Article pubs.acs.org/est
Degradation, Metabolism, and Bound-Residue Formation and Release of Tetrabromobisphenol A in Soil during Sequential Anoxic− Oxic Incubation Jie Liu,† Yongfeng Wang,† Bingqi Jiang,† Lianhong Wang,† Jianqiu Chen,‡ Hongyan Guo,†,§ and Rong Ji*,†,§ †
State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Xianlin Avenue 163, 210023 Nanjing, China ‡ Department of Environmental Science, China Pharmaceutical University, Tongjia Alley 24, 210009 Nanjing, China § Institute for Climate and Global Change Research, Nanjing University, Hankou Road 22, 210093 Nanjing, China S Supporting Information *
ABSTRACT: Tetrabromobisphenol A (TBBPA) is one of the most commonly used flame retardants and has become an environmental contaminant worldwide. We studied the fate of 14C-labeled TBBPA in soil under static anoxic (195 days) and sequential anoxic (125 days)− oxic (70 days) conditions. During anoxic incubation, TBBPA dissipated with a half-life of 36 days, yielding four debromination metabolites: bisphenol A (BPA) and mono-, di-, and tribrominated BPA. At the end of anoxic incubation, all four brominated BPAs completely disappeared, leaving BPA (54% of initial TBBPA) as the sole detectable organic metabolite. TBBPA dissipation was accompanied by trace mineralization (130 days.2,12−14 Under oxic conditions, the half-life of TBBPA in one Nordic clay soil was 65−93 days13 and 36−82% of TBBPA remained in three loamy soils after incubation for 64 days,15 while in oxic sediments TBBPA had a half-life of 9−40 days.1,16 As a halogenated aromatic compound, TBBPA may be used as an electron acceptor by microorganisms and be reductively debrominated to less brominated bisphenol A (BPA) and finally to BPA under anoxic conditions; 12,17 however, debromination metabolites of TBBPA have been reported only in anoxic sediments.2,18 Under oxic conditions, microorganisms may O-methylate TBBPA forming monomethyl and dimethyl TBBPA ethers,1,19 which have been detected in Received: Revised: Accepted: Published: 8348
April 2, 2013 June 1, 2013 July 8, 2013 July 8, 2013 dx.doi.org/10.1021/es4014322 | Environ. Sci. Technol. 2013, 47, 8348−8354
Environmental Science & Technology
Article
sediments.1,20 Metabolites of TBBPA in both anoxic and oxic soil are still unclear. Sequential anoxic−oxic incubation was proposed as a strategy to completely degrade TBBPA from contaminated soil,12 in which TBBPA is first anaerobically debrominated to BPA, which is then degraded during subsequent oxic incubation. This strategy was based on experiments with separate samples12 and has not yet been applied to real environmental samples. Considering the complexity of the real environment, the validity of the sequential anaerobic−aerobic process in removal TBBPA from soil needs to be proved. The objectives of this study were (1) to obtain information about metabolites and fate of TBBPA in soil under anoxic and sequential anoxic−oxic conditions, and (2) to examine the feasibility of sequential anaerobic−aerobic process in remediating soil contaminated with TBBPA. We used 14C-labeled TBBPA to facilitate localization and identification of metabolites and residues of TBBPA, and chose a rice paddy soil, because in reality such soil undergoes frequent alteration of redox environment between anoxic state at water-submerged period and oxic state at dry period.21
(Oxysolve C-400; Zinsser Analytic, Frankfurt, Germany) and the radioactivity in the cocktail was counted by a liquid scintillation counter (LSC) (see below). Then, these tubes were sacrificed for extraction by organic solvents and for analysis of radioactivity in different soil fractions (see below). Incubation under Oxic Conditions. After the incubation under anoxic conditions for 125 days, the soil−water mixture in part of the tubes was transferred to glass incubation vials (25 mm diameter × 120 mm height). The water layer was decanted after centrifugation (720 g), and the soil (the pellet) was mixed using a stainless spatula so that the soil was exposed to the air. The vials were closed with rubber stoppers and incubated further at 27 °C in the dark. The CO2 released from the soil in the incubation vials during the oxic incubation was trapped by 1.0 mL of NaOH (1 M) contained in a 6-mL plastic vial which was suspended from the bottom of the stopper.24 During the incubation, these incubation vials were opened for about 1 min each day to allow a headspace exchange with fresh air. At defined sampling times (15, 30, and 70 days), the radioactivity in the NaOH trap was determined (see below), and three vials were sacrificed for analysis as in the case of the anoxic incubation. Extraction and Fractionation of Soil Samples. After incubation, the Hungate tubes were gently centrifuged (720g, 10 min). The supernatant was decanted from the soil pellets and determined for radioactivity by LSC (see below). The soil pellet was freeze-dried and extracted with 10 mL of methanol (twice) and 10 mL of ethyl acetate (once) by repeated shaking (200 rpm, 1 h) and centrifugation (2900g, 10 min). The supernatants (extracts) were combined and the radioactivity was measured by LSC (see below). The extracts were concentrated on a rotary evaporator to about 1 mL for further analyses using high-performance liquid chromatography coupled to a radioactivity detector (HPLC-LSC) (see below) and liquid chromatography coupled to a mass detector (LCMS) (see below). The radioactivity in water and organic solvent extracts were considered as extractable residues. Preliminary experiments showed that 92.6 ± 0.3% of TBBPA in soil could be recovered as extractable residues by this procedure. The radioactivity remained in the soil after the exhaustive extraction with organic solvents was defined as bound residues of 14C-TBBPA and its metabolites, which was further fractionated into fulvic acids (FA)-, humic acids (HA)-, and humin-bound residues according to their alkaline solubility.25 Briefly, about 0.5 g of soil sample of the residues was extracted with 0.1 M oxygen-free NaOH (2 mL) for 24 h by horizontal shaking (200 rpm). The precipitate after centrifugation (11 000g, 20 min) was defined as humin which was freeze-dried and determined by means of combustion with a biological oxidizer (see below). The supernatants were separated into fulvic (supernatant) and humic (precipitate) acids by acidification with HCl (6 M) to pH 1. The radioactivity in the humic fractions was determined by LSC (see below). Analytic Methods. HPLC-LSC Analysis. HPLC was performed on an Eclipse XDB-C18 column (250 mm × 4.6 mm, 5 μm; Agilent Technology, USA) at 30 °C with an Agilent HPLC 1100 system, equipped with an autoinjector, a degasser, a diode array detector, and an online radio flow detector (Ramona Star; Raytest, Straubenhardt, Germany) with a cell volume of 600 μL. The radio detector had a detection limit of 1 Bq. About 20 μL of samples in methanol was injected. The mobile phase contained water (25%) and methanol (75%),
■
MATERIALS AND METHODS Chemicals. Uniformly 14C-ring-labeled TBBPA (1.48 GBq/ mmol, 99% radiochemical purity and 97% chemical purity) was synthesized from 14C-ring-labeled phenol via BPA as described by Susan et al.22 Nonlabeled TBBPA and BPA were purchased from Sigma Incorporation (Shanghai, China). Other chemicals with chromatographic or analytical grade were purchased from Nanjing Chemical Reagent, LTD (Nanjing, China). Paddy Rice Soil. A gleyic hydragric Anthrosol soil was collected from the Changshu Experimental Station of Chinese Academy of Sciences in Jiangsu Province, China in September 2011. The paddy soil contained 2.2% total organic carbon, 0.12% nitrogen, and had a pH (0.01 M CaCl2) of 7.17. The soil was air-dried, sieved through 20-mesh (0.90 mm), and stored at room temperature before use. Incubation under Anoxic Conditions. Anoxic incubation experiments were conducted in Hungate tubes (16 mm × 125 mm) with butyl rubber stoppers. Two grams of air-dried soil was weighed into the tubes and submerged by 5 mL oxygenfree medium containing NH4Cl (2.7 g/L), MgCl2·6H2O (0.1 g/L), CaCl2·2H2O (0.1 g/L), FeCl2·4H2O (0.02 g/L), K2HPO4 (0.27 g/L), and KH2PO4 (0.35g/L).23 Resazurin was added to three of the tubes as redox indicator at 20 mg/L. The tubes were then sealed under a headspace of N2/CO2 (80:20; v:v) and were incubated in dark at 27 °C. When the water color in the tubes with resazurin changed from pink to colorless, indicating an anoxic state of the soil−water mixture, 16 μL of methanolic stock solution of 14C-TBBPA was added to the other tubes with a microsyringe within a plastic bag filled with N2/CO2 (80:20; v:v), giving a final concentration of 10 mg/kg soil (dry weight) and a radioactivity concentration of 5.55 kBq/ g soil (dry weight). Controls (sterile treatments) were identically prepared using soil−water mixtures which were autoclaved thrice at 121 °C for 30 min on three consecutive days. In total, 50 tubes were prepared and incubated in dark at 27 °C. At defined sampling times (0, 10, 30, 50, 80, 125, 155, and 195 days), about 2 mL of headspace of a batch of samples (total headspace 10 mL) was withdrawn by syringe with valve to measure the amount of 14CO2 generated during incubation. The gas was injected to 2 mL of alkaline scintillation cocktail 8349
dx.doi.org/10.1021/es4014322 | Environ. Sci. Technol. 2013, 47, 8348−8354
Environmental Science & Technology
Article
running isocratically at 1 mL/min, was mixed with scintillation cocktail (Gold Flow multipurpose; Meridian Biotechnologies Ltd., Epsom, UK) running at 2 mL/min for the radio detector. The UV signal was recorded at 220 nm. LC-MS Analysis. LC analysis was performed on the Finnigan Surveyor LC system, using a Beta Basic-C18 column (150 mm × 2.1 mm column, 5 μm; Thermo Scientific, USA). The mobile phase was a mixture of water (25%) and methanol (75%), running isocratically at a flow rate of 0.2 mL/min. About 10 μL of the methanolic extracts was injected. MS analysis was performed on a Finnigan LCQ Advantage MAX ion trap mass spectrometer (Thermo, USA), which was operated with electrospray ionization (−45 kV) in negative ion mode. The ion-transfer capillary temperature was set to 300 °C. Nitrogen was used as sheath gas and auxiliary gas. Determination of Radioactivity. Radioactivity was quantified by a liquid scintillation counter (LS6500; Beckman Coulter, USA) with a detection limit about 0.5 Bq. The quench on radioactivity was corrected by external standards and the chemiluminescence in the samples was negligible. For organic extracts, 0.5 mL of the extracts was mixed with 2 mL of scintillation cocktail (Gold Star multipurpose; Meridian Biotechnologies Ltd., UK). For radioactivity in aqueous samples, 4.5 mL of water samples and 1 mL of alkaline solution (14CO2 trap) were mixed with 10 and 2 mL of cocktail, respectively. For radioactivity in bound residues, 0.5 g of extracted soil was combusted using a biological oxidizer (OX500; Zinsser Analytic, Germany) and the generated 14CO2 was absorbed by 15 mL of alkaline cocktail Oxysolve C-400 (Zinsser Analytic) and counted on LSC. Data Analysis. TBBPA and BPA dissipation data in soil was fitted to the first-order kinetics C = C0e−kt, using SigmaPlot 12.0, where C0 is the initial concentration, C is the concentration at time t, and k is the degradation rate constant. Half-life time (t1/2) was calculated by the equation t1/2 = ln 2/k.
Figure 1. HPLC with UV (A) and radioactivity (B) detection of organic extracts of soil incubated with 14C-TBBPA under anoxic conditions for 80 days. The radioactive peaks represent TBBPA and its debromination metabolites tri-, di-, and mono-BBPA and BPA identified by LC-MS, while the UV peaks indicate all possible compounds (including background matrixes) with UV absorption property.
■
RESULTS AND DISCUSSION Degradation and Metabolism of TBBPA under Anoxic Conditions. The organic extracts of the soils during anoxic incubation of 14C-TBBPA over 195 days were analyzed by HPLC-LSC, which showed TBBPA and four radioactive metabolite peaks on the chromatographs (Figure 1). The metabolites were identified by LC-MS as debromination products of TBBPA, i.e., tri-, di-, and monobromobisphenol A (tri-BBPA, di-BBPA, and mono-BBPA, respectively) and BPA. The identification was based on comparison of retention times on chromatographs and ion fragments on mass spectra with authentic compounds (see Supporting Information). The occurrence of the less brominated intermediates was evidence for a debromination process of TBBPA under the anoxic conditions. The degradation kinetics of TBBPA and formation kinetics of the metabolites during the incubation are illustrated in Figure 2. TBBPA in active soil degraded apparently without a lag period and depleted completely after the anoxic incubation for 195 days (Figure 2A). The degradation rates of TBBPA in the active soil followed the first-order decay kinetics with a kinetic constant of 0.020 ± 0.001 d−1 (t1/2 = 36 ± 3 days). The debromination end product BPA was detected after 30 days and increased to about 54% of the initially applied TBBPA at the end of the anoxic incubation (195 days) (Figure 2A). The accumulation of BPA under the anoxic conditions demonstrated that BPA was not easily degraded under anoxic conditions, which is in accordance with previous studies.2,26
Figure 2. Degradation of TBBPA and formation of metabolites in active and sterilized soil during incubation of TBBPA under static anoxic conditions (195 days for active soil; 155 days for sterilized soil) or sequential anoxic−oxic conditions (0−125 days: anoxic, open symbols; 126−195 days: oxic, closed symbols). The vertical arrows at day 125 indicate the switch of incubation from anoxic to oxic conditions. A: Degradation of TBBPA and formation of BPA in active and sterilized soil; B: formation of intermediates (mono-, di-, and triBBPA) in the active soil. No metabolite was detected in the sterilized soil. The present values are means with standard deviations of three individual experiments.
8350
dx.doi.org/10.1021/es4014322 | Environ. Sci. Technol. 2013, 47, 8348−8354
Environmental Science & Technology
Article
The less brominated BPAs were temporarily detected during the anoxic incubation with a maximum value ranging from 4.4− 6.6% of the applied TBBPA (Figure 2B) and were completely disappeared after 125 days of anoxic incubation, leading to BPA as the sole detectable debromination metabolite of TBBPA in the organic extracts. In the sterilized soil, dissipation of TBBPA was observed at the beginning of incubation, and no significant dissipation of TBBPA was observed during the further anoxic incubation (Figure 2A). This abiotic dissipation of TBBPA was attributed to physical−chemical aging processes that sequester micropollutants in soils.27 During the whole incubation, no metabolite was detected in the sterilized soil (data not shown), indicating that the observed debromination of TBBPA in the active soil was attributed to microbial activities. Degradation of Metabolites under Oxic Conditions. After anoxic incubation of 125 days, part of the samples were exposed to air and subsequently incubated for 70 days under oxic conditions. The alteration of incubation condition strongly affected the degradation of the end metabolite BPA (Figure 2A). Under oxic conditions, the anaerobically accumulated and stable BPA degraded rapidly (Figure 2A), indicating the existence of active BPA-degraders in the soil. This phenomenon was also consistent with the ubiquitous occurrence of BPAdegrading bacteria (including pseudomonads, streptomycetes, and sphingomonads) in oxic environments.28 BPA was readily biodegraded in oxic environments and the microbial population appears to rapidly acclimate to degrade BPA.29 The present study demonstrated that BPA degradation in oxic soil took place rapidly even after long-term incubation under anoxic conditions. Our results confirm the ubiquitous occurrence of bacteria capable of degrading BPA and suggest that BPAdegraders may survive under different environmental conditions. One previous study demonstrated that BPA had a half-life