77. manipulations: delays in pup motor and exploratory behavior are evident in McIntyre, B.S., Barlow, N.J., Foster, P.M., 2001. Androgen-mediated development in adult operant performance. Physiol. Behav. 75, 697–715. male rat offspring exposed to flutamide in utero: permanence and correlation of Butt, C.M., Wang, D., Stapleton, H.M., 2011. Halogenated phenolic contaminants early postnatal changes in anogenital distance and nipple retention with inhibit the in vitro activity of the thyroid-regulating deiodinases in human liver. malformations in androgen-dependent tissues. Toxicol. Sci. 62, 236–249. Toxicol. Sci. 124, 339–347. Menoutis, J., Parisi, A.A., 2002. Testing for dioxin and furan contamination in Chen, J.G., Ahn, K.C., Gee, N.A., Gee, S.J., Hammock, B.D., Lasley, B.L., 2007. triclosan. Cosmet. Toiletries Mag. 117, 75–78. Antiandrogenic properties of parabens and other phenolic containing small Noda, S., Muroi, T., Takakura, S., Sakamoto, S., Takatsuki, M., Yamasaki, K., Tateyama, molecules in personal care products. Toxicol. Appl. Pharmacol. 221, 278–284. S., Yamaguchi, R., 2005. Preliminary evaluation of an in utero-lactation assay Christen, V., Crettaz, P., Oberli-Schrammli, A., Fent, K., 2010. Some flame retardants using 6-n-propyl-2-thiouracil. Arch. Toxicol. 79, 414–421. and the antimicrobials triclosan and triclocarban enhance the androgenic Opazo, M.C., Gianini, A., Pancetti, F., Azkcona, G., Alarcón, L., Lizana, R., Noches, V., activity in vitro. Chemosphere 81, 1245–1252. Gonzalez, P.A., Marassi, M.P., Mora, S., Rosenthal, D., Eugenin, E., Naranjo, D., Christiansen, S., Boberg, J., Axelstad, M., Dalgaard, M., Vinggaard, A.M., Metzdorff, Bueno, S.M., Kalergis, A.M., Riedel, C.A., 2008. Maternal hypothyroxinemia S.B., Hass, U., 2010. Low-dose perinatal exposure to di(2-ethylhexyl) phthalate impairs spatial learning and synaptic nature and function in the offspring. induces anti-androgenic effects in male rats. Reprod. Toxicol. 30, 313–321. Endocrinology 149, 5097–5106. Christiansen, S., Kortenkamp, A., Axelstad, M., Boberg, J., Scholze, M., Jacobsen, P.R., Paul, K.B., Hedge, J.M., DeVito, M.J., Crofton, K.M., 2010a. Short-term exposure to Faust, M., Lichtensteiger, W., Schlumpf, M., Burdorf, A., Hass, U., 2012. Mixtures triclosan dreases thyroxine in vivo via upregulation of hepatic catabolism in of endocrine disrupting contaminants modelled on human high end exposures: young long-evans rats. Toxicol. Sci. 113 (2), 367–379. an exploratory study in rats. Int. J. Androl. 35, 303–316. Paul, K.B., Hedge, J.M., DeVito, M.J., Crofton, K.M., 2010b. Developmental triclosan Crofton, K.M., Paul, K.B., DeVito, M.J., Hedge, J.M., 2007. Short-term in vivo exposure exposure decreases maternal and neonatal thyroxine in rats. Environ. Toxicol. to the water contaminant triclosan: evidence for disruption of thyroxine. Chem. 29, 2840–2844. Environ. Toxicol. Pharmacol. 24, 194–197. Paul, K.B., Hedge, J.M., Bansal, R.R., Zoeller, T., Peter, R., DeVito, M.J., Crofton, K.M., Cuevas, E., Ausó, E., Telefont, M., Morreale de Escobar, G., Sotelo, C., Berbel, P., 2005. 2012. Developmental triclosan exposure decreases maternal, fetal, and early Transient maternal hypothyroxinemia at onset of corticogenesis alters neonatal thyroxine: a dynamic and kinetic evaluation of a putative mode-of- tangential migration of medial ganglionic eminence-derived neurons. Eur. J. action. Toxicology 300, 31–45. Neurosci. 22, 541–551. Pop, V.J., Kuijpens, J.L., van Baar, A.L., Verkerk, G., van Son, M.M., de Vijlder, J.J., Dybdahl, M., Nikolov, N.G., Wedebye, E.B., Jónsdóttir, S.Ó., Niemelä, J.R., 2012. QSAR Vulsma, T., Wiersinga, W.M., Drexhage, H.A., Vader, H.L., 1999. Low maternal model for human pregnane X receptor (PXR) binding: screening of free thyroxine concentrations during early pregnancy are associated with environmental chemicals and correlations with genotoxicity, endocrine impaired psychomotor development in infancy. Clin. Endocrinol. (Oxf.) 50, 149– disruption and teratogenicity. Toxicol. Appl. Pharmacol. 262, 301–309. 155. Flippin, J.L., Hedge, J.M., DeVito, M.J., LeBlanc, G.A., Crofton, K.M., 2009. Predictive Pop, V.J., Brouwers, E.P., Vader, H.L., Vulsma, T., van Baar, A.L., de Vijlder, J.J., 2003. modeling of a mixture of thyroid hormone disrupting chemicals that affect Maternal hypothyroxinaemia during early pregnancy and subsequent production and clearance of thyroxine. Int. J. Toxicol. 28, 368–381. child development: a 3-year follow-up study. Clin. Endocrinol. (Oxf.) 59, 282– Gee, R., Charles, A., Taylor, N., Darbre, P., 2008. Oestrogenic and androgenic activity 288. of triclosan in breast cancer cells. J. Appl. Toxicol. 28, 78–91. Provost, T.L., Juarez de Ku, L.M., Zender, C., Meserve, L.A., 1999. Dose- and age- Ghassabian, A., Bongers-Schokking, J.J., Henrichs, J., Jaddoe, V.W., Visser, T.J., Visser, dependent alterations in choline acetyltransferase (ChAT) activity, learning and W., de Muinck Keizer-Schrama, S.M., Hooijkaas, H., Steegers, E.A., Hofman, A., memory, and thyroid hormones in 15- and 30-day old rats exposed to 1.25 or Verhulst, F.C., van der Ende, J., de Rijke, Y.B., Tiemeier, H., 2011. Maternal 12.5 PPM polychlorinated biphenyl (PCB) beginning at conception. Prog. thyroid function during pregnancy and behavioral problems in the offspring: Neuropsychopharmacol. Biol. Psychiat. 23, 915–928. the generation R study. Pediatr. Res. 69, 454–459. Rodriguez, P.E., Sanchez, M.S., 2010. Maternal exposure to triclosan impairs thyroid Gilbert, M.E., Sui, L., 2006. Dose-dependent reductions in spatial learning and homeostasis and female pubertal development in wistar rat offspring. J. Toxicol. synaptic function in the dentate gyrus of adult rats following developmental Environ. Health A73, 1678–1688. thyroid hormone insufficiency. Brain Res. 1069, 10–22. SCCS 2009. SCCP (Scientific Committee on Consumer Products), Opinion on Gilbert, M.E., Sui, L., 2008. Developmental exposure to perchlorate alters synaptic triclosan, 21 January 2009. transmission in hippocampus of the adult rat. Environ. Health Perspec. 116, Sharlin, D.S., Tighe, D., Gilbert, M.E., Zoeller, R.T., 2008. The balance between 752–760. oligodendrocyte and astrocyte production in major white matter tracts is Haddow, J.E., Palomaki, G.E., Allan, W.C., Williams, J.R., Knight, G.J., Gagnon, J., linearly related to serum total thyroxine. Endocrinology 149, 2527–2536. O’Heir, C.E., Mitchell, M.L., Hermos, R.J., Waisbren, S.E., Faix, J.D., Klein, R.Z., 540 M. Axelstad et al. / Food and Chemical Toxicology 59 (2013) 534–540
Stoker, T.E., Gibson, E.K., Zorrilla, L.M., 2010. Triclosan exposure modulates modulates thyroid hormone-associated gene expression and disrupts estrogen-dependent responses in the female wistar rat. Toxicol. Sci. 117, 45–53. postembryonic anuran development. Aquat. Toxicol. 80, 217–227. Suchy, F.J., Sokol, R.J., Balistreri, W.F., 2007. Liver Disease in Children. Cambridge Vinggaard, A.M., Niemelä, J., Wedebye, E.B., Jensen, G.E., 2008. Screening of University Press, third ed.. 397 chemicals and development of a quantitative structure–activity US EPA. Environmental Protection Agency (2011). Integrated approaches to testing relationship model for androgen receptor antagonism. Chem. Res. Toxicol. 21, and assessment strategy: use of new computational and molecular tools. FIFRA 813–823. Scientific Advisory Panel Consultation, May 24–26, 2011. Arlington (VA): FIFRA Zorrilla, L.M., Gibson, E.K., Jeffay, S.C., Crofton, K.M., Setzer, W.R., Cooper, R.L., Stoker, Scientific Advisory Panel Meeting. Available from: