Assessment of Impact of Internal Exposure to PBDEs on Human

May 12, 2012 - Department of Pediatrics, Soonchunhyang University Hospital, Jangjeon-dong, Geumjeong-gu, Busan 609-735, Republic of Korea...
1 downloads 0 Views 1MB Size
Article pubs.acs.org/est

Assessment of Impact of Internal Exposure to PBDEs on Human Thyroid FunctionComparison between Congenital Hypothyroidism and Normal Paired Blood Un-Jung Kim,† Min-Young Kim,† Yong-Hee Hong,‡ Dong-Hwan Lee,‡ and Jeong-Eun Oh†,* †

Department of Civil and Environmental Engineering, Pusan National University, Jangjeon-dong, Geumjeong-gu, Busan 609-735, Republic of Korea ‡ Department of Pediatrics, Soonchunhyang University Hospital, Jangjeon-dong, Geumjeong-gu, Busan 609-735, Republic of Korea S Supporting Information *

ABSTRACT: In this work, we investigated exposure levels, distribution patterns, and potential harmful impacts of polybrominated diphenyl ethers (PBDEs) on thyroid hormone activity in 26 children with congenital hypothyroidism and their mothers’ pair and 12 normal control pairs. The average concentration of PBDEs in congenital hypothyroidism (median: 22.16 ng/g lipid) was higher than in normal controls (median: 14.76 ng/g lipid), but there was no statistical difference between the two groups. The BDE congeners were dominated by penta- to hepta-BDEs, but the greater brominated congeners (e.g., BDE 197, 196, 207, and 208) were relatively abundant in congenital hypothyroidism. BDE 138 was only observed in the congenital hypothyroidism cases. The maternal transfer and transport ratio of individual BDE congeners was shown for BDE 28 (0.588, p < 0.001), BDE 47 (0.564, p < 0.001), BDE 49 (0.712, p < 0.001) and BDE 119 (0.477, p = 0.002). The thyroid hormones were most obviously influenced by the internal exposure to PBDEs in normal mothers, showing a positive relationship with TSH (0.641 with BDE 154; 0.591 with BDE 153) and FT4 (0.584 with BDE 49; 0.572 with BDE 66) and a negative relationship with T3 (−0.577 with BDE 154) in the normal infants group. No significant correlations were observed in the congenital hypothyroidism cases. possibility of thyroid disruption by PBDEs.6−10 In our previous study,10 we identified a statistically significant correlation between thyroid hormones and PBDE concentration in human blood. However, contradictory results were reported in a case-by-case study,11 demonstrating that it is difficult to identify and define the harmful impacts of PBDE exposure on endocrine disorders. So to further understand these harmful impacts, we conducted a mother−infant pair determination study to identify the potential effect of exposure to PBDEs on human thyroid hormone activity and the potential for transfer of BDE congeners from mothers to infants. In addition, we also evaluated the differences in PBDE internal exposure patterns between a normal group and patients with congenital hypothyroidism, which is a typical metabolic disease. Congenital hypothyroidism is a common pediatric endocrine disorder and occurs in 1 of every 3000 or 4000 children. In addition, this disease is suspected to be derived or caused by EDC exposure during the prenatal period.12 However, there is currently not enough background data to assess whether the risk of PBDEs as endocrine disruptors in the human body is a

1. INTRODUCTION Although there are many types of endocrine-disrupting chemicals (EDCs), persistent organic pollutants (POPs) are regarded as high-risk potential EDCs in human exposure because of their physicochemical properties, such as persistency, the possibility of cross-placental transfer, and ill-defined reactions in the human body related to endocrine function.1 Prenatal exposure to EDCs is considered potentially dangerous and can cause endocrine and other related malfunctions that can influence the development of the fetus and can retain their deleterious effects for their lifetime. In addition, because the nervous, immune, reproductive, and digestive systems of the fetus are not fully developed, the potential disturbance from prenatal exposure to these chemicals is more critical than for adults.2 Therefore, it is important to monitor internal exposure to EDCs in both the mother and fetus, and to assess health status (e.g., thyroid function) during pregnancy, which directly affects the health and development of the fetus and infant. Among EDCs, polybrominated diphenyl ethers (PBDEs), recently designated as regulated compounds at the 2009 Stockholm Convention,3 are among the most widely used flame retardant compounds and have a molecular structure similar to thyroid hormones (THs).4 In addition, it is believed that PBDEs may mimic or compete with THs, which would disrupt thyroid function;5 however, few studies have investigated the © 2012 American Chemical Society

Received: Revised: Accepted: Published: 6261

October 31, 2011 May 8, 2012 May 12, 2012 May 12, 2012 dx.doi.org/10.1021/es2038678 | Environ. Sci. Technol. 2012, 46, 6261−6268

Environmental Science & Technology

Article

Inc., Bellefonte, PA, U.S.) was used for the extraction and cleanup. 2.3. Sample Collection, Storage and Description. Blood samples were collected from 26 infants with congenital hypothyroidism (congenital hypothyroidism is described in the Supporting Information (SI), S1) and their mothers and from 12 normal infants and their mothers, making the total number of collected serum samples 76. Congenital hypothyroidism can occur from altered thyroid hormonal function during pregnancy; however, this disorder may also occur regardless of the health status of the mother. Thus, only congenital hypothyroidism was considered a confirmed metabolic disease in this study. Blood samples from all 38 mother−infant pairs were collected at Soonchunhyang Hospital, Seoul, South Korea, from November 2009 to May 2010, and were collected from volunteers that had agreed to participate. The blood serum samples for infants and mothers were collected during the first 1 to 3 months after birth, with the exception of two congenital hypothyroidism infants from whom collection was delayed 18 and 24 months. The donors were provided a full explanation of the purpose of this research, and the study was approved by a local committee for medical ethics. The ages of sample donors ranged from 23 to 37 years for mothers and from 1 to 29 months for babies. The body weight of babies ranged from 2.21 to 4.20 kg and the body weight of the mothers ranged from 50 to 85 kg (Table S1 of the SI). Total cholesterol (TC), which was based on the sum of the low density lipid (LDL), high density lipid (HDL), and triglyceride (TG), was measured to estimate lipid content in serum blood. All of the samples collected from the donors were moved from the hospital to the laboratory in an ice-packed box and stored at −20 °C before analysis. 2.4. Analytical Procedures. Sample Pretreatment, Extraction and Cleanup. The PBDEs were analyzed as described by Covaci and Voorspoels (2005) with minor modifications.12,13 Briefly, target internal standard mixture was added and homogenized overnight after addition of serum. Sample extraction was performed by solid phase extraction (SPE) with OASIS HLB (6 cm3, 200 mg) with 2 × 3 mL DCM:Hx (1:1) and cleaned up with multisilica gel using 8 mL of DCM. Further details on this procedure are given in the SI (S2). Thyroid Hormone Examination. Serum levels of free thyroxine (FT4), total triiodothyronine (T3), and thyroid stimulating hormone (TSH) of both mother and infant were measured using radioimmunoassay kits (Diagnostic Products Corp., Los Angeles, CA, U.S.). These measurements were taken through neonatal screening tests for infants and after delivery for mothers. The detection limits of the kits in this study were 1 μg/dL and 0.02 μg/dL for total T4 and TSH, respectively. 2.5. Instrumental Analysis. Separation of PBDEs was achieved by HRGC/HRMS (Agilent 6890 GC/JEOL 800 MSD) with a diphenyl (5%)-dimethylpolysiloxane (95%)coated 15 m × 0.25 mm ×0.10 μm column. The flow rate was 1 mL/min using high-quality Helium (99.9999% purity) and the injection volume was 2 μL. HRMS was performed on magnetic-electric double focusing lenses operating in positive EI mode. The temperature gradient started at 110 °C (5 min), then increased to 200 °C at a rate of 40 °C/min for 5.5 min, and then to 320 °C (10 °C/min, 5 min). Electron ionization energy was set at 36 eV for octa- to deca-BDEs, which corresponded to the most sensitive value for detecting higher molecular PBDEs used in electron impact ionization, while a

cause of congenital hypothyroidism or other types of endocrine disorders, especially in the fetus or infant, who are the most vulnerable to EDCs. The only available reports that evaluated the human body reaction to PBDE exposure were conducted using in vitro acute toxicity tests, which revealed a statistical relationship between controlled PBDE exposure (e.g., injection or dose experiments) to animals and thyroid hormone levels.6,7,11 In this study, we wanted to confirm previously observed correlations between internal exposure to PBDE and TH activity in umbilical cord blood and breast milk from Korean women10 in congenital hypothyroidism cases. To achieve this, we conducted a comparison study with blood from congenital hypothyroidism infants and their mothers and normal blood pairs to more accurately assess the harmful impact of PBDE exposure on the human body. To the best of our knowledge, this is the first study in which the concentration and distribution patterns of 27 BDEs from mono- to deca-BDE congeners are compared in normal infants and congenital hypothyroidism infants to assess the differences in internal levels of PBDEs and to estimate the potential harmful impact of PBDEs on thyroid function.

2. MATERIALS AND METHODS 2.1. Target Compounds. In this study, all of the major PBDE congeners (e.g., 27 from mono- to deca- BDEs) originating from potential sources of food stuff, commercial products, and house dust, or other ambient environmental matrices were analyzed in blood sera. The mono- to decaPBDEs included: 4-MoBDE (#3), 2,4-DiBDE (#7), 4,4′-DiBDE (#15), 2,2′,4-TriBDE (#17), 2,4,4′-TriBDE (#27), 2,2′,4,4′TeBDE (#47), 2,2′,4,5′-TeBDE (#49), 2,3′,4,4′-TeBDE (#66), 2,3′,4′,6-TeBDE (#71), 3,3′,4,4′-TeBDE (#77), 2,2′,3,4,4′PeBDE (#85), 2,2′,4,4′,5-PeBDE (#99), 2,2′,4,4′,6-PeBDE (#100), 2,3′,4,4′,6-PeBDE (#119), 3,3′,4,4′,5-PeBDE (#126), 2,2′,3,4,4′-PeBDE (#138), 2,2′,4,4′,5,5′-HeBDE (#153), 2,2′,4,4′,5,6′-HeBDE (#154), 2,3,3′,4,4′,5-HeBDE (#156), 2,2′,3,4,4′,5′,6-HpBDE (#183), 2,2′,3,4,4′,6,6′-HpBDE (#184), 2,3,3′,4,4′,5′,6-HpBDE (#191), 2,2′,3,3′,4,4′,5,6′-OcBDE (#196), 2,2′,3,3′,4,4′,6,6′-OcBDE (#197), 2,2′,3,3′,4,4′,5,5′,6-NoBDE (#206), 2,2′,3,3′,4,4′,5,6,6′-NoBDE (#207), and deca BDE (#209). 2.2. Standards and Reagents. 13C12-4-MoBDE (#3 L), 13 C12-4,4′-DiBDE (#15 L), 13C12-2,4,4′-TriBDE (#28 L), 13C122,2′,4,4′-TeBDE (#47 L), 13C12-2,2′,4,4′,6-PeBDE (#100 L), 13 C12-2,2′,4,4′,5-PeBDE (#99 L), 13C12-2,2′,4,4′,5,6′-HeBDE (#154 L), 13C12-2,2′,4,4′,5,5′-HeBDE (#153 L) and 13C122,2′,3,4,4′,5′,6-HpBDE (#183 L), 13C12-2,2′,3,3′,4,4′,6,6′-OcBDE (#197 L), 13C12-2,2′,3,3′,4,4′,5,6,6′-NoBDE (#207 L), and 13C12DeBDE (#209 L) were used as internal standards (MBDEMXE), while 13C12-2,2′,3,4,4′,5′-HeBDE (#138 L) was used as a recovery standard (PBDE ISS); these compounds were obtained from Wellington Laboratories (Guelph, Ontario, Canada). All solvents used, including acetone, n-hexane (Hx), dichloromethane (DCM), and methanol (MeOH), were pesticide grade (J.T. Baker Co., Phillipsburg, NY, U.S.). Concentrated sulfuric acid (98%), formic acid (99%), and silica gel were from Merck Co. (Darmstadt, Germany). Anhydrous sodium sulfate (Na2SO4) was from Wako Co. (Tokyo, Japan). A Visiprep SPE Vacuum manifold (Supelco 6262

dx.doi.org/10.1021/es2038678 | Environ. Sci. Technol. 2012, 46, 6261−6268

Environmental Science & Technology

Article

9.30 ng/g lipid, and Norway: ∑3−7BDEs: 4.72 ng/g lipid).14 The wider range of PBDE homologues from mono- to decaBDEs, including the seven bioaccumulative congeners (BDE 28, 47, 100, 99, 154, 153 and 183) were analyzed in this study. Among our 23 detected target BDE congeners in this study, the seven bioaccumulative congeners along with BDE 49 and 119 were abundant. Specifically, in the normal infants’ mothers, BDEs were in the following order, 183 ≈ 153 > 49 > 47 > 154 > 119, and in the normal infants, BDEs were in the following order, BDE 183 ≈ 49 > 153 > 119 > 154 > 47. As described, most previous studies of BDEs in human blood serum analyzed only seven BDE congeners (6, 7, 9, 10, 14, 20, 21, 24, 26), making it impossible to compare the whole distribution pattern, but Toms et al. also reported the relatively higher detection of BDE 49 and 119 in the human body.15 3.1.2. PBDEs in Congenital Hypothyroidism Pairs (n = 26). Total PBDE concentration from mono- to deca-BDEs in the 26 congenital hypothyroidism pairs ranged from 2.220 to 1563 ng/ g lipid (mean: 70.92 ng/g lipid), and Σ3−7BDEs concentration ranged from 3.692 to 1563 ng/g lipid (mean: 83.58 ng/g lipid) for mothers (n = 26) and from 2.216 to 860.2 ng/g lipid (58.25 ng/g lipid) for infants (n = 26). There are no previous studies on PBDEs in congenital hypothyroidism patients; thus, we could not adequately compare the concentration levels measured in this study. Compared to that in the normal pairs in this study, the average concentration in the congenital hypothyroidism group was higher, but this difference was not statistically significant. Even though BDE 209 was below the LOQ range (10 pg/ mL) for all of the serum samples in this study, octa- and nonaBDEs were observed in the congenital hypothyroidism cases. Octa-BDEs (e.g., BDE 197 and 196) in congenital hypothyroidism cases were 1.7 to 5.2 times higher than in normal cases. Nona-BDEs (e.g., BDE 207 and 208) were detected only in the congenital hypothyroidism cases and provided the highest contribution to the total BDE concentration of the sample. In addition, the level of BDE 138 was elevated in the congenital hypothyroidism cases but was never detected in the normal cases. In congenital hypothyroidism, as the total PBDE concentration increased, the fraction of heavier BDE congeners increased and the fraction of the usual dominant congeners (e.g., BDE 47, 99, and 100) decreased. Especially, BDE 138, which is not usually reported in the human body, constituted 30% of the BDEs in the congenital hypothyroidism cases. Due to a lack of previous studies on the effects of BDE 138 on the human body or thyroid disruption, we do not know the reason for this high BDE 138 concentration in the congenital hypothyroidism cases; however, this distinctive pattern of BDE distinguished endocrine diseased and normal people. Excluding BDE 138, the 27 BDE congeners were mostly penta-, hexa- and hepta-BDEs, which was similar to the normal pairs. The order of BDEs in the mother’s group was 183≈153 > 119 > 49 > 17 and the order of BDE in the infant’s group was 49 > 183 ≈ 119 > 153 > 47. 3.1.3. Comparison between Normal and Congenital Hypothyroidism Cases. To determine the statistical differences of internal exposure of PBDEs between the normal and congenital hypothyroidism cases, principal component analysis (PCA) was used for the 76 samples. With input data consisting of 19 PBDE congener fractions for each sample, excluding eight congeners with detection frequencies below 30%, two principal components were extracted as PC 1 (33.4%) and PC 2

higher value (42 eV) was preferred for lower molecular monoto hepta-BDEs. 2.6. Quality Assurance and Quality Control (QA/QC). For every batch, procedural blanks, consisting of water, were included to obtain quality control values and to check for cross contamination during the experiment. The average detection concentration of target compounds in the procedural blank was mostly below 10% of the lowest concentration obtained in the blood samples. Detailed QA/QC results are described in the SI (S3). 2.7. Survey and Statistical Analysis. To identify correlations between the internal exposure levels of PBDEs and human metabolic conditions, or other health status indicators, statistical analyses were conducted. SPSS14.0K was used for statistical analysis and all statistical results were based on basic information. To assess the PBDEs exposure, each donor′s basic information was obtained from a survey that included dietary habits, type of hobbies, use frequency of instrument or apparatus, present and past career, and type of working environment. A detailed explanation can be found in the SI (S4). All of the data and information used in our study were anonymous to protect the rights of the mothers and infants. To determine the statistical differences of internal exposure to PBDEs between the normal and congenital hypothyroidism cases, principal component analysis (PCA) was performed with 19 BDE congeners, excluding eight congeners that had detection frequencies below 30% for variables, and 76 serum samples. The concentration below the limit of quantification (LOQ) was replaced with 1/2 LOQ to prevent distortion of the null data, and the data were then normalized to the total BDE concentration. The possibility of maternal transfer of PBDEs was estimated by a paired t test, and the relationship between PBDE exposure and the thyroid hormone level was examined by Pearson’s two-sided correlation.

3. RESULTS AND DISCUSSION 3.1. Concentration Level and Distribution Pattern of PBDEs. In this study, 27 mono- to deca-BDE congeners were analyzed in 76 blood serum samples collected from congenital hypothyroidism infants, their mothers, and normal infants and their mothers. Total PBDE concentration from mono- to decaBDEs in 76 samples ranged from 1.610 to 1563 ng/g lipid (mean: 63.77 ng/g lipid). The detailed concentration profile of each BDE congener is shown in Table 1. 3.1.1. PBDEs in Normal Pairs (n = 12). Total PBDE concentration from mono- to deca-BDEs in 12 normal pairs ranged from 1.610 to 252.9 ng/g lipid (mean: 37.74 ng/g lipid), but after excluding three outlying samples (exceeded 110 ng/g lipid), the PBDE levels in all samples were below 50 ng/g lipid. Because many previous studies report seven bioaccumulative congeners,14 we also used the concentrations of seven bioaccumulative tri- to hepta-BDEs for equal comparison with these studies. The Σ3−7BDEs in blood serum ranged from 1.559 to 50.85 ng/g lipid (mean: 17.72 ng/g lipid) in mothers (n = 12) and from 0.829 to 252.9 ng/g lipid (mean: 46.29 ng/g lipid) in infants (n = 12). The observed concentration level of tri- to hepta-BDEs in normal mothers was compared with other studies, which showed that the levels in this study were lower than those reported in a 2001 study conducted in the U.S. (BDE47: 28.0 ng/g lipid, ∑3−7BDEs: 41.1 ng/g lipid),14 and slightly higher than the values reported in Europe (Netherlands: ∑3−7BDEs: 6263

dx.doi.org/10.1021/es2038678 | Environ. Sci. Technol. 2012, 46, 6261−6268

a

26

12

26

mothers of congenital hypothyroidism

normal babies

congenital hypothyroidism babies

6264 4.218 (