Environ. Sci. Technol. 2005, 39, 8762-8769
Production Mechanism of Hydroxylated PCBs by Oxidative Degradation of Selected PCBs Using TiO2 in Water and Estrogenic Activity of Their Intermediates KEI NOMIYAMA,† TEIJI TANIZAKI,‡ HIROSHI ISHIBASHI,§ KOJI ARIZONO,† AND R Y O T A S H I N O H A R A * ,† Graduate School of Environmental and Symbiotic Sciences, Prefectural University of Kumamoto, 3-1-100, Tsukide, Kumamoto 862-8502, Japan, Kitakyushu City Institute of Environmental Sciences, Shin-ike 1-2-1, Tobata, Kitakyushu 804-0082, Japan, and Center for Marine Environmental Studies (CMES), Ehime University, Bunkyo-cho 2-5, Matsuyama 790-8577, Japan
The oxidative photodegradation behaviors of selected polychlorinated biphenyls (PCBs) [2,3,3′,4′-tetraCB (BZ number: CB56), 2,3′,4′,5-tetraCB (CB70), and 3,4,4′,5-tetraCB (CB81: coplanar PCB)] using titanium dioxide (TiO2) in water were investigated. The main purposes were to clarify the structural relation between the original PCBs and the intermediates derived by TiO2 oxidation and to evaluate the estrogenic activity in the treated PCBs during the oxidative reactions. Approximately 90% of the three tetraCBs decomposed within 120 min. Intermediates by decomposition of three tetraCBs, such as some OH-tetraCBs and OHtriCBs, carboxylic intermediates, phenolic intermediates, and other intermediates produced by the cleavage of a benzene ring were identified and quantified. In the degradation pathways, the produced amounts of OHtetraCB and OH-triCB increased within 60 min of irradiation time. Estrogenic activities of the intermediates from the three tetraCBs in water were assessed by using a yeast twohybrid assay system for human estrogen receptor R (hERR). The maximal estrogenic activities were induced by the solutions of decomposed CB81 with irradiation time at 60 min. We found that the solutions at an irradiation time of 60-120 min contained several 4-OH-tetraCBs and 4-OH-triCBs substituted with OH and Cl at para- and para′positions. It is presumed that the chemical structures of the 4-OH-PCBs are similar to that of 17β-estradiol (β-E2); these intermediates present strong estrogenic activities. Moreover, we learned that there is a high possibility of conversion from some low toxic PCBs congeners to strong estrogenic OH-PCBs.
Introduction Polychlorinated biphenyl (PCB) congeners are industrial chemicals that have been used in electrical capacitors, transformers, lubricants, cooling fluids, flame retardants, * Corresponding author phone: 81-96-383-2929; fax: 81-96-3846765; e-mail:
[email protected]. † Prefectural University of Kumamoto. ‡ Kitakyushu City Institute of Environmental Sciences. § Ehime University. 8762
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 22, 2005
hydraulic fluids, adhesives, and plastics. PCBs and their biological metabolites as environmental contaminants have been shown to affect the endocrine system in human beings and wildlife (1-3). PCB congeners, which are present in commercial mixtures, are lipophilic substances with low water solubility, demonstrating a strong tendency to accumulate in higher organisms. The accumulate PCBs are known to have numerous adverse effects on human health including reproductive enbryotoxicity, oncogenicity, estrogenic endocrine disruption, and even human carcinogenicity (4-5). PCB congeners have been widely detected in the environment; the concentration of PCB congeners is reported at 12-390 pg/m3 in the North Pacific and 72-600 pg/m3 in the North Atlantic (6). However, there is no report concerning degradation products of PCBs and ecotoxicological effects by the products in environment. Within organisms, PCBs are metabolized (7) and oxidized (8) into hydroxylated polychlorinated biphenyls (OH-PCBs). These metabolites are formed by the cytochrome P450 enzymatic system in the human body that generally involves the oxide intermediates (9). OH-PCBs detected in human organs, blood (10-11), fatty tissue (12-13), and milk (14) as well as in wildlife have estrogenic and antiestrogenic effects (7). OH-PCBs may disrupt thyroxin and vitamin A transport and might lead to adverse neurodevelopmental effects (7). OH-PCBs are attracting increasing attention as potentially estrogenicity metabolites of PCBs. The research regarding estrogenic activity among toxicological researchers of OHPCBs has been frequently reported in the past decade (1518). However, there is no information as to what kinds of PCBs produce OH-PCBs showing strong estrogenic activity. PCBs and the OH-PCBs generated from PCBs have the potential to cause a worldwide environmental pollution problem in the future. Therefore, understanding conversion by oxidation of PCBs into OH-PCBs can add to our fundamental knowledge for estimating the fate of PCBs. In this study, we attempted to clarify the production process of OH-PCBs and the relation between OH-PCBs and their original PCBs. We applied titanium dioxide (TiO2) photocatalyst to rapid oxidation of PCBs in water to elucidate degradation pathways in environmental water. TiO2 photocatalysts utilize activated microholes, which are produced on the surface due to irradiation by UV (shorter than 388 nm in wavelength) to decompose organic substances in water and air (19). The degradation pathway and intermediates of some chemicals by TiO2 photocatalysts were reported (2022); however, a detailed degradation pathway has not been confirmed. Although a LC/ESI (electrospray ionization)-MS system and UV were applied to analyze the intermediates, insufficient spectral information of the intermediates was obtained (20-22). We carried out the derivatization of the high polar intermediates which were converted to trimethylsilyl (TMS) derivatives for gas chromatograph/mass spectrometer (GC/ MS) analysis. The identification of the intermediates from the three tetraCBs (CB56, CB70, and CB81) was attempted by mass spectral interpretation of derivatives and by comparison of GC chromatographical retention times between intermediates and reference compounds. The degradation pathway was estimated by identical combinations between identified intermediates and their quantitative transition at a degradation time of three tetraCBs. Moreover, we investigated the estrogen agonist activities of OH-PCBs by degradation of PCBs. The estrogen agonist activities of three tetraCB intermediates by TiO2 were evaluated for cases of both -S9 and +S9 by using a 10.1021/es050791a CCC: $30.25
2005 American Chemical Society Published on Web 10/08/2005
constructed yeast two-hybrid assay system for hERR (2324). These intermediates will be expected with a range of features that might provide structural information on affinity for the hERR, and give some insight into the toxicological implications of PCB oxidation. On the basis of these results, we attempt to clarify the structural relation between the original PCBs and the intermediates derived by TiO2 oxidation. Moreover, we examine whether the difference of estrogenic activity among the produced intermediates appeared to be due to structural difference.
Experimental Section Materials. Three tetrachlorobiphenyls (99.7% purity), 2,3,3′,4′tetraCB (CB56), 2,3′,4′,5-tetraCB (CB70), and 3,4,4′,5-tetraCB (CB81: coplanar PCB), were purchased from GL Science Ind. Co., Tokyo, Japan. The stock solutions (50 mg/L) were prepared in acetone and stored at -20 °C. The novel monohydroxylated PCB 5 standards (4-OH-2′,3,3′-triCB, 2-OH-3,3′,4′-triCB, 2-OH-3′,4′,5-triCB, 4-OH-2′,3,5′-triCB, and 3-OH-2′,4,5′-triCB) were synthesized by thermal diazocoupling between chlorophenol and chloro-aniline diazonium salt. These OH-PCBs were supplied by Dr. T. Okumura (18, 25) (Environ. Poll. Cont. Center, Osaka, Prefectural Government). The novel monohydroxylated PCB 3 standards (4-OH-2′,4′,6′-triCB, 3-OH-2′,4′,6′-triCB, and 2-OH-2′,4′,6′triCB) were obtained from Accu Standard, Inc. (New Haven, CT). As a reagent for derivatization, N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) was purchased from GL Science Ind. Co., Tokyo, Japan. The positive control of -S9 and +S9 tests, 17β-estradiol (β-E2, Sigma, St. Louis, MO) and t-stilbene (T-S, Tokyo Kasei Kogyo Co., Ltd., Tokyo, Japan), were used, respectively, in the yeast two-hybrid assay. Analytical Procedure. The nanostructured TiO2 thin film was immobilized on quartz beads by an advanced sol-gel method (26). The photoreactor used in the study was already reported (27). Each tetraCB solution, (1 mg/L) 500 mL, was introduced into the photoreactor and recirculated while being irradiated with UV light. Portions of sample solutions (400 mL) were sampled periodically to determine any variation of the intermediates’ concentration according to irradiation time. That sample solution was removed completely with a rotary evaporator. Acetone (1 mL) was added to the dried residue of the sample solutions, and the residue was applied onto a silica gel column for separation of unchanged PCBs, OH-PCBs, and other intermediates from spiked PCBs. The first fraction containing the PCBs was eluted with hexane/benzene (1:1, 10 mL), and the second fraction containing the OH-PCBs and other intermediates was collected subsequently with methanol (10 mL). The 2 mL volume of the methanol solution was taken from each of the second fractions and reduced to 0.5 mL with a KD concentrator, and then the internal standard (phenanthrene-d10, Wako Pure Chemical Ind., Co., Osaka, Japan) was added to the sample solution of each fraction and analyzed by GC/MS. The 8 mL volume of the second fraction was used for the estrogenic activity assay. One-fifth of the volume was divided from the sample solutions of the second fraction. These volumes were dried completely by a rotary evaporator. This fraction contains OH-PCBs and other intermediates. Every dried residue of the sample solutions was derivatized by BSTFA as follows. The BSTFA (0.5 mL) was added to the dried residue of the sample solution, and the solution stood for 1 h to derivatize at a temperature of 70 °C. Each 2 µL volume of TMS derivative solution was injected into a GC/MS instrument in a splitless mode. The analysis of three tetraCBs and the intermediates was performed with the GC/MS which consists of HP6890 GC (Agilent Technologies, CA) and JMS-700 (JEOL, Tokyo, Japan). A selected ion monitoring (SIM) mode and a scanning mode (50-500 msu) were operated in electron impact positive (EI+)
ionization mode at a resolution of R ) 1000. A fused silica capillary column DB-5 (30 m × 0.25 mm i.d., 0.1 µm film thickness) (J&W Scientific, Folsom, CA) was used. The temperature program consisted of the following: injector temperature 260 °C, an initial oven temperature of 70 °C for TMS derivatives held for 1.0 min, and heating to 280 °C at 10 °C /min and then to 280 °C held for 6 min. The chloride ion produced from decomposed PCBs was detected with an ion chromatograph (IC, Dionex DX-120, Sunnyvale, CA). Estrogenicity Test Using the Yeast Two-Hybrid Assay. The estrogen agonist activities of identified OH-PCBs and sample solution from three tetraCBs treated by TiO2 at 0, 60, 120, 180, and 240 min of irradiation time were measured by a yeast two-hybrid assay, both with and without possible metabolic activation by rat liver S9 preparation (Kikkoman Company, Noda, Japan). The yeast two-hybrid estrogenicity assay system using yeast cells (Saccharomyces cervisiae Y190) was produced by introduction of the human estrogen receptor (ERR) and the coactivator TIF2 (pGAAD424-TIF-2) (23). Both treatments with S9 and without S9 were adapted to a chemiluminescent receptor gene (for β-galactosidase) method employing a 96-well culture plate (SUMILON, Sumitomo Bakelite, Japan) (28-29). After the remains of the volume of 4/5 were divided from each second fraction, all the solutions were completely dried with the nitrogen current. DMSO (20 µL) was added to the dried residue of the sample solutions, with S9 or without S9 incubation with rat liver S9 mix (37 °C, 1 h), being incubated (30°C, 4 h) with yeast cells that had been preincubated (30 °C, overnight) in a modified synthetic dropout (SD) medium lacking tryptophan and leucine. A mixed solution for inducing chemiluminesence and for enzymatic digestion (Zymolase, Seikagaku, Corp., Tokyo, Japan) was added followed by a light emission accelerator solution. The chemiluminesence produced by released β-galactosidase was measured with a 96-well plate luminometer (Luminescencer-JNR AB-2100, ATTO Bio-instrument, Tokyo, Japan). Agonist activity was recorded as EC × 10 which was defined as the concentration of the sample solution producing a chemiluminescent signal 10 times that of the blank control. The inverse of the obtained EC × 10 values of β-E2 and T-S was set to 100. Similar procedures were taken for other samples to calculate the β-E2 relative activity. Computational Chemistry. The MOPAC (version 2000) program in this study is provided by CAChe Scientific Inc. (Fujitsu, Corp., Japan). The AM1 (Austin Model 1) level Hamiltonian parameter is used to optimize stable structures (30). This program was used to obtain optimum geometries, frontier electron density, and atom partial charges. An initial position for a possible OH radical attack was estimated from calculations of frontier electron densities and partial electric charges of all carbon atoms in the PCB structures.
Results and Discussion Photodegradation Curves of Three TetraCBs Using TiO2. Photodegradation curves of three tetraCBs at irradiation time by using TiO2 are shown in Figure 1. About 10% of the three tetraCBs in water were decomposed within 240 min of UV irradiation without the TiO2 photocatalyst. With the photocatalyst, approximately 90% of the initial concentrations are decomposed within 180 min of UV irradiation. The first-order decomposition rate (min-1) constants calculated by an equation for three tetraCBs were CB56, 1.7 × 10-3; CB70, 1.0 × 10-3; and CB81, 0.5 × 10-3, respectively. The degradation of CB81 with TiO2 was faster than that of the other two tetraCBs. Since CB81 has three vicinal chlorines which are electron donors to carbon atom of a benzene ring, it is estimated that the chemical structures might be electrically unstable. The concentration variations of discharged chloride ion in the TiO2 photocatalytic degradation of the three tetraCBs VOL. 39, NO. 22, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8763
FIGURE 1. Photodegradation of three tetraCBs (1 mg/L) using TiO2 Symbols represent the means and standard deviations (n ) 3). rapidly increased within 60 min (data not shown). This was presumed that OH radicals attacked Cl-C bonding points of PCBs in the first stage of degradation pathways. There was no difference in dechlorination reaction rates among these three tetraCBs. Mass spectral data of PCBs and their main intermediates obtained by GC/MS of the reference compounds corresponding to those intermediates data are listed in Table 1. The molecular ions and characteristic ions of some hydroxytetrachlorobiphenyls (OH-tetraCBs), hydroxy-trichlorobiphenyls (OH-triCBs), carboxylic intermediates, and phenolic intermediates typically clearly appear in the mass spectra, as observed in reference compounds derivatized by BSTFA. For example, m/z 346 (the molecular ions of TMS-O-triCBs), 331 (M+ - CH3), 293, 279, 93, and 73 (TMS) coincided with those of reference compounds. Moreover, we confirmed TMS-O-triCBs from intermediates by using GC retention time. In addition, the molecular ion and fragment ions of dichlorobenzoic acid (DCBA) produced by cleavage of PCBs follow: m/z 262 (M+), 247 (M+ - CH3), 173 (M+- O - TMS, 145 (M+ - COO - TMS) and 73 (TMS). The values for OHdichlorobenzoic acid (OH-DCBA) are as follows: m/z 352 (M+), 337 (M+ - CH3), 308 (M+ - COO -), 200, and 73 (TMS) were decided.
Relative Molar Distributions of Three TetraCBs and Their Major Intermediates. Figure 2 shows the relative intensity of three tetraCBs and their major intermediates by GC/MS analysis as a factor of irradiation time. Almost all the produced amount of OH-PCB increased within 60 min of irradiation time, and gradually decreased thereafter. The produced amount of these OH-PCBs was less than 10% of the initial tetraCBs of the peak area on mass chromatograms before the beginning of UV irradiation. For example, the maximum concentration of 4-OH-2′,3,3′-triCB and 2-OH-3,3′,4′triCB in degradation pathway of CB56 were 43 and 12 µg/L at 60 min of UV irradiation time, respectively. The 4-OH-2′, 3,5′-triCB and 2-OH-3′,4′,5-triCB in degradation pathway of CB70 were 30 and 14 µg/L at 60 min of UV irradiation time, respectively. As for DCBA, chlorobenzoic acid (CBA), OHBA, and other intermediates, the amount of production increased after 60 min. In other words, fluctuations of these intermediates according to the irradiation time reveal the possible some step-flows in the degradation pathways of tetraCBs. We calculated the electric charge distributions and frontier electron density on carbon atoms for the three tetraCBs with the MOPAC program. These calculation results are shown in Figure 3. High frontier electron density points of carbon atoms were C1, C1′ (0.188-0.241), and para-positions of C4, C4′ (0.183-0.292) of three tetraCBs. Next, the high frontier electron density points were carbon atom ortho-positions of C2 on the CB56 and CB70. These calculation results will serve basic information in the elucidation of the degradation mechanism of three tetraCBs. These sites are those expected to be the most likely sites of attack by neutral OH radicals (21-22). The definite degradation pathways by oxidization of PCBs with TiO2 were determined by OH-PCBs references, these calculation results, and transition of TIC intensity according to UV irradiation time. Proposed Degradation Pathways of CB56 and CB70. The detailed degradation pathways of CB56 and CB70 in water with TiO2 are shown in Figure 4. One OH-tetraCB and two OH-triCBs were produced by the photodegradation of CB56, and one OH-tetraCB and three OH-triCBs were produced by the photodegradation of CB70, respectively. The chemical structures of 4-OH-2′,3,3′-triCB, 2-OH-3,3′,4′-triCB, 2-OH-
TABLE 1. Mass Spectra Data (m/z) of the Intermediates from Decomposed Three TetraCBs with BSTFA Derivatization Detected by GC/MSa PCBs and intermediates
characteristic ions (m/z)
Rt
185 (M+ - H2Cl3)
150 (M+ - H2Cl4)
331 (M+ - CH3) 295 (M+ - CH3 - HCl)
279
93
21.65 380 (M+) 22.07
365 (M+ - CH3) 329 (M+ - CH3 - HCl)
313
93
432 (M+) 398 (M+) 22.45 324 (M+) 474 (M+) 17.7 352 (M+) 12.42 262 (M+) 11.45 9.58 228 (M+) 282 (M+) 5.35 220 (M+) 8.51 262 (M+)
417 (M+ - CH3) 383 (M+ - CH3) 309 (M+ - CH3) 459 (M+ - CH3) 337 (M+ - CH3) 247 (M+ - CH3)
404 (M+ - C2H6) 368 (M+ - C2H6) 295 (M+ - H - CO) 439 M+ - Cl 308 (M+ - COOH) 204
360 309 261 - Cl - CO) + 308 M - C6H4OH 200 173 (M+ - OH - TMS)
73 73 73 73 73 145
213 (M+ - CH3) 267 (M+ - CH3) 205 (M+ - CH3) 247 (M+ - CH3)
169 (M+ - COOCH3) 223 (M+ - COOCH3) 177 (M+ - COOH) 218 (M+ - COOH)
139 193 147 (M+ - TMS) 172
73 73 73 147
(M+)
CB56 CB70 CB81
20.55 292 20.19 21.44
3-OH-2′,4,5′-triCB 2-OH-3′,4′,5-triCB 2-OH-3,3′,4-triCB 4-OH-2′,3′,5′-triCB (4-OH-3,4′,5-triCB)* (4-OH-3′,4′,5′-triCB)* 4-OH-2,3′,3′-triCB unknown OH-triCBs
18.56 346 (M+) 18.6 17.85 18.82 18.9 19.21 19.42
(4-OH-2,3,3′,4′-tetraCB)* (4-OH-2,3′,4′,5-tetraCB)* unknown OH-tetraCBs dihydroxy-triCBs dihydroxy-diCBs OH-diCDF HPPCB OH-dichlorobenzoic acid 2,3-dichlorobenzoic acid (DCBA) 2,5-DCBA 4-CBA OH-benzoic acids (OH-BA) glycolic acid succinic acid
56
(M+
- HCl)
220 (M+ - 2(HCl))
(M+
a Compounds marked with asterisks (*) indicate that the chemical structures of the intermediates in the parenthesis were estimated from progress of degradation pathways.
8764
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 22, 2005
FIGURE 3. Frontier electron densities and partial electric charges on the carbon atoms of three tetraCBs by MOPAC program calculation (parentheses indicate partial electric charges).
FIGURE 2. Time course of the major intermediates appearing in the GC/MS chromatogram during the photodegradation of the three tetraCBs in water (Co ) internal standard: 1 mg/L of phenanthrened10, T ) 25 °C). Each peak area was normalized by that of the initial three tetraCBs peak areas (Co/C: 100). Their detailed characteristics are listed in Table 1. 3′,4′,5-triCB, 4-OH-2′,3,5′-triCB, and 3-OH-2′,4,5′-triCB were determined by comparison with GC retention time and mass spectra of reference compounds. The OH radicals directly attacked the two benzene rings of CB56, which were changed to 4-OH-2,3,3′,4′-tetraCB by the substitutional reaction of the hydroxyl group followed by dehydration primarily occurring in C4 position carbon atoms with a high frontier electron density (0.210). As for 4-OH-2,3,3′,4′-tetraCB, many products from the degradation pathway of CB56 were confirmed and were estimated to be major intermediates. Similarly, the substitutional reaction of the hydroxyl group followed by dechlorination might primarily occur at high frontier electron density points (C4′ and C2) with 0.274 and 0.182 on the CB56; consequently, 4-OH-2′,3,3′-triCB and 2-OH-3,3′,4′-triCB might be produced. Since we did not have reference compounds for determining one OH-CB70, the exact chemical structure is unknown. However, the carbon atom at position C4 of the benzene ring of CB70 has a high frontier electron density (0.183). This site is expected to be the most likely point of OH radical attack. Therefore, the unknown OH-CB70 was estimated to be 4-OH-2,3′,4′,5-tetraCB. The 2-OH-3′,4′,5-triCB was detected as a major intermediate in the degradation
pathway of CB70. According to the production mechanism, the carbon at ortho-position to the biphenyl is an atom at a high electron density point (0.207); this position is expected to be the most likely position of OH radical attack. The chemical structures, including an exact substitutional position of OH and Cl, have not been determined, although many kinds of polyhydroxylated PCBs were tentatively detected by GC/MS. Furthermore, the polyhydroxylated PCBs were changed to CBA, DCBA, OH-DCBA, and other intermediates as fragment products. The 2,3-DCBA was detected as a quantitatively major intermediate in the degradation pathway of CB56. It is expected that 2,3-DCBA is produced by the OH radicals attack to C1′ atoms that bear the high electron density point (0.236) of CB56. The 2-OH-CB70 evolved to the new intermediates by the consecutive attack of OH radicals. The detection of hydroxylated dichlorinated dibenzofuran (OH-diCDF) suggests that two molecules of 2-OH-CB70 probably condensed through dehydrochlorination. The mass spectrum of OH-diCDF has the molecular ion as well as a characteristic fragment peak pattern (25) similar to that of PCDF (Table 1). The mass spectrum of the trimethylsilylated OH-phenoxytetraCB (HPPCB) observed at a retention time of 22.21 min consists of the molecular ion and fragment ions at m/z 474 (M+), 472 (M+ - 2), 476 (M+ + 2), and 478 (M+ + 4) by TMS derivatization. Owing to four chlorine atoms, 459 (M+ - CH3), 439 (M+ - Cl), and 308 (M+ - C6H4OH) are indicative of the formation of HPPCB by an additional reaction of OH-tetraCBs and phenolic compounds following an attack of OH radicals. It is estimated that these intermediates were produced through the opening of an aromatic ring by the OH radical attack. Consequently, succinic acid and glycolic acid which are highly polar aliphatic compounds are produced in the final stage. From the above result, it is estimated that the stable hydroxylation with dechlorination occurs before the cleavage of the C-C bond of biphenyls. In other words, this suggests that more polyhydroxy-phenolic intermediates VOL. 39, NO. 22, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8765
FIGURE 4. Proposed degradation pathways of CB56 and CB70 in a sample solution using TiO2. The chemical structures of the intermediates in the parenthesis were estimated from progress of degradation pathways (FED: frontier electron density). without chlorine are detected later than chlorophenolic intermediates in the degradation pathway of PCB with TiO2. Proposed of Degradation Pathways of CB81. The detailed degradation pathways of CB81 in water with TiO2 are shown in Figure 5. Two OH-triCBs were produced by the photodegradation of the CB81 between 1 and 2 h of irradiation. In this study, as some reference compounds to enable us to determine two OH-triCBs were not available, the exact chemical structures were unknown. However, significantly more OH-triCB was produced than other intermediates. This suggests that the steric congestion of the 3′,4′,5′-arrangement enhances the reactivity of the 4′-position chlorine due to 3′and 5′-chorines which are located in the vicinity of the 4′chlorine (31-32). In short, we concluded that the high frontier electron density point (0.292) of the carbon atom combined with 4′-position chlorine has stronger reactivity than other carbon atoms. From the above result, the unknown OHtriCB was estimated to be 4-OH-3,4′,5-triCB. The reaction that hydroxylation will occur preferentially through nucleophilic substitution accompanied by a two-step reaction (like an SN1 reaction) was assumed. The other unknown OH-triCB was produced by hydroxylation with dechlorination in water. The carbon atom combined with chlorine at position C4 to the benzene ring is the atom that bears the high frontier electron densities (0.198). This site is expected to be the most likely point of neutral OH radical attack. From this result, it was estimated that the other unknown intermediate of OH-triCB might be 4-OH-3′,4′,5′-triCB. By the consecutive attack of OH radicals, it is possible that two OH-triCBs are formed with substitutional reaction to dihydroxy-diCBs. The 4-CBA was detected as a major intermediate in the degradation pathway of CB81. According to the production mechanism, the carbon at position C1′ of the biphenyl is the 8766
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 22, 2005
FIGURE 5. Proposed degradation pathways of CB81 sample solution using TiO2. The chemical structures of the intermediates in the parenthesis were estimated from progress of degradation pathways (FED: frontier electron density). atom bearing the high electron density point (0.241). As for this point, a cleavage of the benzene ring by OH radical attacks is expected.
FIGURE 6. Dose-response curves for sample solutions produced by decomposition of three tetraCBs at various irradiation times using the agonist test (-S9: CB56-a, CB70-b, CB81-c) and (+S9: CB56-d, CB70-e, CB81-f) for the yeast two-hybrid assay for human estrogen receptor r. Symbols represented the means and standard deviations (n ) 3). The y-axis represents the ration of a sample solution to a blank control (baseline) in chemiluminescence (CLN) intensity (T/B) of β-galactosidase.
TABLE 2. Estrogenic Activities of Eight Hydroxy-Chlorinated Biphenyl Standards by Yeast Two-Hybrid Assaysa -S9 test
+S9 test
chemical name
av of EC × 10 (M) n)3
SD (M)
4-OH-2′,3,5′-triCB 4-OH-2′,3,3′-triCB 2-OH-3′,4′,5-triCB 2-OH-3,3′,4′-triCB 3-OH-2′,4,5′-triCB 4-OH-2′,4′,6′-triCBc 3-OH-2′,4′,6′-triCBc 2-OH-2′,4′,6′-triCBc positive control (β-E2) positive control (T-S)
5.6 × 10-7 6.8 × 10-7 NA. NA NA 1.17 × 10-8 4.4 × 10-7 weak 1.02 × 10-10 NA
2.2 × 10-8 3.2 × 10-8
0.018 0.015
3.9 × 10-9 1.8 × 10-8
0.87 0.021
1.8 × 10-11
% activity relative to β-E2
100
av of EC × 10 (M) n)3
SD (M)
% activity relative to β-E2
NA weakb NA NA NA NA NA NA NAd 6.0 × 10-7
3.8 × 10-8
0.019
a
Positive control for estrogen assay in absence of S9 preparation was 17β-estradiol (β-E2). Positive control for estrogen assay in the presence of S9 preparation was trans-stilbene (T-S). NA is defined as not estrogenic activity < 1 mM tested. In all OH-PCB analysis, the data represent the mean and standard deviation (SD) (n ) 3). b Since estrogenic activity is weak, EC × 10 values were not able to be calculated. Percentage activity relative to β-E2 is calculated from (EC × 10 values β-E2/EC × 10 values test compound) × 100. c The estrogenic activities of 4-OH-2′,4′,6′-triCB, 3-OH-2′,4′,6′-triCB, and 2-OH-2′,4′,6′-triCB were measured for the comparison with intermediates produced by photodegradation of the three tetraCBs. d Here, NA indicates not estrogenic activity < 200 nM tested.
The 4-OH-BA was found as a fragment product in the sample solution after a long UV-irradiation time. This intermediate originating from a cleavage of the aromatic ring in dihydroxy-diCBs and 4-OH-3′,4′,5′-triCB of the aromatic ring was also detected in the final solution. These molecular ions of intermediates typically appeared in each mass spectrum similar to mass spectra of reference compounds. Moreover, these phenolic and carboxylic compounds were produced with a cleavage of an aromatic ring by the OH radical attack. Thus, aromatic intermediates were sequentially decomposed to aliphatic small fragments such as succinic acid and glycolic acid. Since succinic acid and glycolic acid were always produced as a relatively stable intermediate, it was assumed that succinic acid and glycolic acid were major intermediates in the last stage of degradation pathways of CB81. Estrogenic Activities of Sample Solutions Obtained by Decomposition of PCBs Using TiO2. To evaluate the potential
estrogenic activities induced by degradation of PCBs in water environment, it is important to investigate the estrogenic activities of these intermediates. A yeast two-hybrid assay system based on the ligand-dependent interaction of two proteins, a human estrogen receptor R (hERR), and a coactivator were used to assess the estrogenic potency. Yeast cells were exposed to the sample solution from the degradation process, and the results for the samples decomposed for 0, 60, 120, 180, and 240 min revealed β-galactosidase activity as shown in parts a-c of Figure 6 (without rat liver S9 treatment) and parts d-f of Figure 6 (with rat liver S9 treatment). The weak estrogenic activity was found for decomposed CB56 and CB70 solutions at 60-120 min (Figure 6a,b); however, this activity vanished in the case of the S9 rat liver treatment (Figure 6d,e). We found that strong estrogenic activity was observed for decomposed CB81 solutions at 60-120 min (Figure 6c). It is presumed that this strong estrogenic activity originates from 4-OH-3,4′,5-triCB VOL. 39, NO. 22, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8767
and 4-OH-3′,4′,5′-triCB included in first stage of degradation pathways at 60-120 min of UV irradiation time. These compounds vanished in the case of the S9 rat liver treatment (Figure 6f). These decomposed PCB solutions irradiated with UV at 180 min lost estrogenic activity due to the advance of the decomposition. It was estimated that estrogenic activities of these decomposed PCB solutions were due to the counterbalance of two or more intermediates produced by the decomposition of PCBs. It is reported that some dichlorophenolic and phenolic carboxylic compounds have estrogenic activity as well as OH-PCBs (24). In this study, we confirmed estrogenic activities of OHPCBs appeared in the degradation experiment, necessary to identify specific OH-PCBs having estrogenic activities. Therefore, we examined estrogenic activities for eight kinds of OHPCBs including five kinds of OH-PCBs which were detected in the degradation experiment. Table 2 shows the estrogenicity of eight OH-PCBs (1 mM: 4-OH-2′,3,5′-triCB, 4-OH-2′,3,3′-triCB, 2-OH-3′,4′,5-triCB, 2-OH-3,3′,4′-triCB, 3-OH-2′,4,5′-triCB, 4-OH-2′,4′,6′-triCB, 3-OH-2′,4′,6′-triCB, and 2-OH-2′,4′,6′-triCB) by the yeast twohybrid assay. It was found that 4-OH-PCBs (4-OH-2′,3,5′triCB and 4-OH-2′,3,3′-triCB) of the intermediates by three tetraCBs’ photodegradation showed estrogenic activities compared with other OH-PCB intermediates in the case of the agonist (-S9) test. The 4-OH-2′,3,3′-triCB showed weak estrogenic activity in the absence of the S9 mix. The relative estrogenic activities EC × 10 (M) of the para-hydroxy-PCBs for hERR were estimated to be 4-OH-2′,3,5′-triCB (5.6 × 10-7), 4-OH-2′,3,3′-triCB (6.8 × 10-7), and 4-OH-2′,3,3′-triCB (+S9: weak estrogenic activity), respectively. These values were 4-OH-2′,3,5′-triCB (0.018), 4-OH-2′,3,5′-triCB (0.015), and 4-OH-2′,3,3′-triCB (+S9: weak) based on the relative activities to the positive control substance (β-E2: 100), respectively. The estrogenic activities of 4-OH-2,3′,4′-triCB, 4-OH2′,3,3′,6-tetraCB, 2-OH-2′,3,3′-triCB, and 3-OH-2, 2′, 3′,4,6pentaCB were 0.11, 0.0017, 0.0037, and 0.0007, respectively, in the case that β-E2 was set to 100 by using hERR (-S9) (17). In case of hERR (+S9), the estrogenic activities of 4-OH2′,3,3′,6-tetraCB and 3-OH-2, 2′, 3′,4,6-pentaCB were 0.014 and 0.0033, respectively. It was surmised that the estrogenicity of 4-OH-PCB as a mimic for β-E2 might be due to the parasubstituted phenol ring (15-18). PCB congeners have been considered as toxic chemicals, and classified with toxic levels. We found that some low toxic PCB congeners converted to high estrogenic OH-PCBs, especially para-position OH-PCBs by oxidative degradation.
Acknowledgments This study was a “High-Tech Research Center” project for private universities,with a matching fund subsidy from NEXT (Ministry of Education, Culture, Sports, Science and Technology), 2002-2006. The authors are indebted to Professor M. Koga of Prefectural University of Kumamoto, Japan, for his helpful advice and discussion. The authors gratefully acknowledge the assistance of Dr. T. Okumura of the Environmental Pollution Control Center, Osaka Prefectural Government, Japan, for use of 31 OH-PCB reference standards.
Literature Cited (1) Darnerud, P. O.; Morse, D. C.; Klasson-Wehler, E.; Brouwer, A. Binding of a 3,3’,4,4’-tetrachlorobiphenyl (CB-77) metabolite to fetal transthyretin and effects on fetal thyroid hormone levels in mice. Toxicology 1996, 106, 105-114. (2) Fielden, M. R.; Chen, I. Chittim, B.; Safe, S. H.; Zacharewski, T. R. Examination of the estrogenicity of 2,4,6,2′,6′-pentachlorobiphenyl (PCB 104), its hydroxylated metabolite 2,4,6,2′,6′pentachloro-4-biphenylol (HO-PCB 104), and a further chlorinated derivative, 2,4,6,2′,4′,6′-hexachlorobiphenyl (PCB 155). Environ. Health Perspect. 1997, 105, 1238-1248. 8768
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 22, 2005
(3) Brouwer, A.; Morse, D. C.; Lans, M. C.; Schuur, A. G.; Murk, A. J.; Klasson-wehler, E.; Bergman, A° .; Visser, T. J. Interactions of persistent environmental organohalogens with the thyroid hormone system: Mechanisms and possible consequences for animal and human health. Toxicol. Ind. Health. 1998, 14, 5984. (4) Danse, I. R.; Jaeger, R. J.; Kava, R.; Kroger, M.; London, W. M.; Lu, F. C.; Maickel, R. P.; Mcketta, J. J.; Newell, G. W.; Shindell, S.; Stare, F. J.; Whelan, E. M. Position paper of the American Council on Science and Health: public health concerns about environmental polychlorinated biphenyls (PCBs). Ecotoxicol. Environ. Saf. 1997, 38, 71-84. (5) Soto, A. M.; Sonnenschein, C.; Chung, K. L.; Fernandez, M. F.; Olea, N.; Serrano, F. O. The E-SCREEN assay as a tool to identify estrogens: An update on estrogenic environmental pollutants. Environ. Health Perspect. 1995, 103 (7), 113-122. (6) Iwata, H.; Tanabe, S.; Sakai, N.; Tatsukawa, R. Distribution of persistent organochlorines in the oceanic air and surface seawater and the role of oceans on their global transport and fate. Environ. Sci. Technol. 1993, 27, 1080-1098. (7) Letcher, R. J.; Klasson, W. E.; Bergman, A° . Methyl sulfone and hydroxylated metabolites of polychlorinated biphenyls. In The Handbook of Environmental Chemistry 3K; Paasivirta, J, Ed.; Springer-Verlag: Berlin. 2000; pp 315-359. (8) Ko¨ller, G.; Mo¨der, M.; Czihal, K. Peroxidative degradation of selected PCB: a mechanistic study. Chemosphere 2000, 41, 1827-1834. (9) Li, H.; Boon, P. J.; Lewis, E. W.; Berg, M.; Nyman, M.; Letcher, J. R. In vitro metabolism and inhibition studies to allocate the biotransformation of different metabolic groups of PCBs and testosterone to specific cytochrome P450 enzymes in liver microsomes of Baltic grey seal (Halichoerus grypus). Environ. Toxicol. Chem. 2003, 22, 636-644. (10) Sandau, C. D.; Ayotte, P.; Dewailly, EÄ .; Duffe, J.; Norstrom, R. J. Pentachlorophenol and Hydroxylated Polychlorinated Biphenyl Metabolites in Umbilical Cord Plasma of Neonates from Coastal Populations in Que´bec. Environ. Health Perspect. 2002, 110 (4), 411-417. (11) Bergman, A° .; Klasson, W. E.; Kuroki, H. Selective retention of hydroxylated PCB metabolites in blood. Environ. Health Perspect. 1994, 102, 464-469. (12) Weistrand, C.; Nore´n, K. Methylsulfonyl metabolites of PCBs and DDE in human tissues. Environ. Health Perspect. 1997, 105, 644-649. (13) Guvenius, M. D.; Hassanzadeh, P.; Bergman, A° .; Nore´n, K. Metabolites of polychlorinated biphenyls in human liver and adipose tissue. Environ. Toxicol. Chem. 2002, 21, 2264-2269. (14) Nore´n, K.; Lunde´n, A° .; Pettersson, E.; Bergman, A° . Methylsulfonyl metabolites of PCBs and DDE in human milk in Sweden, 19721992. Environ. Health Perspect. 1996, 104, 766-772. (15) Connor, K.; Ramamoorthy, K.; Moore, M.; Mustain, M.; Chen, I. Safe, S.; Zacharewski, T.; Gillesby, B.; Joyeux, A.; Balaguer, P. Hydroxylated polychlorinated Biphenyls (PCBs) as estrogens and antiestrogens: Structure-activity relation ships. Toxicol. Appl. Pharmacol. 1997, 145, 111-123. (16) Schultz, T. W. Estrogenicity of biphenylols: Activity in the yeast gene activation assay. Bull. Environ. Contam. Toxicol. 2002, 68, 332-338. (17) Shiraishi, F.; Okumura, T.; Nomachi, M.; Serizawa, S.; Nishikawa, J.; Edmonds, J. S.; Shiraishi, H.; Morita, M. Estrogenic and thyroid hormone activity of series of hydroxyl-polychlorinated biphenyls. Chemosphere 2003, 52, 33-42. (18) Arulmozhiraja, S.; Shiraishi, F.; Okumura, T.; Iida, M.; Takigami, H.; Edmonds, S. J.; Morita, M. Structural requirements for the interaction of 91 hydroxylated polychlorinated biphenyls with estrogen and thyroid hormone receptors. Toxicol. Sci. 2005, 84, 49-62. (19) Ollis, D. F.; Hsiao, C. Y.; Budiman, L.; Lee, C. L. Heterogeneous photoassisted catalysis: conversion of perchloroethylene, dichloroethane, chloroacetic acids, and chlorobenzene. J. Catal. 1984, 88, 89-96. (20) Ohko, Y.; Ando, I.; Niwa, C.; Tatsuma, T.; Yamamura, T.; Nakashima, T.; Kubota, Y.; Fujishima, A. Degradation of Bisphenol-A in Water by TiO2 Photocatalyst. Environ. Sci. Technol. 2001, 35, 2365-2368. (21) Horikoshi, S.; Tokunaga, A.; Hidaka, H.; Serpone, N. Environmental remediation by an integrated microwave/UV-illumination technique VII. Thermal/non-thermal effects in the microwave-assisted photocatalyzed meneralization of bisphenol A. J. Photochem. Photobiol., A 2004, 162, 33-40.
(22) Watanabe, N.; Horikoshi, S.; Kawabe, H., Sugie, Y.; Zhao, J.; Hidaka, H. Photodegradation mechanism for bisphenol A at the TiO2/H2O interfaces. Chemosphere 2003, 52, 851-859. (23) Nishikawa, J.; Saito, K.; Goto, J.; Dakeyama, F.; Matsuo, M.; Nishihara, T. New screening methods for chemicals with hormonal activities using interaction of nuclear hormone receptor with coactivator. Toxcol. Appl. Pharmacol. 1999, 154, 76-83. (24) Nishihara, T.; Nishikawa, J.; Kanayama, T.; Dakeyama, F.; Saito, M.; Imagawa, M.; Takatori, S.; Kitagawa, T.; Hori, S.; Utsumi, H. Estrogenic activities of 517 chemicals by yeast two-hybrid assay. J. Health Sci. 2000. 46 (4), 282-298. (25) Okumura, T.; Shibata, J. Investigation on the behavior of PCBs in the environment. Research Report of the Environmental Pollution Control Center. Osaka, Prefectural Government, Japan, 1975, (in Japanese). (26) Kato, K.; Tsuzuki, A.; Torii, Y.; Taoda, H.; Butsugan, Y. Morphology of thin anatase coating prepared from alkoxide solutions containing organic polymer, affecting the photocatalytic decomposition of aqueous acetic acid. J. Mater. Sci. 1995, 30, 837841. (27) Tanizaki, T.; Kadokami, K.; Shinohara, R. Catalytic photodegradation of endocrine disrupting chemicals using titanium dioxide photosemiconductor thin films. Bull. Environ. Contam. Toxicol. 2002, 68, 732-739.
(28) Shiraishi, F.; Shiraishi, H.; Nishikawa, J.; Nishikawa, T.; Morita, M. Development of a simple operational estrogenicity assay system using the yeast two-hybrid system. J. Environ. Chem. 2000, 10, 57-64 (in Japanese). (29) Miyahara, M.; Ishibashi, H.; Inudo, M.; Nishijima, H.; Iguchi, T.; Guillette, J. L., Jr.; Arizono, K. Estrogenic activity of a diet to estrogen receptors -R and -β in an experimental animal. J. Health Sci. 2003, 49, 481-491. (30) Stewart, J. J. P. Application of localized molecular orbitales to the solution of semiempirical self-consistent field equations. J. Quantum Chem. 1996, 58, 133-146. (31) Yao, Y.; Kakimoto, K.; Ogawa, H.; Kato, Y.; Kadokami, K.; Shinohara, R. Further study on the photochemistry of nonortho substituted PCBs by UV irradiation in alkaline 2-propanol. Chemosphere 2000, 40, 951-956. (32) Chang, F.; Chiu, T.; Yen, J.; Wang, Y. Dechlorination pathways of ortho-substituted PCBs by UV irradiation in n-hexane and their correlation to the charge distribution on carbon atom. Chemosphere 2003, 51, 775-784.
Received for review April 26, 2005. Revised manuscript received September 5, 2005. Accepted September 9, 2005. ES050791A
VOL. 39, NO. 22, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8769
