Environ. Sci. Technol. 2010, 44, 3149–3154
In Vitro Assay Shows That PCB Metabolites Completely Saturate Thyroid Hormone Transport Capacity in Blood of Wild Polar Bears (Ursus maritimus) A R N O C . G U T L E B , * ,†,‡,§,⊥ P E T E R C E N I J N , ‡ MARTIN VAN VELZEN,‡ ELISABETH LIE,§ ERIK ROPSTAD,§ J A N N E C H E U T N E S K A A R E , †,§ TINA MALMBERG,| ÅKE BERGMAN,| GEIR W. GABRIELSEN,¶ AND JULIETTE LEGLER‡ Department Environment and Agro-biotechnologies, Centre de Recherche PublicsGabriel Lippmann, 41, rue du Brill, L-4422 Belvaux, Grand-duchy of Luxembourg, National Veterinary Institute, Postboks 750 Sentrum, NO-0106 Oslo, Norway, Institute for Environmental Studies, VU University Amsterdam, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands, Norwegian School of Veterinary Science, P.O. Box 8146 Dep., NO-0033 Oslo, Norway, Department of Environmental Chemistry, Stockholm University, SE-106 91 Stockholm, Sweden, and Norwegian Polar Institute, NO-9296 Tromsø, Norway
Received October 5, 2009. Revised manuscript received March 8, 2010. Accepted March 10, 2010.
Persistent chemicals accumulate in the arctic environment due to their chemical reactivity and physicochemical properties and polychlorinated biphenyls (PCBs) are the most concentrated pollutant class in polar bears (Ursus maritimus). Metabolism ofPCBandpolybrominatedbiphenylether(PBDE)flame-retardants alter their toxicological properties and these metabolites are known to interfere with the binding of thyroid hormone (TH) to transthyretin (TTR) in rodents and humans. In polar bear plasma samples no binding of [125I]-T4 to TTR was observed after incubation and PAGE separation. Incubation of the plasma samples with [14C]-4-OH-CB107, a compound with a higher binding affinity to TTR than the endogenous ligand T4 resulted in competitive binding as proven by the appearance of a radio labeled TTR peak in the gel. Plasma incubation with T4 up to 1 mM, a concentration that is not physiologically relevant anymore did not result in any visible competition. These results give evidence that the binding sites on TTR for T4 in wild living polar bears are completely saturated. Such saturation of binding sites can explain observed lowered levels of THs and could lead to contaminant transport into the developing fetus.
* Corresponding author phone: e-mail: +352 470261 481; fax: +352 470264; e-mail:
[email protected]. ⊥ Centre de Recherche Public - Gabriel Lippmann. † National Veterinary Institute. ‡ VU University Amsterdam. § Norwegian School of Veterinary Science. | Stockholm University. ¶ Norwegian Polar Institute. 10.1021/es903029j
2010 American Chemical Society
Published on Web 03/26/2010
Introduction Polar bears (Ursus maritimus) have received extensive attention during recent years due to the fact that high levels of some pollutants such as polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs), or chlordanes were found in their tissues and organs (1-3) and the fear that rising temperatures may add additional stress to this apex predator species (4-6). Ultimately, accumulation is resulting in high exposure of top predators such as polar bears with persistent pollutants. Among the compounds still increasing are the widely used PBDE flame-retardants (7, 8), while PCB concentrations are still very high in polar bears (3, 9). Correlations of exposure to a range of persistent organic pollutants (POPs) and effects in polar bears, particularly on the immune system and reproduction have been shown (10-14). Other biological effects for which a correlation with POP exposure was shown in polar bears are decreased vitamin A and hormone levels (thyroid hormone, sex steroids) (2, 11, 15, 16). Metabolism of compounds like PCBs or PBDEs alters their kinetics (17, 18) and metabolite toxicity is becoming a factor of increasing scientific interest (19, 20). Hydroxylated PCB metabolites (OH-PCBs) were found in polar bears in very high concentrations (9, 21, 22) and are known to interfere with thyroid hormone (TH) transport in laboratory animals (23) and bind to recombinant TTR from arctic birds (24, 25). Thyroid hormone is important in development and in many regulatory functions in vertebrate species (26) and it is well established that environmental contaminants can affect TH status (23). Thyroid hormones, 3,3′,5,5′-tetrathyroxine (T4) and 3,3′,5-triiodothyronine (T3) are released and secreted into the plasma and are transported in mammals mainly bound to thyroxine-binding-globulin (TBG) or transthyretin (TTR) and the remainder bound to albumin with the relative importance of the three proteins depending on the species in question (27). TTR is synthesized both in the liver and the brain and is suggested to be involved in the transport of T4 over the bloodbrain barrier and the maternal to fetal transport through the placenta (23, 28, 29). Many hydroxylated metabolites of POPs show high binding affinity for TTR that can be even higher than those of the endogenous ligand T4. This results in inhibition of T4-binding to TTR and in addition a disruption of the complex formed by TTR with the retinol binding protein (RBP) altering both plasma levels of TH and retinol in mammals (29, 30). The in vivo effect of the high binding affinity of xenobiotics such as OH-CBs to TTR is hypothesized to result in retention of these compounds in plasma, facilitating transport of compounds over the placenta to the fetal compartment and decreased maternal and fetal plasma T4 levels (23). Indeed, experimental exposure of pregnant rats to PCBs or OH-CBs resulted in decreased T4 levels (23) and enrichment of both 4-OH-CB107 and 4-OH-CB187 in the fetus compared with the female (30). Importantly, OHPCBs have been found in wildlife (including polar bear) serum (9, 17, 22). As T4 levels in polar bear have been suggested to be altered due to exposure to PCBs (2, 15, 16), the hypothesis that hydroxylated metabolites present in polar bear plasma saturate and competitively block the binding sites for T4 on TTR was investigated. The binding of radiolabeled T4 to TTR in plasma was tested in polar bear plasma with human plasma as a reference and the binding of [14C]-4-OH-CB107 with known in vitro TTR binding activities higher than the endogenous ligand T4 were analyzed. In a final experiment, VOL. 44, NO. 8, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3149
increasing and nonphysiological high concentrations of unlabeled T4 were added to the incubation mixture in an attempt to see whether bound xenobiotics could be exchanged at all by the natural ligand.
Experimental Section Animals. Polar bears (three males: age in years: 6, 17, and 21; three females: age in years: 7, 7 plus two cubs, 8 plus two cubs) were located with a helicopter on Svalbard (1995-1996) and immobilized with Zoletil (Virbac, Carros, France) in a solution of 200 mg/mL at a dosage of 5-10 mg/kg of body mass (31). Blood samples were collected from the vena femoralis into evacuated, heparinised containers,and stored cool and dark until centrifugation (3000 rpm for 10 min) within 8 h of collection. Plasma was isolated and stored at -20 °C until analysis. A vestigial premolar tooth was extracted from all bears for age determination (32). Presence and age of cubs, and biologic parameters such as body length and axial girth were recorded. The Norwegian Animal Research Authority approved all capture and handling methods used. Chemicals. We purchased 2,2-bis(4-chlorophenyl)-1,1,1trichloroethane (4,4′-DDE), 2,2′,4,4′,5,5′-hexachlorobiphenyl (CB-153), 2,2′,3,3′,4,5,5′,6-octachlorobiphenyl (CB-198), 2,4,6tribromophenol (TBP); 2,4,5-trichlorophenol, and pentachlorophenol (PCP) from Dr. Ehrenstorfer (Augsburg, Germany), 2,2′,4,5′,6-pentachlorobiphenyl (CB-103) was purchased from Accustandard (New Haven, CT). n-hexane (ultraresi), sulfuric acid (p.a.), and methyl-tert-butyl ether (ultraresi) were purchased from Baker (Deventer, Netherlands). Ethereal diazomethane, used for the derivatization of phenolic compounds, was generated from N-methyl-N-nitroso-ureum and 50% w/v potassium hydroxide (p.a.) (Riedel-deHae¨n, Seelze, Germany) in diethyl ether (p.a.) (Merck, Darmstadt, Germany). Diazomethane is a hazardous compound that needs to be handled carefully and by authorized personnel only. The metabolites 2,3,3′,4′,5-pentachloro-4-biphenylol (4-OH-CB107), 2,2′,3,4′,5,5′-hexachloro-4-biphenylol (4-OHCB146), 2,2′,3,4′,5,5′,6-heptachloro-4-biphenylol (4-OHCB187), and 2,4-dibromophenoxy-4,6-dibromo-2-phenol (6OH-BDE47) and radiolabeled [14C]-4-OH-CB107 were synthesized and purified in house at the Department of Environmental Chemistry, Stockholm University as described elsewhere (33-35). Chemical Analysis. Polar bear plasma samples were analyzed for PCBs and organochlorine pesticides at two different laboratories: the Laboratory of Environmental Toxicology, Norwegian School of Veterinary SciencesNVH, Oslo, Norway, and for PCB metabolites and flame retardants at the Institute for Environmental StudiessIVM, Vrije Universiteit Amsterdam, The Netherlands. The analytical work was pursued in 2004. For comparison of the results from the two laboratories, DDE and CB-153 were analyzed independently and linear correlation coefficients and slopes for DDE (0.91, 0.99) and CB-153 (0.89, 0.99) indicate good agreement between measurements. Organochlorine Pesticide and PCB Analysis. The polar bear plasma samples in the present study were analyzed for POPs as part of 357 plasma samples from several studies on polar bears during 1998-2000 (12). Briefly, plasma samples (approximately 8 g) were extracted with cyclohexane and acetone and cleaned up with ultrapure sulphuric acid (36). The whole extracts were used to determine the percentage extractable fat gravimetrically. Aliquots of the final extract were analyzed by gas chromatography with electron capture detection (GC-ECD) instruments equipped with two capillary columns (SPB-5 and SPB1701, 60 m, 0.25 mm i.d. and 0.25 µm film layer; Supelco, Inc., Bellefonte, PA). Quantification was performed using CB 29, CB-112, and CB-207 as internal standards in each sample. Details on analytic procedures, equipment for 3150
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 8, 2010
chromatographic separation, detection, and calculation are given in ref 36. Standard procedures were used to ensure adequate quality assurance and control, and the precision, linearity, and sensitivity of the analyses were within the laboratory’s accredited requirements. A blank sample was included in each batch to test for interference. Reproducibility was continuously tested by analyzing the POP levels in the laboratory’s own reference sample (seal blubber). The mean percentage of the true value and its coefficient of variation (CV) was 103 and 11% (n ) 22). Recoveries and CV of individual pesticides and PCB congeners in spiked blood samples varied from 72 to 116% and 6.3 to 10.0%, respectively (n ) 83). Detection limits for individual POPs ranged from 0.002 to 0.02 ng/g ww. In all samples, the following POPs were determined: hexachlorobenzene (HCB), R-, β and γhexachlorocyclohexanes (HCHs), oxy-, trans-, and cis-chlordane, trans-nonachlor, p,p′-DDE, o,p′-DDD, p,p′-DDD, o,p′DDT, p,p′-DDT, mirex, and 34 PCB congeners (IUPAC numbers: 28, 31, 47, 52, 56, 66, 74, 87, 99, 101, 105, 110, 114, 118, 128, 136, 137, 138, 141, 149, 151, 153, 156, 157, 170, 180, 183, 187, 189, 194, 196, 199, 206, 209). Sixteen PCB congeners that could be quantified in all samples were selected for this study. These congeners were CB-47, 99, 118, 128, 137, 138, 153, 156, 157, 170, 180, 183, 187, 189, 194, and 206. In addition 4,4′-DDE, HCB, R-HCH, β-HCH, oxychlordane and transnonachlor were quantified in these samples. All concentrations are expressed on a ww basis (ng/g ww) to promote comparison to OH-CBs. Analysis of Halogenated Phenolic Compounds (HPCs). We analyzed 4-OH-CB107, 4-OH-CB146, 4-OH-CB187, 6-OHBDE47, and TBP in the samples. Plasma samples (approx 5 g) were extracted with hexane:MTBE (1:1 v/v) and separated into a neutral and phenolic fraction by liquid/liquid extraction with 0.5 M KOH in 50% ethanol (37). The phenolic fraction was first derivatized with diazomethane overnight. Both fractions were than treated with sulfuric acid and subsequently cleaned over a sulfuric acid silica column (1 g, 33% H2SO4 w/w). The final extracts were analyzed on a GC-µECD (Agilent 6890) equipped with a CPSil8CB column (Varian, 25 m × 0.15 mm × 0.12 µm). Quantitation was performed using 2,4,5-trichlorophenol and 4-OH-CB193 as internal standards. A blank human plasma sample and an internal human reference sample were simultaneously analyzed for quality control. No OH-PCBs were found in the blank, and the OH-PCBs found in the reference sample were within the limits of the control chart. The recoveries of 4-OH-CB193 in the blank and reference sample were between 80 and 100%, whereas it varied between 600 and 3000% in the polar bear samples. This was recently explained by the discovery that 4-OH-CB193 is in fact a metabolite present in the polar bear plasma (22, 38). The reported concentrations are not corrected for recovery. Plasma Protein Separation and [125I]-T4 Competition Binding. To determine the binding of persistent pollutants present in plasma and their binding to proteins ex vivo, plasma proteins in samples from six polar bears were separated by polyacrylamide gel electrophoresis (PAGE) as described (29) about 6 years after the blood sampling had occurred. Binding affinities to TTR stored for similar periods did not indicate changes due to aging. In addition, the determination of [125I]-T4-competitive binding to specific plasma proteins was performed as described (29, 39). In short, plasma samples for gel slices (40 µL) were mixed 1:2 with a 50 mM Tris/38 mM glycine buffer (pH 8.3) containing 4.5% saccharose. Plasma samples for [125I]-T4 competition binding (25 µL) were incubated overnight with 100.000 cpm [125I]-T4 (in 5 µL 50 mM tris-HCL buffer, pH 8.0) at 4 °C. Aliquots of 20 µL were run on a 10% native separating gel (Protean II Ready Gel 20-Well Gels, Bio-Rad, Veenendal, Netherlands) for 4 h at 4 °C at a constant current of 50 mA. Each gel also
FIGURE 1. Distribution of [125I]-T4-derived radioactivity after incubation with human TTR (1a, left) and human plasma (1b, right) after PAGE. 1 ) albumin; 2 ) thyroxine-binding-globulin (TBG); 3 ) transthyretin (TTR); 4 ) free 125I. contained plasma samples for protein staining (5 µL) and pure BSA and human TTR as a reference. After electrophoresis, the part of the gel containing the reference proteins was stained in 0.04% Coomassie Brilliant Blue in 3.5% perchloric acid for 60 min, and subsequently destained with 7% acetic acid for 24 h to determine the position of the proteins on the gel. The part of the gel for radioactivity measurements was frozen on the glass plate at -20 °C overnight. The acrylamide gel was subsequently sliced into 2 mm pieces by a standardized procedure. Gel slices containing plasma samples incubated with [125I]-T4 were placed in RIA tubes and counted directly in a γ-counter (Cobra Auto Gamma counter, Canberra Packard). The PAGE gel profile was made by plotting the [125I]-T4-radioactivity against the migration distance on the gel, thereby showing relative migrations distances. In Vitro Competition Binding Studies with [14C]-Labeled Compounds. The in vitro potency of [14C]-4-OH-CB107 to compete with compounds already present in the wild living animals and occupying T4-binding sites on TTR was performed as previously described (39). Plasma samples for [14C]4-OH-CB107-T4 competition binding (25 µL) were incubated overnight with 10 000 cpm [14C]-labeled compound (in 5 µL 50 mM Tris-HCL buffer, pH 8.0) at 4 °C. Aliquots of 20 µL were run on a 10% native separating gel and further treated as described above. Gel slices containing [14C]-4-OH-CB107 derived radioactivity were first eluted by incubating the gel slices in tubes with 1 mL water overnight at 4 °C. Four mL of scintillation fluid was added (Ultima Gold, Packard) the next day and the amount of radioactivity in each gel slice was quantified by liquid scintillation. The PAGE gel profile was again made by plotting the radioactivity against the migration distance on the gel, thereby showing relative migrations distances.
FIGURE 2. Distribution of [125I]-T4-derived radioactivity after incubation with polar bear plasma after PAGE. 1 ) albumin; 2 ) thyroxine-binding-globulin (TBG); 4 ) free 125I; please note that transthyretin (TTR, peak 3) is missing.
Results Six plasma samples (three males, three females; age range 6-21 years) were available from free ranging polar pears from Svalbard, Norway. Protein separation of samples following incubation with [125I]-T4 revealed the expected binding of [125I]-T4 to pure human TTR in buffer solution (Figure 1a), as well as binding to TTR, thyroxin binding globulin (TBG) and albumin, respectively, in the case of incubation of [125I]-T4 with human plasma (Figure 1b). Peaks were identified based on comigration of unlabeled reference compounds. Unbound radioactivity is found at the front of the gel. These two incubations were used as references and method controls. In the case of polar bear plasma samples, PAGE separation following plasma incubation with [125I]-T4 revealed no binding of [125I]-T4 to TTR, with the majority of radioactivity bound to albumin/TBG and again unbound radioactivity on the front (Figure 2). This result was identical in all samples investigated (data not shown).
FIGURE 3. Distribution of [14C]-derived radioactivity after in vitro incubation of polar bear plasma with [14C]-4-OH-CB107 after PAGE. 1 ) albumin; 2 ) thyroxine-binding-globulin (TBG); 3 ) transthyretin (TTR). In order to investigate whether the results as shown in Figure 2 imply that TTR is not present in polar bears at all or if the binding site on TTR is blocked by compounds with binding affinities higher than the endogenous ligand T4, in vitro incubations of polar bear plasma with [14C]-4-OH-CB107 with known high potency to bind to TTR were performed. This in vitro [14C]- 4-OH-CB107 competition with polar bear plasma and separation of plasma proteins by PAGE showed two radioactive peaks for [14C]-4-OH-CB107, one at the position of TTR and one representing radioactivity bound to albumin/TBG (Figure 3). This indicates that [14C]-4-OHCB107 known to bind with high affinity to TTR is able to VOL. 44, NO. 8, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3151
TABLE 1. Concentrations (ng.g-1Wet Weight) of Contaminants in Polar Bear Plasma from Three Males and Three Femalesa
FIGURE 4. Distribution of [125I]-T4-derived radioactivity after in vitro incubation of polar bear plasma with [125I]-T4 and T4 (total concentration of T4 ) 1 mM) after PAGE. 1 ) albumin; 2 ) thyroxine-binding-globulin (TBG); please note that transthyretin (TTR, peak 3) is missing. compete with compounds already present after in vivo exposure or the binding site. To test the hypothesis that the binding sites on TTR in the polar bear plasma samples investigated are indeed occupied by compounds with a higher binding affinity than T4, an in vitro incubation competition assay with increasing concentrations of unlabeled-T4 ranging from nM to mM in the presence of constant amounts of [125I]-T4 was performed. Even in the presence of 1 mM T4, which is 3 orders of magnitude higher than in the competition assays usually performed (Figure 2), no radioactivity was found on the position of TTR (Figure 4). At this stage, chemical analyses revealed the presence of 16 PCB congeners in the plasma of the polar bears, whereas 18 congeners had concentrations lower than limit of detection with the method used for the analysis of ΣPCB34. CB153 and CB180 were the dominating PCB congeners accounting for 62 ( 2.7% of the ΣPCB34 (Table 1). PCB congeners 99, 138, 170, and 194 also had concentrations of more than 1 ng/g ww in most samples and together with CB153 and CB180, they accounted for 94.8 ( 0.8% of the sum concentrations of PCBs measured in plasma samples. Concentrations of CB153 ranged from 19.6 ng/g ww to 125 ng/g ww and CB180 ranged from 15.9 ng/g ww to 81.0 ng/g ww in the six analyzed polar bear plasma samples. The sum of the three hydroxylated PCB metabolites (ΣOH-CBs), 4-OH-CB107, 4-OH-CB146, and 4-OH-CB187 ranged from 73.2 ng/g ww to 377 ng/g ww. This concentration was identical or higher than those of the PCBs ranging from 0.98 to 4.3 times the ΣPCB34 concentrations (Table 1). Concentrations of the other compounds analyzed were often below the limit of detection (6-OH-BDE47, PCP, TBP) or present only in low concentrations (Table 1).
Discussion A range of phenolic compounds, among which OH-CBs metabolites, has been shown to bind to TTR from fish, amphibians, birds, and mammals (30, 40-42). High concentrations of OH-CB congeners were found in the polar bear plasma samples from Svalbard. These compounds are known to have binding affinities to TTR higher than the endogenous ligand T4 (29, 30, 39) and [125I]-T4 (10 µM) did not show any binding to TTR when incubated with plasma from free-ranging polar bears from Svalbard, Norway (Figure 2). In contrast, [14C]-4-OH-CB107 that has a higher binding affinity than T4 was able to bind to TTR in polar bear plasma (Figure 3). Contrary to that observation incubation of polar 3152
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 8, 2010
parameter
mean
range
gender age (years) PCB 47 PCB 99 PCB 118 PCB 128 PCB 137 PCB 138 PCB 153 PCB 156 PCB 157 PCB 170 PCB 180 PCB 183 PCB 187 PCB 189 PCB 194 PCB 206 4-OH-CB 107 4-OH-CB 146 4-OH-CB 187 6-OH-BDE 47 TBP p,p′-DDE PCP HCB R-HCH β-HCH oxychlordane trans-nonachlor
3/3 11 0.5 13.4 0.2 0.3 0.9 12.9 57.3 1.7 1.6 15.9 39.9 0.8 0.1 0.4 7.7 1.2 13.6 166 68.7
