Article pubs.acs.org/est
Tissue-Specific Accumulation, Depuration, and Transformation of Triphenyl Phosphate (TPHP) in Adult Zebrafish (Danio rerio) Guowei Wang, Zhongkun Du, Hanyan Chen, Yu Su, Shixiang Gao,* and Liang Mao State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing 210093, China S Supporting Information *
ABSTRACT: Understanding bioaccumulation and metabolism is critical for evaluating the fate and potential toxicity of compounds in vivo. We recently investigated, for the first time, the bioconcentration and tissue distribution of triphenyl phosphate (TPHP) and its main metabolites in selected tissues of adult zebrafish. To further confirm the metabolites, deuterated TPHP (d15-TPHP) was used in the exposure experiments at an environmentally relevant level (20 μg/L) and at 1/10 LC50 (100 μg/L). After 11−14 days of exposure to 100 μg/L of d15-TPHP, the accumulation and excretion of d15TPHP reached equilibrium, at which point the intestine contained the highest d15-TPHP (μg/g wet weight, ww) concentration (3.12 ± 0.43), followed by the gills (2.76 ± 0.12) > brain (2.58 ± 0.19) > liver (2.30 ± 0.34) ≫ muscle (0.53 ± 0.04). The major metabolite of d15-TPHP, d10-diphenyl phosphate (d10-DPHP), was detected at significantly higher contents in the liver and intestine, at levels up to 3.0−3.5 times those of d15-TPHP. The metabolic pathways of TPHP were elucidated, including hydrolysis, hydroxylation, and glucuronic acid conjugation after hydroxylation. Finally, a physiologically based toxicokinetic (PBTK) model was used to explore the key factors influencing the bioaccumulation of d15-TPHP in zebrafish. These results provide important information for the understanding of the metabolism, disposition, and toxicology of TPHP in aquatic organisms.
■
INTRODUCTION
Several studies have shown that TPHP induces toxicological effects on embryonic development,23,24 lipid metabolism,5 thyroid hormone secretion,25,26 the neurological system27,28 and the immune system.29,30 Leisewitz et al.31 stated that TPHP is acutely toxic to aquatic organisms, and Lassen and Lokke13 showed that TPHP was the most acutely toxic triaryl phosphate to fish, shrimp, and daphnia. The toxicity index (96 h LC50) of TPHP for rainbow trout is 0.36 mg/L, which is lower than the estimated value of 0.84 mg/L based on its hydrophobicity (log Kow: 4.59),32,33 indicating that TPHP is more acutely toxic to rainbow trout than expected. The disposal of TPHP-treated vinyl fabric upholstery into a pond would result in a sufficiently high concentration of TPHP to poison fish.13 Investigations on the in vivo accumulation and transformation of a compound are essential for evaluating its toxicity. To date, only limited studies have focused on the bioaccumulation of TPHP in wild fish.34,35 Lassen and Lokke13 reported that the maximum level of TPHP detected in a fish body was 600 μg/kg, and Evenset et al.36 reported levels of 5.7−13 μg/kg in fish liver and 0.3−3.2 μg/kg in fish muscle, respectively. Kim et al.16
Because brominated flame retardants (BFRs) are being gradually phased out on account of their persistence, bioaccumulation, and toxicity,1,2 the use of alternative and replacement flame retardants, such as organophosphate flame retardants (OPFRs), has increased significantly. As one of the most popular OPFRs used in unsaturated polyester resins, triphenyl phosphate (TPHP) is widely used in polyvinyl chloride, electronic thermoplastics, casting resins, and commercial mixtures.3−6 Because TPHP is an additive flame retardant and does not form chemical bonds with substrate materials, it can be easily released into the aquatic environment.6−8 In fact, TPHP has been found in various natural and engineered aquatic systems, including surface water, groundwater, wastewater, and even drinking water.7,9−12 TPHP has even been reported in rivers at a level of 7.9 μg/L in Denmark13 and has been detected as the dominating OPFR in three wastewater treatment plants in Norway at concentrations ranging from 3.1 to 14 μg/L in the influents and 1.7 to 3.5 μg/L in the effluents.14 Recently, Li et al.15 detected TPHP as one of the predominant OPFRs in drinking water from eight cities in China and found that the level was up to 84.1 ng/L. TPHP has also been detected frequently in humans and aquatic organisms,16−22 which has aroused great attention regarding its potential adverse effects. © 2016 American Chemical Society
Received: Revised: Accepted: Published: 13555
September 16, 2016 November 13, 2016 November 15, 2016 November 15, 2016 DOI: 10.1021/acs.est.6b04697 Environ. Sci. Technol. 2016, 50, 13555−13564
Article
Environmental Science & Technology reported that the whole-body levels of TPHP in fish from Manila Bay ranged from 91 ng/g lipid weight (lw) (demersal fish) to 350 ng/g lw (pelagic fish). Ma et al.37 also found that the muscle contents of TPHP were up to 45.7 ng/g lw in catfish (Clarias fuscus) and grass carp (Ctenopharyngodon idellus) from the Pearl River Delta region in southern China. A few researchers have examined the metabolites of TPHP hydrolysis in urine samples and other phase I or phase II metabolites in the microsomes of terrestrial organisms, such as rats or humans.18,38−41 However, the transformation and fate of TPHP in fish is largely unknown. This may be partly attributed to the complicated procedures required for such studies and the lack of analytical standards for the analysis of TPHP metabolites in tissue samples. In this study, a protocol for the extraction and quantification of TPHP and its major metabolites from fish tissues and water samples was developed. Based on this reliable analytical method, we examined the accumulation and depuration of TPHP in the brain, intestine, gill, liver, and muscle of zebrafish at two exposure levels. Potential metabolites in the liver and water samples were identified using high-resolution liquid chromatography−qualitative quadrupole time-of-flight mass spectrometry (LC−QTOF), and the tissue distribution of the main metabolites during equilibrium and the depuration period were determined using quantitative liquid chromatography−tandem mass spectrometry (LC−MS/MS) in the multiple reaction monitoring (MRM) mode. Finally, a physiologically based toxicokinetic (PBTK) model was used to calculate the partition coefficients of d15TPHP in various tissues and to evaluate the effects of transformation on bioaccumulation.
liter of water. During the exposure period, half of the test solution was renewed daily. Fish were fed with pellet food and checked daily for abnormal behavior, disease, and mortality. Uneaten food and feces were siphoned from the tanks shortly after feeding to avoid dietary uptake and adsorption. Solvent control (0.01% DMSO) experiments were also conducted. Triplicate experiments were conducted for each treatment group. After exposure for 3, 7, 11, 14, 16, and 19 days, triplicate samples (each sample with 20 fish) in each treatment group were randomly removed and euthanized instantly. The weight and length of each individual fish were measured. Fish was dissected, and major tissues, including the brain, gills, intestines, liver, and muscle (0.5 g of muscle for analysis), were separated, weighed, and stored at −80 °C until analysis. After the fish were removed, triplicate water samples were collected for determination of the d15-TPHP concentration in the aqueous phase. Depuration experiments were conducted in a manner similar to that employed during the uptake experiment. After exposure for 19 days, fish were transferred to clean water and cultured for 7 days with daily renewal of water for depuration. Triplicate water and fish samples in each treatment group were sampled after 3 and 7 days of depuration. Extraction of d15-TPHP and Metabolites from Tissues. To explore an economic, fast, and effective method to quantify TPHP and its metabolites in fish tissues, we optimized the extraction and cleanup procedures with various solutions and finally developed a reliable method for analyzing TPHP and its major metabolite as described below. The detailed procedures used for tissue extraction are provided in Figure S1A,B. The sampled tissues from 20 fish were separated into two groups to measure the concentrations of d15-TPHP and metabolites. Tissues from 10 fish were first spiked with 20 ng of nondeuterated TPHP and homogenized in 3 mL of C-hex/ Etac (1:1 v/v) and anhydrous sodium sulfate (4 g of Na2SO4 per 1 gram of wet weight) with a homogenizer (WiseTis HG-15A, Daihan Scientific Co. Ltd.). The homogenates were ultrasonicated (15 min, 20 °C) and centrifuged (5 min, 4000 rpm), and then, the supernatants were collected. This procedure was repeated three times for each sample. The supernatants of each sample were combined and pretreated on a gel permeation chromatography column (GPC; J2 Scientific, AccuPrep MPS, 3 cm id, 20 cm length) to remove lipids using C-hex/Etac (1:1 v/v) as the elution solvent. The resulting extracts containing TPHP were passed through an Etac (5 mL) prewashed SPE cartridge (CNWBOND NH2) at a flow rate of 0.25 mL/min and then concentrated to a final volume of 100 μL under a gentle nitrogen stream for further analysis. For the examination of d10-DPHP and other metabolites, the sampled tissues from the remaining 10 fish were spiked with 20 ng of nondeuterated DPHP, acidified with 0.5 mL of 2 M HCl and homogenized in 3 mL of C-hex/Etac (1:1 v/v). After bath sonication and centrifugation using the same procedure described above, the pH value of the combined supernatant was adjusted to 7.2 ± 0.2 with 0.2 M KOH dissolved in 50% ethanol. The mixtures were bath sonicated for 10 min and centrifuged for 5 min at 3000 rpm. The upper layer of organic solvent was concentrated to approximately 4 mL by rotary evaporation and then to a final volume of 100 μL (dissolved in MeOH−H2O, 30:70 v/v) for quantitative analysis of hydrophobic metabolites in tissues. For polar metabolites, the remaining aqueous solution (containing KOH) was again subjected to acidification, bath sonication, and centrifugation using the same procedure described above. The supernatant
■
MATERIALS AND METHODS Chemicals and Materials. TPHP, CAS no. 115-86-6 (purity ≥99%); diphenyl phosphate, CAS no. 838-85-7 (DPHP, ≥ 99%); and d15-TPHP, CAS no. 1173020-30-8 (≥98%) were purchased from Sigma-Aldrich (St. Louis, MO); and deuterated diphenyl phosphate, CAS no. 1477494-97-5 (d10-DPHP, ≥ 95%) was purchased from J&K Scientific Ltd. (Beijing, China). Methanol (MeOH, HPLC-grade) was purchased from Merck Chemicals (Germany). Other HPLC-grade solvents, including cyclohexane (C-hex), hexane, dichloromethane, dimethyl sulfoxide (DMSO), ethyl acetate (Etac), and acetone, were purchased from TEDIA. The SPE cartridges (CNWBOND NH2, 500 mg/3 mL; CNWBOND HC-C18, 500 mg/6 mL; poly−Sery HLB, 200 mg/6 mL; Poly−Sery PSD, 250 mg/6 mL; and Poly− Sery PWAX, 500 mg/6 mL) were purchased from ANPEL (Shanghai, China). Uptake and Depuration Experiments. Adult zebrafish (Danio rerio) (length 37 ± 4 mm; weight 380 ± 40 mg (n = 300)) obtained from the Institute of Hydrobiology, Chinese Academy of Sciences, were acclimated to the laboratory conditions as described by Du et al.5 before experiments. To further confirm the metabolites, d15-TPHP was used for exposure experiments, while nondeuterated TPHP and DPHP served as internal standards. A semistatic aqueous exposure fish test was performed according to a modified standard test protocol (OECD, 305).42 Prior to exposure experiments, a stock solution was prepared by dissolving 10 mg of d15-TPHP in 10 mL of DMSO, followed by dilution with clean water to yield exposure concentrations of 20 (±0.4, n = 3) and 100 (±2.3, n = 3) μg/L, which were chosen on the basis of the reported environmental concentration and acute 96 h LC50 for zebrafish.5 Zebrafish were transferred to glass tanks containing 30 L of d15-TPHP solution and were cultured at 24 ± 1 °C at a fish-to-water loading of 1.9 ± 0.2 g fish (wet weight) per 13556
DOI: 10.1021/acs.est.6b04697 Environ. Sci. Technol. 2016, 50, 13555−13564
Article
Environmental Science & Technology (organic layer) was collected and concentrated to a final volume of 100 μL (dissolved in MeOH−H2O, 30:70 v/v) for further analysis. Extraction of d15-TPHP and Metabolites from Water. To measure the concentrations of d15-TPHP in the test media, 100 mL aliquots of water samples were diluted to 1000 mL with fresh water and spiked with 2 and 10 μg of TPHP for the 20 and 100 μg/L groups, respectively. The spiked test medium was passed through a MeOH (5 mL) prewashed Poly−Sery PSD SPE cartridge at a flow rate of 2 mL/min and was then eluted with 7 mL of Etac. After residual water was removed by passing the sample through anhydrous sodium sulfate, the eluate was concentrated to volumes of 10 and 20 mL for the 20 and 100 μg/L groups, respectively, by rotary evaporation. The concentration of d15-TPHP in water was determined using GC−MS. The metabolites in the test medium were extracted with a Poly− Sery PWAX SPE cartridge using the method described by Van den Eede et al.43 and were finally analyzed by LC−MS/MS or LC−Q-TOF. The detailed procedures used for water sample preparation are provided in the Supporting Information and are outlined in Figure S1C,D. Quantification of d15-TPHP and Metabolites. Quantification of d15-TPHP was performed using a gas chromatograph− mass spectrometer (GC−MS) in the selected-ion-monitoring (SIM) mode. Several selected samples were analyzed in full-scan (FS) mode for d15-TPHP confirmation. The retention time and mass spectrum of d15-TPHP and TPHP were confirmed using the commercial standard (FigureS1E,F). The detailed procedure used for d15-TPHP quantification is provided in the Supporting Information. Qualification of the metabolites of d15-TPHP was performed using LC−Q-TOF. The parameters used for the MS2 spectrum were optimized according to the volume of the molecule and ion intensity of the daughter and its parent. Subsequently, the optimized parameters were also used for quantitative analysis of the metabolites in the tissues by LC− MS/MS. The detailed values of the optimized parameters used for the GC−MS or LC−MS/MS analysis, and the quantitative ions of all of the tested compounds are listed in Table S1. The detailed procedures used for the qualitative and quantitative analysis of the metabolites are provided in the Supporting Information. Data Analyses. The experimental uptake and depuration rates were calculated using first-order models,44−46 and the detailed procedure and results are shown in the Supporting Information and Table S5. The toxic ratio,33 defined as the ratio of a chemical’s LC50 estimated from a QSAR for baseline toxicity and measured from the experiment, was also calculated and is described in the Supporting Information. The PBTK model,47−50 including five tissue compartments, the gill, muscle (carcass), brain, liver, and intestine, was used to model the accumulation of d15-TPHP in zebrafish. Chemical accumulations in all tissues were considered to be blood-flowlimited, and the partition coefficients of d15-TPHP between tissues and blood in the PBTK model were estimated using the analyzed results. Metabolism was set in the liver compartment. Because d15-TPHP was well-metabolized in zebrafish, the metabolic effect (or transformation effect) of d15-TPHP on bioaccumulation was assessed in the liver. As venous blood flowing out of the intestine joins the portal vein and enters the liver,48,49 the main metabolites from the transformation of d15TPHP in liver were excreted into the exposure water through the gills, intestines, and other tissues, such as the kidneys, skin, and so on (Figure S2). The concentrations of the metabolites in water
were used to calculate the liver transformation rate (RAM) of d15TPHP in this paper rather than the in vitro metabolizing method.48,51,52 The total cumulative concentration of d15-TPHP in the liver through passive diffusion without metabolic elimination was calculated as CLTot, and the metabolism effect from the d15-TPHP to d10-DPHP conversion was described by the transformation proportion (T, %) of the transformation concentration to the total cumulative concentration in the liver. The detailed processes used to calculate these parameters are provided in the Supporting Information. Quality Assurance, Quality Control, and Statistical Analysis. All of the containers and droppers used in the experiments were composed of glass to avoid sample contamination. The method quantification limits (MQLs) were defined as 10 times the ratio of the signal to instrument noise (10 S/N). Concentrations that were less than the MQLs were reported as not detected (ND). The MQLs for d15-TPHP and d10-DPHP were 6.3 and 2.2 ng/g, respectively, determined based on 0.1 g of wet tissue sample and 0.20 and 0.14 ng/L, respectively, determined based on a 1 L water sample. To ensure the accuracy of the analytical procedures, nondeuterated TPHP and DPHP were added as surrogate standards throughout the entire analytic procedure. Sample blanks and spiked quality control with known concentrations were routinely analyzed in multiple-steps procedure to make sure the signal strengths and retention times of the analytes and internal standards consistency. The recoveries of TPHP and DPHP were assessed by spiking a standard solution at three levels in fish tissue and two levels in water samples. All of them were in the range of 85.5% to 96.2% (Table S2). Precision was evaluated as the relative standard deviation (RSD) of replicate measurements (n = 5) and indicated good overall reproducibility. The concentrations of d15-TPHP and d10-DPHP in tissue and water samples were calculated according to their corresponding calibration curves with the surrogate standard. Standard curves exhibited good and reproducible linearity, with R2 values of 0.995 and 0.999 for d10-DPHP and d15-TPHP, respectively. d15-TPHP and d10-DPHP and their internal standard concentrations in the fish food and blanks were below the MQLs. All data are expressed as the means and standard deviations. One-way ANOVA with Tukey’s multiple comparison tests was used to determine the statistical significance of differences in the levels of d15-TPHP and metabolites among the treatment groups. Significant difference was set at p < 0.05.
■
RESULTS Tissue-Specific Accumulation and Depuration of d15TPHP. The concentrations of d15-TPHP in tissue and water samples were monitored (Figure 1). The concentrations of d15TPHP in tissues and water in the control groups were all below the MQLs. While d15-TPHP was detected in the intestines, gills, brain, and muscle of zebrafish, the maximum accumulation appeared at approximately 11 days for the 100 μg/L groups and at 14 days for the 20 μg/L groups (Figure 1b−e). After 11 or 14 days, the d15-TPHP levels in tissues slightly decreased, and an equilibrium was assumed from 14−19 days in the 100 μg/L groups because the levels of d15-TPHP in each of the selected tissues were not significantly different (ANOVA, p > 0.05). Interestingly, the d15-TPHP level in the liver reached a maximum at 3 days or before and then decreased by 54.8% at equilibrium in the 100 μg/L groups (Figure 1a). Although three replicates with 20 fish each were used, individual differences, such as the fish gender, volume of the specific organs, and moisture content, still 13557
DOI: 10.1021/acs.est.6b04697 Environ. Sci. Technol. 2016, 50, 13555−13564
Article
Environmental Science & Technology
Table 1. Bioconcentration Factors, Transformation Parameters, and Partition Coefficients For the d15-TPHP in Zebrafish parameter BCFww
a
BCFlwb,c
RAM (pg/min/fish) partition coefficient (L/kg)
CLTot (μg/kg) T (%)
liver brain intestine gill muscle liver brain intestine gill muscle liver liver−blood (PLB) intestine−blood (PIB) muscle−blood (PMB) brain/blood (PBB) liver liver
20 μg/L group
100 μg/L group
128.6 ± 10.2 149.9 ± 4.3 157.4 ± 21.6 139.2 ± 32.8 32.1 ± 6.9 1545 ± 122 2468 ± 71 3674 ± 504 3579 ± 843 1968 ± 423 200.4 ± 9.3 0.46 ± 0.04
55.5 ± 5.3 62.4 ± 4.6 75.2 ± 5.2 66.8 ± 4.2 12.8 ± 1.6 653 ± 62 957 ± 70 1748 ± 121 1895 ± 119 783 ± 98 1464.9 ± 84.7 0.22 ± 0.02
d
0.43 ± 0.03
0.21 ± 0.01
0.09 ± 0.01
0.04 ± 0.01
0.42 ± 0.01
0.17 ± 0.01
1136.5 ± 59.5 16.7 ± 1.8
2931.9 ± 190.2 21.6 ± 2.1
a BCFww, bioconcentration factor on a wet weight basis. bBCFlw, bioconcentration factor on a lipid weight basis. cThe lipid contents in tissues were measured using a gravimetric method.66 dMean and standard deviation values were calculated from three replicate samples in equilibrium.
Figure 1. Tissue distribution of d15-TPHP in zebrafish. The concentrations in the tissues in (a)−(e) and the actual concentrations of the aqueous phase (f) in the 20 (±0.4; n = 3) and 100 (±2.3; n = 3) μg/L exposure groups. The dashed line represents the day before depuration. The data points are the mean and standard deviation values that were calculated from triplicate samples. Asterisks (*) indicate a significant difference in the level on the accumulation days (3−11 days) compared to the days in equilibrium (14−19 days) in the respective tissues (ANOVA, p < 0.05).
retention time of 16.12 (Figure 2, M1) and was confirmed by mass spectral (MS2) analysis (Figure 3, M1). The presence of d10-DPHP was also verified by another exposure experiment with nondeuterated TPHP, which showed a similar fragment pattern with a mass-to-charge ratio shift corresponding to deuterium (Figure S3−2, M′1). By the subsequent monitoring of the corresponding fragments and molecular ions, we identified five other metabolites in the liver with chromatographic retention times of approximately 12.7 and 14.9, 19.3, 17.9, 16.9, and 16.4 and 17.4 min (Figure 2). This elution order was consistent with the nondeuterated TPHP experiment (Figure S3−1). The five compounds were identified as monohydroxylated diphenyl phosphate (M2), mono- (M4) and dihydroxylated (M5) d15TPHP, and their glucuronic acid conjugated metabolites after hydroxylation (M6 and M7) on the basis of their mass-to-charge ratios, isotope pattern, and deuterated fragment ions in the mass spectrum (Figure 3), which were further confirmed based on the similar fragment pattern of the metabolites in the nondeuterated TPHP exposure experiment (Figure S3−2, M′2, M4′−M7′). In addition, a compound with mass and isotope patterns consistent with a monophenyl phosphate (MPHP) (m/z = 178.0) was identified at a retention time of 3.7 min in the exposure medium (Figure 3, M3). The masses, chemical structures and derived properties of the seven metabolites in the liver in LC−Q-TOF analysis (retention time (RT) and reaction) are summarized in Table S6. It should be noted that the contributions were based on the relative peak areas and may greatly differ from the actual contributions based on concentrations that were measured quantitatively. In addition, mass spectrometry analysis did not provide exact positions of additional substituents on the aryl ring. Consequently, the structures depicted for monohydroxylated, dihydroxylated, and glutathione-conjugated d15-TPHP metabo-
made the data fluctuate, with the relative standard deviation ranging from 2% to 13%. The body burdens of d15-TPHP in selected tissues showed similar trends for both exposure groups (Figure 1). The bioconcentration factors (BCFs) were calculated based on the ratio of the concentration in selected tissues and water in equilibrium (14−19 days) and are shown in Table 1 on the basis of the wet weight (BCFww) and lipid weight (BCFlw). After 19 days, the exposed fish was transferred to fresh water, and most of the d15-TPHP was depurated in the first 3 days (19− 22 days) (Figures 1a−e). The half-lives of d15-TPHP in tissues, determined based on this depuration rate, were all below 20.5 h (Table S5). LC−Q-TOF Detection of Metabolites Derived from d15TPHP. The decrease of d15-TPHP in the tissues after equilibrium prompted us to analyze these tissues for the presence of metabolites. To maximize the recovery of a variety of potential metabolites, tissue homogenates were acidified before being subjected to C-hex−Etac (1:1) extraction. The identification of d15-TPHP metabolites by matching accurate masses and isotope patterns was achieved through analysis by LC−Q-TOF. The liver was chosen as a priority tissue to search for metabolites because of the observed quick decrease in the accumulated d15-TPHP after 3 days of exposure. Deuterated d10-DPHP was detected at a 13558
DOI: 10.1021/acs.est.6b04697 Environ. Sci. Technol. 2016, 50, 13555−13564
Article
Environmental Science & Technology
the others (M2 and M4−M7) were presented with relative peak areas due to the lack of available standards (Figure S5). Most of the six metabolites were detected in the liver and intestine, while their concentrations in the brain and muscle were below the MQLs. Although the amount of M1 and M2 detected in the gills was nearly 1/3 of that in the intestine, the other four metabolites were not found in the gills. The levels of metabolites M2 and M4−M7 (16 days) in the liver and intestine were compared on the basis of their chromatographic peak areas of the MS2 ions (Figure 5d,e). The concentration of hydrolysis metabolites M1 and M2 were higher in the intestine than in the liver, while the levels of oxidation metabolites and glucuronic acid-conjugated metabolites after hydroxylation, M4−M7, showed a reversed trend. The levels of metabolites in the tissues of both exposure groups were rapidly decreased after 3 days of depuration, which was consistent with the findings for the parent d15-TPHP. Rapid excretion of metabolite d10-DPHP by the fish resulted in a large amount of d10-DPHP in the water, up to 1/4−1/3 of the level of d15-TPHP. For the not put back sampling, the number of fish in the tanks decreased after each sample point. Although the level of d10-DPHP in the water decreased with the number of fish (Figure 5a, in water), the transformation rate of each fish (RAM) in equilibrium (14−19 days) was stabilized at 200.4 ± 9.3 pg/min/ fish and 1464.9 ± 84.7 pg/min/fish, respectively, for the 20 and 100 μg/L groups (Table 1), indicating that each exposed fish functioned as a tiny biochemical reactor to degrade d15-TPHP at a constant rate. Partition Coefficients and Transformation Proportion of d15-TPHP in Zebrafish. The transformation proportion (T) in the liver and partition coefficients between tissues and blood were calculated in the PBTK model (Table 1). The partition coefficients of liver−blood, intestine−blood, and brain−blood samples ranged from 0.17 ± 0.01 to 0.46 ± 0.04, which were more than those of muscle−blood samples, which ranged from 0.04 ± 0.01 to 0.09 ± 0.01. The transformation proportion (T) represented the extent of the transformation effect on bioaccumulation. A low proportion indicated that biotransformation consumed little of the chemical, which should have been accumulated in the tissues if no metabolism occurred, and implied that biotransformation had little impact on bioaccumulation. A high proportion had the opposite meaning. In the liver of zebrafish, the transformation to d10-DPHP reduced the total cumulative d15-TPHP in the 20 and 100 μg/L groups by 16.7 ± 1.8% and 21.6 ± 2.1%, respectively.
Figure 2. Identification of the metabolites of d15-TPHP in the liver by LC−Q-TOF. Ion chromatograms with Gaussian smoothing are shown and indicate the approximate retention times of all seven identified metabolites (M2, t = 12.75 + 14.93 min; M7, t = 16.44 + 17.45 min). The structures of the monohydroxylated, dihydroxylated, monoglucuronidated metabolites after hydroxylation and glucuronidated metabolites after dihydroxylation are representative examples because mass spectrum does not give information in the exact positions of the hydroxyl and glucuronic acid substituents.
■
DISCUSSION To determine the impact of OPFRs on aquatic life, we identified the acute toxicity of nine OPFRs and found that TPHP actively induced cardiotoxicity during zebrafish embryogenesis and disturbed carbohydrate metabolism, lipid metabolism, and the DNA damage repair system in the adult zebrafish liver.5,23 However, in these studies, the details of the uptake and distribution of these OPFRs and their additional metabolites in tissues were not characterized. In the present study, the BCFww and BCFlw of d15-TPHP in different tissues ranged from 32.1 to 157.4 and 1545 to 3674, respectively, in the 20 μg/L exposure group, which were not significantly (p > 0.05) related to the lipid contents in the tissues (Figure S6c,d). It seems that the uptake of TPHP may be not a simple partition process between the water and lipids of fish.53 However, when the measured concentration in liver CL (Figure S6c,d) was substituted for the total cumulative concentration in
lites should be considered to be examples of potential regioisomers containing the indicated substituents on the triphenyl ring system. The structures and proposed pathways for the formation of these main metabolites in zebrafish are shown in Figure 4. Moreover, we examined all samples for evidence of 10 additional potential metabolites, including sulfate conjugates after hydroxylation, methoxylation after hydroxylation, and hydrolysates after glucuronidation (Figure S4). However, none of these additional metabolites (M8−M17) were detectable or were below the detection limit in our samples. Distribution and Depuration of Metabolites in Tissues and Water. In addition to the parent d15-TPHP, we tried to quantify the distributions of the six main metabolites detected in the liver in other tissues during the equilibrium and depuration periods. Metabolite M1 (d10-DPHP) was quantified as the actual concentration with the commercial standard (Figure 5b,c), and 13559
DOI: 10.1021/acs.est.6b04697 Environ. Sci. Technol. 2016, 50, 13555−13564
Article
Environmental Science & Technology
Figure 3. Results of a LC−Q-TOF analysis of the d15-TPHP metabolites. The present structures are representative examples of the metabolites because mass spectrum does not give information in the exact positions of the hydroxyl and glucuronic acid substituents.
liver (CLTot) (Figure S6e,f), the accumulated concentrations increased with the lipid contents in the muscle, brain, and liver, indicating that passive diffusion was important for d15-TPHP accumulation. The gills and intestines accumulated more d15-
TPHP than expected on the basis of their lipid contents, which was possibly because the gills could accumulate d15-TPHP through direct contact with suspended particles in the water via inhalation, as exemplified by other pollutants, such as PBDE and 13560
DOI: 10.1021/acs.est.6b04697 Environ. Sci. Technol. 2016, 50, 13555−13564
Article
Environmental Science & Technology
Figure 4. Proposed pathway for the in vivo biotransformation of d15-TPHP in zebrafish based on the metabolites identified by LC−Q-TOF. The dashed arrows represent the way that d10-DPHP is likely oxidized to d10-DPHP-OH. The present structures are representative examples of the metabolites because mass spectrum does not give information in the exact positions of the hydroxyl and glucuronic acid substituents.
PCB,50,54 and the fact that the intestine could accumulate d15TPHP by partitioning it to the intestinal contents.55 The actual concentration of d15-TPHP in the exposure medium was 6.7−8.2 μg/L in the 20 μg/L group, which was similar to the reported environmental level.7,13,14 The accumulation of TPHP in the selected tissues explains why TPHP can be frequently detected in wildlife. This is the first study to provide details on in vivo phase I hydrolysis and oxidation as well as phase II metabolism of TPHP in zebrafish. The application of isotope-labeled d15-TPHP made us further confirmed the existence of these metabolites. The structures of the identified metabolites indicated that oxidation of d15-TPHP to d15-TPHP-OH (M4 and M5) (Figure 4) could occur in vivo, which was presumably catalyzed by Cytochrome P450 enzymes.40 The d15-TPHP-OH thus formed would be subject to further biotransformation catalyzed by SULTs and UGTs. Although d10-DPHP was the major metabolite (Table S6), the effects of the minor oxidation metabolites (like dihydroxylated d15-TPHP (M5)) cannot be ignored. The presence of dihydroxylated species implied the formation of catechols and quinone intermediates, which could be more toxic than their parents.56 For the study of its fate and potential toxicities, an understanding of the distribution and concentrations of a chemical and its metabolites throughout the body could not only help researchers understand its uptake, metabolism, and excretion but also may provide insight into the pharmacokinetics of its metabolites and their chemical interactions in the body.53 The high level of d15-TPHP and undetectable level of metabolites in the brain suggested that TPHP was able to cross the blood− brain barrier, but the metabolites were not, which was reasonable considering the less hydrophobic and more polar properties of its metabolites.9,57 Obviously, the distribution of the parent d15TPHP and its metabolites in the liver and intestine suggested that the hepatobiliary system (liver−bile−intestine) would be an important compartment for the metabolism and excretion of TPHP in zebrafish. When the concentration of d15-TPHP and its
major metabolite, d10-DPHP, are compared in the liver and intestine in equilibrium, it is surprising to find that d10-DPHP was detected at a maximum concentration of 9.2 ± 1.0 μg/g ww, which was 3.0−3.5 times those of parent d15-TPHP (Table S7). The toxic effect of a chemical in an organism is apparently dependent on its internal concentration and metabolites. The high concentrations of d15-TPHP and metabolites observed in different organs might cause various toxic effects, such as inducing neurotoxicity in the brain,27 reducing the circulating bile acid concentrations in the intestine,58 and disturbing the carbohydrate metabolism and lipid metabolism in the liver.5,23 Chemicals with a toxic ratio of ≥10 are assumed to have a specific mode of toxic effect.33 The observed toxic ratio (0.82) of TPHP in zebrafish indicated that the reported toxicity in other studies may simply represent a nonspecific effect due to its bioaccumulation. However, the high acute toxicity of TPHP to another fish (rainbow trout, LC50 of 0.36 mg/L)32 and other aquatic organisms, such as algae,32 suggest that more attention should be paid when it is released into water ecosystems. The finding that there was a higher concentration of d10DPHP than parent d15-TPHP in the liver and intestine indicated that intensive transformation may occur in the fish body. Nichols and other authors51,52,59 demonstrated that biotransformation could strongly impact the extent to which hydrophobic organic chemicals accumulated in fish. Our results showed that 85.1− 94.4% of d15-TPHP was transformed to d10-DPHP (Table S6), which was confirmed by the detected concentration of d10-DPHP in water. The use of the concentration of the main metabolite in water to evaluate the in vivo transformation rate of fish in equilibrium may be a more-objective method to evaluate the metabolic effects on the bioaccumulation than studies of the in vitro hepatic clearance.48,51,52 However, missing the quantification of other metabolites and of the subsequent metabolites from the first metabolites in water would make the calculated transformation proportions lower than the actual values. The transformation of 16.7−21.6% of the total accumulated d15TPHP to d10-DPHP demonstrated that taking the biotransfor13561
DOI: 10.1021/acs.est.6b04697 Environ. Sci. Technol. 2016, 50, 13555−13564
Article
Environmental Science & Technology
that the BCFww of TPHP in fathead minnows was 420 ± 25 and 218 ± 55 using a static test and nonlinear regression methods and that the uptake rate constant (k1) was 15.4 (h−1) and the depuration half-life was 30 h. For a BCFww of 32.1−157.4, the uptake rate constant was >7.9 (h−1) and the half-life (t1/2) was
