61, 76 – 82 (2001) Copyright © 2001 by the Society of Toxicology
Effects of Short-Term in Vivo Exposure to Polybrominated Diphenyl Ethers on Thyroid Hormones and Hepatic Enzyme Activities in Weanling Rats Tong Zhou,* David G. Ross,† Michael J. DeVito,† and Kevin M. Crofton‡ ,1 *Curriculum in Toxicology, University of North Carolina, Chapel Hill, North Carolina; †Experimental Toxicology Division and ‡Neurotoxicology Division, National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711 Received September 7, 2000; accepted January 2001
Key Words: polybrominated diphenyl ether; thyroid hormone; hepatic enzyme activity; rat; benchmark dose.
Polybrominated diphenyl ethers (PBDEs), used as flame retardants, are ubiquitous environmental contaminants. PBDEs act as endocrine disruptors via alterations in thyroid hormone homeostasis. We examined thyroid hormone concentrations and hepatic enzyme activity in weanling rats exposed to three commercial PBDE mixtures: DE-71, DE-79, and DE-83R. Female Long-Evans rats, 28 days old, were orally administered various doses of DE-71, DE-79, or DE-83R for 4 days. Serum and liver samples were collected 24 h after the last dose and analyzed for serum total thyroxine (T 4), triiodothyronine (T 3), thyroid-stimulating hormone (TSH), hepatic microsomal ethoxy- and pentoxy-resorufinO-deethylase (EROD and PROD), and uridinediphosphate-glucuronosyltransferase (UDPGT) activities. The PBDE-treated groups did not exhibit significant changes in body weight; however, increased liver weights, as well as 10- to 20-fold induction in EROD and 30- to 40-fold induction in PROD were found in the DE-71– and DE-79 –treated animals. DE-71 and DE-79 caused dose-dependent depletion of T 4, accompanied by up to 3- to 4-fold induction in UDPGT activities. Serum total T 4 was decreased a maximum of 80% for DE-71 and 70% for DE-79 in the highest dose, with benchmark doses (BMDs) of approximately 12.74 mg/kg/day for DE-71 and 9.25 mg/kg/day for DE-79. Dose-related effects in serum T 3 levels were less apparent, with maximal reductions of 25–30% at the highest dose for both DE-71 and DE-79. The two mixtures showed no effect on serum TSH levels. Benchmark dose analysis revealed that the two mixtures were comparable in altering thyroid hormone levels and hepatic enzyme activity. DE-83R was not effective in altering any of the measured parameters. The present study suggests that short-term exposure to some commercial PBDE mixtures interferes with the thyroid hormone system via upregulation of UDPGTs.
Polybrominated diphenyl ether (PBDE) mixtures are manufactured as flame retardants for commercial products such as electronic equipment and textiles (IPCS, 1994). Extrusion of PBDEs can form toxic polybrominated dibenzofurans and polybrominated dibenzodioxins (Luijk et al., 1992). Bioaccumulation, due to persistence and lipophilicity, has led to the detection of PBDEs in many species of wildlife (Andersson and Blomkvist, 1981; de Boer et al., 1998) as well as humans (Sjodin et al., 1999; Meneses et al., 1999). In human breast milk samples and some wildlife species, where sufficient data are available, concentrations of PBDEs are increasing (Meironyte´ et al., 1999; Sellstrom et al., 1993). Indeed, concentrations of PBDEs in human breast milk increased over 50-fold from 1972 to 1997 (Meironyte´ et al., 1999; Nore´n and Meironyte´, 2000). Decabromodiphenyl ether (deca-BDE) accounts for 75% of total PBDEs usage (Nordic Council of Ministers, 1998) and is the major environmental contaminant found in sediment. However, 2,2⬘,4,4⬘-tetra-BDE (IUPAC: BDE-47) and 2,2⬘,4,4⬘,5penta-BDE (BDE-99) are the predominant contaminants found in wildlife and human samples (Kierkegaard et al., 1999; Lindstrom et al., 1999; Meironyte´ et al., 1999; Sellstrom et al., 1993; Strandman et al. 1999; Sjodin et al., 1999). PBDEs are structurally similar to polychlorinated biphenyls (PCBs), dioxins (TCDDs), and thyroid hormones, and therefore may act as endocrine disruptors via interference with thyroid hormone (TH) homeostasis (Brouwer et al., 1998; Hooper and McDonald, 2000). Chronic exposure to deca-BDE results in an increased incidence of thyroid hyperplasia and tumors in mice, but not rats (NTP, 1986). Reduction in plasma thyroxine (T 4) has been reported in young rodents (both rats and mice) following a 14-day exposure to commercial PBDE mixtures (Bromkal 70 or DE-71) or BDE-47 (Fowles et al., 1994; Darnerud and Sinjari, 1996; Hallgren and Darnerud, 1998). In all cases, however, plasma thyroid stimulating hormone (TSH) concentrations were not affected. Information on the mechanism by which PBDEs decrease T 4
The information in this document has been funded wholly (or in part) by the U.S. Environmental Protection Agency. It has been subjected to review by the National Health and Environmental Effects Research Laboratory and approved for publication. Approval does not signify that the contents reflect the views of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use. 1 To whom correspondence should be addressed at Neurotoxicology Division, MD-74B, National Health and Environmental Effects Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711. Fax: (919) 541-4849. E-mail: [email protected]
POLYBROMINATED DIPHENYL ETHERS AND THYROID HORMONES
TABLE 1 Chemical Composition and Doses
Mixture composition Dose (mg/kg/day)
penta-, tetra, hexa-, tri-BDE 0.3, 1, 3, 10, 30, 100, 300
octa-, hexa-, penta-, tri-BDE 0.3, 1, 3, 10, 30, 60, 100
deca-BDE (⬎ 98%) 0.3, 1, 3, 10, 30, 60, 100
Note. See Sjodin (2000) for composition of different commercial mixtures of PBDEs.
is limited. In vitro data suggest that hydroxylated PBDE congeners will displace T 4 from transthyretin, a plasma transport protein (Meerts et al., 2000). Marsh et al. (1998) demonstrated in vitro binding of two hydroxylated PBDE congeners to human TR-␣1 and TR-␤. PBDEs have also been demonstrated to induce both phase I and II metabolic enzymes. Induction of uridinediphosphate-glucuronosyltransferase (UDPGT), ethoxyresorufin-O-deethylase (EROD), and pentoxyresorufin-O-deethylase (PROD) activities have been found in both mice and rats exposed to various commercial mixtures and/or BDE-47 (Carlson, 1980a,b; Fowles et al., 1994; Hallgren and Darnerud, 1998; von Meyerinck et al., 1990). Carlson (1980a) concluded that the induction potency of the commercial mixtures was negatively correlated to the degree of bromination. Furthermore, some PBDE mixtures were suggested to be either solely phenobarbital-type inducers, such as DE-71 and DE-79 (Carlson, 1980a), or mixed-type inducers (i.e., phenobarbital and dioxin type) of xenobiotic metabolism, such as Bromkal 70 (von Meyerinck et al., 1990). BDE-47 has been shown to be a mixed-type inducer (Hallgren and Darnerud, 1998). Recently, some PBDE congeners (BDE-47, -77, and -138) were shown to exhibit Ah receptor–antagonistic activity in vitro (Meerts et al., 1998). Information on the induction of UDPGT is limited to one study using single doses of the three commercial mixtures; Carlson (1980a) found increased UDPGT activity with DE-71 and DE-79, but not DE-83R. Clearly, the data available for PBDE effects on thyroid homeostasis are limited. The present studies were conducted to contrast and compare the dose-response relationships of short-term exposure to three commercial PBDE mixtures (DE-71, DE-79, and DE-83R) on T 4, triiodothyronine (T 3), and TSH concentrations, as well as hepatic enzymatic activities (EROD, PROD, and UDPGT). These mixtures are representative of the three types of commercial mixtures used worldwide (Sjodin, 2000). These data are needed to increase our understanding of mechanism(s) by which PBDEs interfere with thyroid hormone homeostasis. MATERIALS AND METHODS Animals. A total of 188 Long-Evans female rats at 23 days of age were obtained from Charles River Laboratories Inc. (Raleigh, NC), and allowed 5 days of acclimation in an American Association for Accreditation of Laboratory Animal Care (AAALC)-approved animal facility prior to being treated. Two animals were housed per plastic cage (45 cm ⫻ 24 cm ⫻ 20 cm), with heat-treated pine shavings bedding. They were maintained at 21 ⫾ 2°C with
50 ⫾ 10% humidity on a photoperiod of 12L:12D (0600 –1800 h). Food (Purina Rodent Chow, Barnes Supply Co., Durham, NC ) and tap water were provided ad libitum. All animal procedures were approved by the U.S. EPA Institutional Animal Care and Use Committee. Chemicals and treatment. DE-71 (lot 7550OK20A), DE-79 (lot 8525DG01A), and DE-83R (lot 7480DL10C) were generously supplied by the Great Lakes Chemical Corporation (West Lafayette, IN) (Table 1). The compositions of the different lots were reported to contain 58.1% penta-BDE and 24.6% tetra-BDE for DE-71; 30.7% octa-BDE and 45.1% hepta-BDE for DE-79 (Carlson, 1980a); and more than 98% deca-BDE for DE-83R (Table 1). Analysis of the DE-71 mixture compared to another commercial penta-BDE (Bromkal 70-5DE) suggests that the composition may be closer to 35% tetra and 45% penta (see Sjodin, 2000). The dosing solution (or suspension for DE-83R at concentrations of 3 mg/kg/day or above) was prepared by mixing the compounds with corn oil, sonicating for 30 min at 40°C, and diluting in series with corn oil to the desired concentrations (Table 1). These doses were selected on the basis of results of pilot studies that measured body weights, liver weights, and thyroid hormone concentrations in small numbers of subjects (data not shown). Dosing solutions were administered in 1.0 ml/kg corn oil. All chemicals used in enzyme assays were purchased from Sigma Chemical Co. (St. Louis, MO) and were of the highest grade commercially available. Rats at 28 days of age were orally administrated DE-71, DE-79, or DE-83R (n ⫽ eight/dose, except for 0.3 mg/kg/day for DE-71 where n ⫽ 4) for 4 consecutive days. Separate sets of animals, including controls, were used for each commercial mixture. Young female rats were chosen to be consistent with ongoing research on the effects of short-term exposure to PCBs, TCDD, dibenzofurans, and PBDEs. Rats were semirandomly assigned to treatment groups by counterbalancing for body weights. Body weights were recorded and dosing volumes adjusted on a daily basis. On the fifth day, approximately 24 h after the last dose, the animals were killed between 0900 and 1030 h, and trunk blood and liver were collected. Serum was obtained after clotting whole blood on ice for approximately 1.5 h, followed by centrifugation at 2500 rpm at 4°C for 20 min. Serum and liver samples were stored at – 80°C until analysis. Thyroid hormone assay. Serum total T 4 and T 3 were measured in duplicate by standard radioimmunoassay (RIA) kits (Diagnostic Products Corporation, Los Angeles, CA). Serum TSH concentrations were analyzed in duplicates with a double antibody RIA method (Zoeller and Rudeen, 1992). Rat TSH RIA kits (including rat TSH antiserum NIDDK-antirat TSH-RIA-6; rat TSH reference preparation NIDDK-rTSH-RP3) were obtained through NIDDK’s National Hormone & Pituitary Program. I-125–labeled TSH was purchased from Covance Laboratories (Vienna, VA). Intraassay and interassay coefficients of variance for all assays were below 10%. Hepatic enzyme activities assay. Liver microsomal fractions were prepared as described previously (DeVito et al., 1993). Microsomal protein concentrations were determined using a protein assay kit (Bio-Rad, Richmond, CA) with bovine serum albumin as the standard. Hepatic microsomal EROD (a marker for CYP1A1 activity) activity was assayed using the method of DeVito et al. (1993). A similar method (DeVito et al., 1993) was used to measure hepatic microsomal PROD (a marker of CYP2B activity) activity, using pentoxyresorufin as a substrate. All substrate concentrations were 1.5 nM. Both EROD and PROD values were calculated as picomoles (pmol) resorufin per milligram protein per minute, and graphically represented as percentage of control activity.
ZHOU ET AL.
TABLE 2 Absolute Values for Control Groups (mean ⴞ SEM) for Serum Total T 4, T 3, Liver Microsomal EROD, PROD, and UDPGT Activities
T4 T3 EROD PROD UDPGT
50.12 ⫾ 4.43 154.70 ⫾ 12.90 251.74 ⫾ 27.42 15.02 ⫾ 1.61 0.87 ⫾ 0.12
48.41 ⫾ 3.08 110.80 ⫾ 2.39 295.00 ⫾ 36.59 15.62 ⫾ 2.01 0.76 ⫾ 0.06
58.42 ⫾ 2.36 137.46 ⫾ 13.10 195.18 ⫾ 22.29 8.64 ⫾ 0.59 0.71 ⫾ 0.08
Note. Values for T 4 are given in ng/ml; values for T 3 are in ng/dl. EROD, pmol resorufin/mg protein/min; PROD, pmol resorufin/mg protein/min; UDPGT, pmol T 4-G/mg protein/min.
Hepatic microsomal T 4-UDPGT activity was assayed based on the method of Visser et al. (1993). 125I–labeled T 4 was purchased from NEN Life Science Products Inc. (Boston, MA). Aliquots of liver microsomes were diluted to 2 mg protein/ml with 100 mM Tris-HCl (pH ⫽ 7.8) containing 5 mM MgCl 2. Then 100 l diluted microsomes was incubated without detergent at 37°C, with purified 1 M 125I-labeled T 4 (around 50,000 dpm) as substrate, 100 M 6-n-propyl-2-thiouracil (for preventing de-iodination), and 5 mM UDPGA (as cofactor, or no UDPGA for blanks) over a 30-min period. The reaction was stopped by the addition of 200 l ice-cold methanol, followed by centrifugation at 4°C at 3000 rpm for 5 min. Two hundred microliters of supernatant was transferred to a microtube and mixed with 750 l 0.1N HCl. The T 4-glucuronyl product (T 4-G) was then separated by chromatography on Supelco Filtration Column filled with 2 ml lipophilic Sephadex LH-20 in gel suspension. The column was eluated with 2 ml of 0.1N HCl and 8 ml deionized water, and the collected fractions were counted for radioactivity for 1 min on the gamma counter. The calculated UDPGT activity was expressed as picomoles T 4-G per milligram protein per minute. Data analysis. All data were analyzed as the percentage of the respective control means. Absolute control values are presented in Table 2. One-way ANOVAs followed by Duncan’s Multiple Range Tests were used to compare differences among treatment groups, with acceptable significance levels set at p ⬍ 0.05. All data are presented as mean ⫾ SEM (n ⫽ 8, except for 0.3 mg/kg/day for DE-71 where n ⫽ 4). No-observed-effect-levels (NOELs) were defined as the lowest dose without significant effect. Benchmark dose (BMD) estimates were determined for alterations in thyroid hormones and hepatic enzyme activities using the U.S. EPA Benchmark Dose Software (BMDS Version 1.2). The Hill model was chosen to fit these continuous data according to the following equations:
y共 x兲 ⫽ e 0 ⫹
共e max*x n 兲 共b n ⫹ x n 兲
for responses that increase with the dose
y共 x兲 ⫽ e 0 ⫺
共e max*x n 兲 共b n ⫹ x n 兲
for responses that decrease with the dose
No visible signs of toxicity were noted after the short-term PBDE treatments. No treatment-related effects on body weight gain were found for the three mixtures: [DE-71: F(7,52) ⫽ 1.48, p ⬍ 0.1963; DE-79: F(7,56) ⫽ 1.92, p ⬍ 0.0832; DE83R: F(7,56) ⫽ 1.39, p ⬍ 0.2283]. However, there were dose-related increases in liver-to-body weight ratios following DE-71 [F(7,52) ⫽ 26.20, p ⬍ 0.0001] and DE-79 [F(7,56) ⫽ 19.24, p ⬍ 0.0001] treatment (Fig. 1). Liver-to-body weight ratios increased 47% in the 300 mg/kg/day DE-71 group and 30% in the 100 mg/kg/day DE-79 group. In contrast, the fully brominated DE-83R did not affect liver-to-body weight ratios at any dose [F(7,56) ⫽ 0.91, p ⬍ 0.5078]. Serum total T 4 concentrations were dramatically reduced in a dose-dependent manner for both DE-71 and DE-79 (Fig. 2A). Serum T 4 was decreased 80% in the 300 mg/kg/day dose of DE-71 and 70% for the 100 mg/kg/day dose of DE-79. Effects of PBDEs on serum total T 3 concentrations were much smaller than effects on T 4 (Fig. 2B). Reduction of serum T 3 was only 30% for the 300 mg/kg/day dose of DE-71 and 25% for DE-79 at the highest dose of 100 mg/kg/day. These conclusions were supported by significant main effects of treatment for T 4 [DE71: F(7,52) ⫽ 25.78, p ⬍ 0.0001; DE-79: F(7,56) ⫽ 26.30, p ⬍ 0.0001] and T 3 [DE-71: F(7,52) ⫽ 4.40, p ⬍ 0.0007; DE-79: F(7,56) ⫽ 3.99, p ⬍ 0.0013]. Neither serum T 4 or T 3 concentrations were affected by DE-83R (Fig. 2; all p values ⬎ 0.5744). Serum total TSH concentrations were not changed by any PBDE treatments (data not shown; all p values ⬎ 0.5881). Marked induction (approximately 20-fold in the highest dose) in hepatic microsomal EROD activity was seen in the DE-71–treated groups (Fig. 3A). Induction of EROD activity
where y is the response; x is the dose; e 0 is the estimated background response level; e max is the maximal increase or decrease from background; b is the ED 50; and n is the shape parameter. The benchmark effect levels were set at 20% decreases for the thyroid hormone data and 50% increases for the liver enzyme data. The BMDs were calculated from the Hill fits to the data. The BMDL (lower-bound confidence limit) was calculated as the 95% lower confidence interval.
FIG. 1. Dose-effect curves for the effects of a 4-day exposure to three commercial PBDE mixtures (DE-71, DE-79, DE-83R) on liver:body weight ratio (expressed as percentage of mean control values) following 4 days of treatment. Data are presented as group means ⫾ SE; symbols with no apparent error bars have error estimates too small to see or hidden by the symbol. *Significantly different from the respective control (p ⬍ 0.05).
POLYBROMINATED DIPHENYL ETHERS AND THYROID HORMONES
0.0001; DE-79: F(7,56) ⫽ 12.94, p ⬍ 0.0001]. In contrast, no significant induction in UDPGT [F(7,56), p ⬍ 0.2991] was found in the DE-83R treated groups. Although in the DE-83R treated rats there was a statistically significant main effect of treatment on PROD [F(7,56) ⫽ 2.64, p ⬍ 0.0270], there was no dose dependence to the effect and only the 1 and 30
FIG. 2. Dose-effect curves for the effects of a 4-day exposure to three commercial PBDE mixtures (DE-71, DE-79, DE-83R) on serum total T 4 (A) and T 3 (B) concentrations following 4 days of treatment. Data were expressed as percentage of mean control values. Data are presented as group means ⫾ SE; symbols with no apparent error bars have error estimates too small to see or hidden by the symbol. *Significantly different from the respective control (p ⬍ 0.05).
by DE-79 was less pronounced (about 10-fold) in comparison to that by DE-71. These findings were supported by significant main effects of treatments [DE-71: F(7,52) ⫽ 243.75, p ⬍ 0.0001; DE-79: F(7,56) ⫽ 94.82, p ⬍ 0.0001]. Hepatic microsomal PROD activity increased dramatically after DE-71 [26fold, F(7,52) ⫽ 44.57, p ⬍ 0.0001] or DE-79 [36-fold, F(7,56) ⫽ 71.58, p ⬍ 0.0001] treatment (Fig. 3B). Baseline hepatic microsomal PROD activity was much lower than EROD (average of 13 ⫾ 1 pmol resorufin/mg/min) (Table 2). Hepatic microsomal UDPGT activity showed dose-related increases for DE-71 and DE-79 (Fig. 3C). DE-71 induced UDPGT activity by 5-fold (relative to controls) in the highest dose of 300 mg/kg/day (0.87 ⫾ 0.12 vs. 4.11 ⫾ 0.59 pmol T 4-G/mg/min). DE-79 caused a 2.5-fold increase in UDPGT activity from 0.76 ⫾ 0.06 for controls to 1.96 ⫾ 0.24 pmol T 4-G/min/mg protein in the highest dose of 100 mg/kg/day. One-way ANOVA revealed significant effects of treatment on UDPGT for both mixtures [DE-71: F(7,52) ⫽ 19.38, p ⬍
FIG. 3. Dose-effect curves for the effects of a 4-day exposure to three commercial PBDE mixtures (DE-71, DE-79, DE-83R) on liver microsomal EROD (A), PROD (B), and UDPGT (C) activities expressed as percentage of mean control values, following 4 days of treatment. Data are presented as group means ⫾ SE; symbols with no apparent error bars have error estimates too small to see or hidden by the symbol; *Significantly different from the respective control (p ⬍ 0.05).
ZHOU ET AL.
TABLE 3 BMD and BMDL Values for DE-71 and DE-79 DE-71
T4 T3 EROD PROD UDPGT
10 30 3 3 10
12.74 32.94 2.88 0.81 9.51
6.95 8.56 1.82 a 0.54 a 5.83 a
3 30 10 3 10
9.25 53.38 3.66 0.53 21.17
5.29 11.98 2.45 0.40 a 11.03
Note. All values are mg/kg/day. Values were calculated from data expressed as percentage of control. a The points at the two highest doses were excluded in curve fitting.
mg/kg/day groups were different from the controls. The same lack of a dose dependence was also found for EROD [F(7,56) ⫽ 4.14, p ⬍ 0.001], where maximal induction was only 1.8-fold in the 1 mg/kg/day group. NOELs and model estimates for BMDs and BMDLs are shown in Table 3. NOELs and BMDs are fairly similar for most end points. The comparison of BMDs and NOELs for T 4 and UDPGT values were the exception. BMD estimates of T 4 and UDPGT were better estimates of potency compared to NOELs, based on visual inspection of the data. DISCUSSION
DE-71 and DE-79 caused hypothyroxinemia in weanling female rats following a 4-day exposure. This finding is consistent with other short-term and long-term studies on technical or pure PBDEs in rodents (Carlson, 1980a,b; Darnerud and Sinjari, 1996; Fowles et al., 1994; Hallgren and Darnerud, 1998). Reduction in circulating total T 3, but to a lesser extent than T 4, was also noted as a result of DE-71 or DE-79 exposure. The reduction in T 4 and T 3 was coincident with the induction of hepatic phase I enzymes CYP1A1 (shown by increased EROD activity), CYP2B (shown by increased PROD activity), and phase II UDPGTs. In contrast, DE-83R, which consists of more than 98% of the highly brominated deca-BDE, had no significant effects on levels of thyroid hormones or UDPGT activity. There was no biologically significant effect of DE-83R on PROD or EROD activity. In our study, NOEL values for liver and T 4 effects occurred within the range (2–18 mg/kg/day) found by other subacute and subchronic studies of commercial PBDEs (DE-71, Bromkal 70, commercial octaBDE, and penta-BDE) on rats (Nordic Council of Ministers, 1998). DE-71 and DE-79 had similar effects on thyroid hormones. Both compounds decreased circulating total serum T 4 concentrations in a dose-dependent manner, reaching 75– 80% decreases at the highest doses of DE-71 (300 mg/kg/day) and DE-79 (100 mg/kg/day). Total serum T 3 was affected to a much smaller degree, with only 25–30% decreases. There were
no effects of either mixture on TSH concentrations. This pattern of effects is consistent with other reports of exposure to PBDE mixtures and individual congeners. Darnerud and Sinjari (1996) demonstrated decreased total plasma T 4 in both rats and mice exposed for 14 days to 18 or 36 mg/kg/day of Bromkal 70. These same authors also exposed mice to 18 mg/kg/day BDE-47 and found a 31% decrease in total plasma T 4. Hallgren and Darnerud (1998) found decreases in both total and free plasma T 4 with no increase in TSH following a 14-day exposure of female rats to 18 mg/kg/day BDE-47. Mice exposed for 14 days to up to 72 mg/kg/day DE-71 had decreased free and total serum T 4 (Fowles et al., 1994). These data clearly demonstrate that PBDEs impact circulating concentrations of T 4. Two conclusions can be drawn from the hepatic enzyme assays. The first conclusion is that both DE-71 and DE-79 are mixed-type inducers of hepatic enzymatic activity. Second, these two mixtures may decrease circulating concentrations of T 4, at least partially, by increasing hepatic glucuronidation. Evidence in support of the conclusion of mixed induction includes increases in both EROD (induced by activation of the Ah receptor) and PROD activities [induced via interactions of multiple genes associated with the phenobarbital responsive unit (Ganem et al., 1999)] (see Fig. 3). These findings are consistent with a number of previous investigations. Early work by Carlson (1980a,b) showed that both 14- and 90-day exposure to penta- and octa-BDE mixtures increased hepatic benzo-[a]-pyrene and p-nitroanisole metabolism. Consistent with our findings, Carlson (1980a) also failed to find any hepatic enzyme induction with 14-day exposures to a decaBDE mixture. von Meyerinck et al. (1990) found increased EROD and benzphetamine N-demethylation activity in hepatic tissue from mice treated for 14 days with Bromkal 70. Increased EROD and PROD activity were found in mice exposed to DE-71 for 14 days, but only increased PROD was found following acute exposure (Fowles et al., 1994). The small amount of induction of EROD in these mice (about 3.3-fold), compared to the 20-fold induction in rats found in the current study, could be due to species differences and/or the differences in the commercial mixtures used. More recently, Hallgren and Darnerud (1998) reported increased EROD (3-fold) and PROD (14-fold) activities in rats after 14 days of exposure to BDE-47. One of the mechanisms responsible for thyroid hormone depletion is induction of the phase II UDPGT enzymes in liver and subsequent biliary elimination of thyroxine as T 4-glucuronide (for review see: Mackenzie et al., 1997). Hallgren and Darnerud (1998) demonstrated a small increase in T 4-gluronidation following exposure to BDE-47. DE-71 and DE-79 exposure resulted in increased napthol glucuronidation in rats exposed for 14 days (Carlson, 1980a). T 4 glucuronidation is associated with several isozymes of UDPGTs (Visser et al., 1993), one of which (UDPGT 1A6 isozyme) is controlled by the arylhydrocarbon (Ah) receptor signal transduction pathway (Brouwer et al., 1998). Hepatic UDPGT induction has been
POLYBROMINATED DIPHENYL ETHERS AND THYROID HORMONES
suggested as the dominant mechanism for thyroid hormone depletion by dioxinlike polyhalogenated aromatic hydrocarbons (PHAHs) (Kohn et al., 1996; Schuur et al., 1997). However, the mechanisms by which PBDEs (a class of PHAHs) interfere with thyroid hormone homeostasis are not well studied and are often inconclusive. In our study, a dose-related reduction in plasma T 4 concentrations was consistent with a dose-related induction in hepatic UDPGT activity, suggesting that T 4 glucuronidation was one factor contributing to the reduction in serum T 4. DE-79 showed potency comparable to DE-71 in plasma T 4 depletion, but induced UDPGT only approximately half as much as DE-71. This suggests mechanisms other than T 4 glucuronidation may also contribute to the plasma T 4 depletion. The thyroid gland itself may be affected by bromine, if liberated from the PBDE, which would inhibit iodine uptake into the thyroid (Velicky et al., 1998). However, this is not necessarily a tenable argument, as iodine uptake inhibitors such as perchlorate cause decreases in both T 3 and T 4 (Wolf, 1998), and in the present study, DE-71 and DE-79 affected T 4 predominantly. Interference with the pituitary– thyroid axis is not a likely mechanism of PBDEs, as the present results did not show significant changes in TSH concentrations in spite of the considerable reduction in serum T 4 and modest decreases in serum T 3. TSH increases have been reported following 4-day exposures to other xenobiotics (O’Connor et al., 1999), so the lack of TSH increase may not be due to the short duration of exposure. Alternatively, hydroxylated PBDEs have been shown in vitro to displace T 4 from binding to the plasma transport protein, transthyretin (Meerts et al., 2000). This could result, hypothetically, in increased hepatic catabolism of thyroid hormones (Meerts et al., 2000). However, whether the hydroxylated PBDEs tested by Meerts et al. (2000) are metabolites of the PBDE mixtures examined in the present study is unknown. The lack of effect of DE-83R, which consists of more than 98% of the highly brominated deca-BDE, is consistent with negative findings from other short-term dosing studies in rats. A 14-day exposure to deca-BDE in rats failed to find any evidence of induction of CYP proteins or UDPGT activity (Carlson, 1980a). This may be due to low solubility of this compound in aqueous solutions as well as the corn oil vehicle, and very limited absorption of these very highly brominated congeners. In rats orally administered specific congeners, more than 86% of the dose for 2,2⬘,4,4⬘-tetra-BDE (Orn and LassonWehler, 1998) and 57% of the dose for 2,2⬘,4,4⬘,5-penta-BDE (Hakk et al., 1999) were retained after 72 h. A 12-day feeding study with deca-BDE found absorption of less than 1% of the administered dose (el Dareer et al., 1987). Chronic exposures to deca-BDE, however, have revealed thyroid and liver tumors (NTP, 1986). These results suggest that prolonged exposure to the deca-BDE can result in the accumulation of this chemical or its metabolites in rodents. Induction of hepatic enzymes and disruption of thyroid hormones may, therefore, be likely following longer exposures. A comparison of NOELs and BMDs suggest that the latter
may provide better potency estimates for comparisons of doseeffect functions. For example, the NOELs for T 4 reduction are 10 and 3 mg/kg/day for DE-71 and DE-79, respectively. The BMDs for the same groups were 12.7 and 9.2 mg/kg/day. This difference between NOELS and BMDs is primarily due to increased variability in the DE-71– dosed animals compared to the DE-79 animals, causing the ANOVA-based approach to require greater differences between treated and control means to reach statistical significance. In the BMD approach, this variation does not lead to an increased estimate and therefore provides a better approximation of potency. Another disparity between NOEL and BMD estimates can been seen in the UDPGT data. The NOEL estimates indicate that DE-79 is more potent than DE-71; however, BMD estimates indicate the opposite. This difference is due primarily to dose spacing. The BMD model can extrapolate between doses and thus provides a more accurate estimate of the potency (Allen et al., 1994; Foster and Auton, 1995). In summary, our data show that 4-day in vivo DE-71 or DE-79 exposures induced hypothyroxinemia and UDPGT activity in weanling female rats. DE-71 and DE-79 demonstrated comparable potency and efficacy in terms of decreasing levels of serum thyroid hormones (T 4 and T 3), but different capability in inducing hepatic enzyme activities. Both DE-71 and DE-79 were mixed inducers of hepatic biotransformation enzymes, but DE-71 showed more dioxin-like effects while DE-79 showed more phenobarbital-like effects. The T 4-depleting effects of DE-71 and DE-79 are likely to involve multiple mechanisms of interference in addition to thyroid hormone metabolism. DE-83R (mostly deca-BDE) was ineffective in altering levels of serum thyroid hormones or hepatic EROD, PROD, or UDPGT activity, most likely due to the limited absorption of this highly brominated mixture. ACKNOWLEDGMENTS This research was funded by the U.S. EPA/UNC Toxicology Research Program Training Agreement CT 902908, with the Curriculum in Toxicology, University of North Carolina at Chapel Hill. Technical assistance from Ms. Michele M. Taylor and Ms. Elena Craft is also greatly appreciated. We are also grateful for help with the TSH assay from Drs. Tom R. Zoeller and Amy L. S. Dowling.
REFERENCES Allen B. C., Kavlock R. J., Kimmel C. A., and Faustman E. M. (1994). Dose-response assessment for developmental toxicity. II. Comparison of generic benchmark dose estimates with no observed adverse effect levels. Fundam. Appl. Toxicol. 23, 487– 495. Andersson, O., and Blomkvist, G. (1981). Polybrominated aromatic pollutants found in fish in Sweden. Chemosphere 10, 1051–1060. Brouwer, A., Morse, D. C., Lans, M. C., Schuur, A. G., Murk, A. J., KlassonWehler, E., Bergman, A. and Visser, T. J. (1998). Interactions of persistent environmental organohalogens with the thyroid hormone system: Mechanisms and possible consequences for animal and human health. Toxicol. Ind. Health 14, 59 – 84. Carlson, G. P. (1980a). Induction of xenobiotic metabolism in rats by shortterm administration of brominated diphenyl ethers. Toxicol. Lett. 5, 19 –25.
ZHOU ET AL.
Carlson, G. P. (1980b). Induction of xenobiotic metabolism in rats by brominated diphenyl ethers administered for 90 days. Toxicol. Lett. 6, 207–212. Darnerud, P. O., and Sinjari, T. (1996). Effects of polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs) on thyroxine and TSH blood levels in rats and mice. Organohalogen Comp. 29, 316 –319. de Boer, J. Wester, P. G., Klamer, H. J. C., Lewis, W. E., and Boon, J. P. (1998). Do flame retardants threaten ocean life? Nature 394, 28 –29. DeVito, M. J., Maier, W. E., Diliberto, J. J., and Birnbaum, L. S. (1993). Comparative ability of various PCBs, PCDFs, and TCDD to induce cytochrome P450 1A1 and 1A2 activity following 4 weeks of treatment. Fundam. Appl. Toxicol. 20, 125–130. el Dareer, S. H., Kalin, J. R., Tillery K. F., and Hill, D. L. (1987). Disposition of decabromobiphenyl ether in rats dosed intravenously or by feeding. J. Toxicol. Environ. Health 22, 405– 415. Foster P. M., and Auton T. R. (1995). Application of benchmark dose risk assessment methodology to developmental toxicity: an industrial view. Toxicol. Lett. 82–83, 555–559 Fowles, J. R., Fairbrother, A., Baecher-Steppan, L., Kerkvliet, N. I. (1994). Immunologic and endocrine effects of the flame-retardant pentabromodiphenyl ether (DE-71) in C57BL/6J mice. Toxicology 86, 49 – 61. Ganem, L. G., Trottier, E., Anderson, A. and Jefcoate, C. R. (1999). Phenobarbital induction of CYP2B1/2 in primary hepatocytes: endocrine regulation and evidence for a single pathway for multiple inducers. Toxicol. Appl. Pharmacol. 155, 32– 42. Hallgren, S., and Darnerud, P. O. (1998). Effects of polybrominated diphenyl ethers (PBDEs), polychlorinated biphenyls (PCBs) and chlorinated paraffins (CPs) on thyroid hormone levels and enzyme activities in rats. Organohalogen Comp. 35, 391–394. Hakk, H., Larsen, G., Klasson-Wehler, E., Orn, U., and Bergman, A. (1999). Tissue disposition, excretion, and metabolism of 2,2⬘,4,4⬘,5-pentabromodiphenyl ether (BDE-99) in male Sprague-Dawley rats. Organohalogen Comp. 40, 337–340. Hooper, K., and McDonald, T. A. (2000). The PBDEs: An emerging environmental challenge and another reason for breast-milk monitoring program. Environ. Health Perspect. 108, 387–392. IPCS (1994). Environmental Health Criteria 162: Brominated diphenyl ethers. International Programme on Chemical Safety, World Health Organization, Geneva. Kierkegaard, A., Sellstrom, U., Bignert, A., Olsson, M., Asplund, L., Jansson, B., and de Wit C. (1999). Temporal trends of a polybrominated diphenyl ether (PBDE), a methoxylated PBDE, and hexabromocyclododecane (HBCD) in Swedish biota. Organohalogen Comp. 40, 367–370. Kohn, M. C., Sewall, C. H., Lucier, G. W., and Portier, C. J. (1996). A mechanistic model of effects of doxin on thyroid hormones in the rat. Toxicol. Appl. Pharmacol. 165, 29 – 48. Lindstrom, G., Wingfors, H., Dam, M., and van Bavel, B. (1999). Identification of 19 polybrominated diphenyl ethers (PBDEs) in long-finned pilot whale (Globicephala melas) from the Atlantic. Arch. Environ. Contam. Toxicol. 36, 355–363. Luijk, R., Govers, H. A. J., and Nelissen, L. (1992). Formation of polybrominated dibenzofurans during extrusion of high-impact polystyrene/decabromodiphenyl ether/antiomony(III) oxide. Environ. Sci. Technol. 26, 2191– 2198. Marsh, G., Bergman, A., Bladh, L. G., Gillner, M., and Jakobsson, E. (1998). Synthesis of p-hydroxybromodiphenyl ethers and binding to the thyroid receptor. Organohalogen Comp. 37, 305–308. Mackenzie, P. I., Owens, I. S., Burchell, B., Bock, K. W., Bairoch, A., Belanger, A., Fournel-Gigleux, S., Green, M., Hum, D. W., Iyanagi, T., Lancet, D., Louisot, P., Magdalou, J., Roy Chowdhury, J., Ritter, J. K., Schachter, H., Tephly, T. R., Tipton, K. F., and Nebert, D. W. (1997). The UDP glycosyltransferase gene superfamily: Recommended nomenclature update based on evolutionary divergence. Pharmacogenetics 7, 255–269.
Meerts, I. A. T. M, Luijks, E. A. C., Marsh, G., Jokobsson, E., Bergman, A., and Brouwer, A. (1998). Polybrominated diphenyl ethers (PBDEs) as Ahreceptor agonists and antagonists. Organohalogen Comp. 37, 147–150. Meerts, I. A., van Zanden, J. J., Luijks, E. A., van Leeuwen-Bol, I., Marsh, G., Jakobsson, E., Bergman, A., and Brouwer, A. (2000). Potent competitive interactions of some brominated flame retardants and related compounds with human transthyretin in vitro. Toxicol. Sci. 56, 95–104. Meironyte´, D., Nore´n K., and Bergman, A. (1999). Analysis of polybrominated diphenyl ethers in Swedish human milk. A time-related trend study, 1972– 1997. J. Toxicol. Environ. Health 58, 329 –341. Meneses, M., Wingfors, H., Schuhmacher, M., Domingo, J. L., Lindstrom, G., and van Bavel, B. (1999). Polybrominated diphenyl ethers detected in human adipose tissue from Spain. Chemosphere 39, 2271–2278. Nore´n, K., and Meironyte´, D. (2000). Certain organochlorine and organobromine contaminants in Swedish human milk in perspective of past 20 –30 years. Chemosphere 40, 1111–1123. Nordic Council of Ministers (1998). Polybrominated diphenyl ethers: food contamination and potential risks. TemaNord 1998:503, Copenhagen. NTP (1986). Toxicology and carcinogenesis studies of decabromodiphenyl oxide (CAS No. 1163-19-5) in F344/N rats and B6C3F1 mice (feed studies). National Toxicology Program, Technical Report Series, No. 309. O’Connor, J.C., Frame, S. R., Davis, L. G., and Cook, J. C. (1999). Detection of thyroid toxicants in a tier I screening battery and alterations in thyroid endpoints over 28 days of exposure. Toxicol. Sci. 51, 54 –70. Orn, U, and Klasson-Wehler, E. (1998). Metabolism of 2,2⬘,4,4⬘-tetra-bromodiphenyl ether in rat and mouse. Xenobiotica 28, 199 –211. Schuur, A. G., Boekhorst, F. M., Brouwer, A. and Visser, T. J. (1997). Extrathyroidal effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin on thyroid hormone turnover in male Sprague-Dawley rats. Endocrinology 138, 3727– 3734. Sellstrom, U., Jansson, B., Kierkegaard, A., and de Wit, C. (1993). Polybrominated diphenyl ethers (PBDE) in biological samples from the Swedish environment. Chemosphere 26, 1703–1718. Sjodin, A., Hagmar, L., Klasson-Wehler, E., Kronholm-Diab, K., Jakobsson, E., and Bergman, A. (1999). Flame retardant exposure: polybrominated diphenyl ethers in blood from Swedish workers. Environ. Health Perspect. 107, 643– 648. Sjodin, A. (2000). Occupational and dietary exposure to organohalogen substances, with special emphasis on polybrominated diphenyl ethers. Doctoral dissertation, Stockholm University. Strandman, T., Koistinen, J., Kiviranta, H., Vuorinen, P. J., Tuomisto, J., Tuomisto, J., and Vartiainen, T. (1999). Levels of some polybrominated diphenyl ethers (PBDEs) in fish and human adipose tissue in Finland. Organohalogen Comp. 40, 355–358. Velicky J., Titlback, M., Lojda Z., Duskova J., Vobecky M., Strbak V., and Raska, I. (1998). Long-term action of potassium bromide on the rat thyroid gland. Acta Histochem. 100, 11–23. Visser, T. J., Kaptein, E., van Toor, H., van Raaij, J. A. G. M., van den Berg, K. J., Joe, C., van Engelen, J. G. M., and Brouwer, A. (1993). Glucuronidation of thyroid hormone in rat liver: effects of in vivo treatment with microsomal enzyme inducers and in vitro assay conditions. Endocrinology 135, 2177–2186. von Meyerinck, L., Hufnagel, B., Schmoldt, A., and Benthe, H. F. (1990). Induction of rat liver microsomal cytochrome P-450 by the pentabromo diphenyl ether Bromkal 70 and half-lives of its components in the adipose tissue. Toxicology 61, 259 –274. Wolf, J. (1998). Perchlorate and the thyroid gland. Pharmacol. Rev. 50, 89 –105. Zoeller, R. T., and Rudeen, P. K. (1992). Ethanol blocks the cold-induced increase in thyrotropin-releasing hormone mRNA in paraventricular nuclei but not the cold-induced increase in thyrotropin. Mol. Brain Res. 13, 321–330.