Introduction

Abstract Introduction Results Discussion References

PCBs¹ are ubiquitous environmental contaminants that have been accumulated in the tissues of wildlife and humans (1). The rate of metabolism on PCBs is governed by the extent and position of chlorine substitution, as well as the metabolic capacity that is depending on animal species (2). The CBs with two adjacent meta/para unsubstituted carbon atoms facilitate the metabolism to give OH products or their conjugates (3). Above all, the CBs with a 2,5-substitution produce both sulfur-containing and OH metabolites (4, 5). This pattern may stabilize the 3,4-epoxide product from P450 oxidation and allow more reaction with glutathione, leading to MeS and MeSO₂ metabolites via mercapturic acid pathway (6, 7).

PCB metabolites are generally excreted from animals, but some of them have been shown to be selectively retained in different tissues or blood. For example, the MeSO₂-CBs have been identified in the blubber of a variety of mammals (8, 9), as well as in human tissues (10, 11). To date, the only MeSO₂-CBs identified in wildlife are those originating from CBs with a 2,5- or 2,3,6-substitution. Among them, only a few 3-MeSO₂-CBs were shown to have a strong and specific retention in the liver of seals (Halichoerus grypus) and otters (Lutra lutra) from the Swedish environment.² In contrast, some 4-MeSO₂-CBs were observed to be strongly retained in the bronchial mucosa and/or in the kidney cortex of mice (12). The binding mechanism for the 4-MeSO₂-CBs was found to be due to a strong binding to a uteroglobin-like protein (13-15). Our recent study has shown that several 3-MeSO₂-CBs are potent phenobarbital-like inducers of hepatic microsomal drug-metabolizing enzymes in rats, whereas their isomeric 4-MeSO₂-CBs are inactive (16). On the other hand, certain hydroxylated metabolites of PCB have also been regarded as potentially hazardous compounds. For example, NIH shift metabolites (e.g. 4-OH-CBs with 3,5-chlorine substitution) were selectively retained in the blood of mammals (17) and associated with alterations in thyroid hormone levels (18).

The biological and toxicological effects of PCB metabolites are likely different, depending on the position of substituents and the degree of chlorination. However, there is no information on the comparative profiles of metabolites from different CBs. The present investigation therefore is undertaken to better characterize the rate of fecal excretion and tissue distribution of both hydroxy and sulfur-containing metabolites formed in rats dosed with TriCB, TetraCB, PentaCB, and HexaCB, with a 2,5-substitution in one ring and with at least 4-substitution in the other (fig. 1).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Structures of the chlorinated biphenyls studied.

Materials and Methods

Chemicals. TriCB, TetraCB, PentaCB, and HexaCB (>99% pure, respectively) were prepared by Cadogan diaryl coupling reaction (19) of 4-chloro-, 3,4-dichloro-, 2,4,5-trichloro-, and 2,3,4,5-tetrachloroaniline with p-dichlorobenzene, respectively. The 3- and 4-MeSO₂ derivatives of four CBs were synthesized as described previously (20). 2,3,3,4,4,5,5-Heptachlorobiphenyl and 3-MeSO₂-4-methyl-2,3,4,5,5-pentachlorobiphenyl were used as internal standards for the analysis of metabolites. All solvents and chemicals used were of analytical grade.

Synthesis of Metabolites. The 3- and 4-MeS derivatives of four CBs were prepared by the coupling reaction, according to the method of Cadogan (19), of appropriate chlorinated anilines with 2,5-dichlorothioanisole. The thioanisole was prepared by diazotization of 2,5-dichloroaniline with sodium nitrite and subsequent reaction with sodium methanethiolate (Tokyo Kasei Kogyo Co. Ltd., Tokyo, Japan). The isomers formed were separated to each other by an ODS column (100 × 22 mm i.d., Kusano Kagakukikai, Tokyo, Japan), with acetonitrile:water (9:1, v/v) as a mobile phase. The 3- and 4-MeO derivatives were synthesized from chlorinated anilines and 2,5-dichloroanisole. Some of the MeS and MeO derivatives have been reported partly elsewhere (20-23). The spectral assignments for the derivatives are shown herein. The GC retention times relative to 2,3,3,4,4,5,5-heptachlorobiphenyl are also shown.

3-MeS-TriCB. MS (70 eV): m/z (relative abundance, %) = 306 (35), 304 (100), 302 (95, M⁺), 269 (11, M⁺-SH), 252 (23, M⁺-CH₃Cl), 217 (10, M⁺-CH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 2.50 (3H, s, CH₃), 7.06 (1H, d, J = 2.0 Hz, 4-H), 7.07 (1H, d, J = 2.0 Hz, 6-H), 7.33 (2H, d, J = 8.6 Hz, 2-H and 6-H), 7.40 (2H, d, J = 8.6 Hz, 3-H and 5-H). Relative GC retention time, 0.791.

4-MeS-TriCB. MS (70 eV): m/z (%) = 306 (33), 304 (100), 302 (98, M⁺), 287 (23, M⁺-CH₃), 252 (30, M⁺-CH₃Cl), 217 (17, M⁺-CH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 2.51 (3H, s, CH₃), 7.19 (1H, s, 3-H), 7.30 (1H, s, 6-H), 7.35 (2H, d, J = 8.8 Hz, 2-H and 6-H), 7.40 (2H, d, J = 8.8 Hz, 3-H and 5-H). Relative GC retention time, 0.797.

3-MeS-TetraCB. MS (70 eV): m/z (%) = 340 (56), 338 (100), 336 (78, M⁺), 303 (11, M⁺-SH), 286 (15, M⁺-CH₃Cl). ¹H-NMR (500 MHz, chloroform-d);  = 2.51 (3H, s, CH₃), 7.06 (1H, d, J = 2.3 Hz, 4-H), 7.10 (1H, d, J = 2.3 Hz, 6-H), 7.23 (1H, dd, J = 8.1, 2.0 Hz, 6-H), 7.48 (1H, d, J = 2.2 Hz, 2-H), 7.50 (1H, d, J = 8.1 Hz, 5-H). Relative GC retention time, 0.928.

4-MeS-TetraCB. MS (70 eV): m/z (%) = 340 (55), 338 (100), 336 (76, M⁺), 321 (21, M⁺-CH₃), 286 (30, M⁺-CH₃Cl). ¹H-NMR (500 MHz, chloroform-d);  = 2.52 (3H, s, CH₃), 7.20 (1H, s, 3-H), 7.26 (1H, dd, J = 8.5, 2.0 Hz, 6-H), 7.30 (1H, s, 6-H), 7.50 (1H, d, J = 8.2 Hz, 5-H), 7.51 (1H, d, J = 2.2 Hz, 2-H). Relative GC retention time, 0.939.

3-MeS-PentaCB. MS (70 eV): m/z (%) = 374 (60), 372 (100), 370 (58, M⁺), 337 (10, M⁺-SH), 320 (10, M⁺-CH₃Cl). ¹H-NMR (500 MHz, chloroform-d);  = 2.51 (3H, s, CH₃), 7.00 (1H, d, J = 2.3 Hz, 4-H), 7.14 (1H, d, J = 2.3 Hz, 6-H), 7.34 (1H, s, 6-H), 7.59 (1H, s, 3-H). Relative GC retention time, 1.105.

4-MeS-PentaCB. MS (70 eV): m/z (%) = 374 (58), 372 (100), 370 (55, M⁺), 355 (12, M⁺-CH₃), 320 (20, M⁺-CH₃Cl). ¹H-NMR (500 MHz, chloroform-d);  = 2.52 (3H, s, CH₃), 7.20 (1H, s, 3-H), 7.23 (1H, s, 6-H), 7.36 (1H, s, 6-H), 7.59 (1H, s, 3-H). Relative GC retention time, 1.128.

3-MeS-HexaCB. MS (70 eV): m/z (%) = 408 (82), 406 (100), 404 (52, M⁺), 371 (7, M⁺-SH), 354 (10, M⁺-CH₃Cl). ¹H-NMR (500 MHz, chloroform-d);  = 2.52 (3H, s, CH₃), 6.99 (1H, d, J = 2.3 Hz, 4-H), 7.15 (1H, d, J = 2.3 Hz, 6-H), 7.31 (1H, s, 6-H). Relative GC retention time, 1.154.

4-MeS-HexaCB. MS (70 eV): m/z (%) = 408 (83), 406 (100), 404 (52, M⁺), 354 (14, M⁺-CH₃Cl). ¹H-NMR (500 MHz, chloroform-d);  = 2.53 (3H, s, CH₃), 7.20 (1H, s, 3-H), 7.21 (1H, s, 6-H), 7.32 (1H, s, 6-H). Relative GC retention time, 1.175.

3-MeO-TriCB. MS (70 eV): m/z (%) = 290 (30), 288 (98), 286 (100, M⁺), 243 (27, M⁺-COCH₃), 173 (20, M⁺-COCH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 3.94 (3H, s, CH₃), 6.92 (1H, s, 4-H), 7.02-7.41 (5H, m). Relative GC retention time, 0.674.

4-MeO-TriCB. MS (70 eV): m/z (%) = 290 (30), 288 (96), 286 (100, M⁺), 271 (35, M⁺-CH₃), 243 (32, M⁺-COCH₃), 173 (31, M⁺-COCH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 3.93 (3H, s, CH₃), 7.02 (1H, s, 3-H), 7.32-7.41 (5H, m). Relative GC retention time, 0.682.

3-MeO-TetraCB. MS (70 eV): m/z (%) = 324 (52), 322 (100), 320 (81, M⁺), 277 (17, M⁺-COCH₃), 207 (12, M⁺-COCH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 3.94 (3H, s, CH₃), 6.92 (1H, d, J = 2.0 Hz, 4-H), 6.95 (1H, d, J = 2.0 Hz, 6-H), 7.25 (1H, dd, J = 8.5, 2.0 Hz, 6-H), 7.50 (1H, d, J = 2.2 Hz, 2-H), 7.50 (1H, d, J = 8.2 Hz, 5-H). Relative GC retention time, 0.780.

4-MeO-TetraCB. MS (70 eV): m/z (%) = 324 (45), 322 (100), 320 (77, M⁺), 305 (25, M⁺-CH₃), 277 (22, M⁺-COCH₃), 207 (20, M⁺-COCH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 3.94 (3H, s, CH₃), 7.03 (1H, s, 3-H), 7.25 (1H, dd, J = 8.0, 2.5 Hz, 6-H), 7.32 (1H, s, 6-H), 7.50 (1H, d, J = 2.2 Hz, 2-H), 7.50 (1H, d, J = 8.5 Hz, 5-H). Relative GC retention time, 0.788.

3-MeO-PentaCB. MS (70 eV): m/z (%) = 358 (62), 356 (100), 354 (55, M⁺), 311 (20, M⁺-COCH₃), 241 (24, M⁺-COCH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 3.95 (3H, s, CH₃), 6.86 (1H, d, J = 2.2 Hz, 4-H), 6.99 (1H, d, J = 2.2 Hz, 6-H), 7.35 (1H, s, 6-H), 7.59 (1H, s, 3-H). Relative GC retention time, 0.901.

4-MeO-PentaCB. MS (70 eV): m/z (%) = 358 (66), 356 (100), 354 (60, M⁺), 339 (15, M⁺-CH₃), 311 (27, M⁺-COCH₃), 241 (35, M⁺-COCH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 3.95 (3H, s, CH₃), 7.04 (1H, s, 3-H), 7.26 (1H, s, 6-H), 7.36 (1H, s, 6-H), 7.59 (1H, s, 3-H). Relative GC retention time, 0.915.

3-MeO-HexaCB. MS (70 eV): m/z (%) = 392 (80), 390 (100), 388 (58, M⁺), 345 (15, M⁺-COCH₃), 275 (31, M⁺-COCH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 3.95 (3H, s, CH₃), 6.84 (1H, d, J = 2.2 Hz, 4-H), 7.00 (1H, d, J = 2.2 Hz, 6-H), 7.31 (1H, s, 6-H). Relative GC retention time, 1.082.

4-MeO-HexaCB. MS (70 eV): m/z (%) = 392 (82), 390 (100), 388 (52, M⁺), 373 (11, M⁺-CH₃), 345 (18, M⁺-COCH₃), 275 (15, M⁺-COCH₃Cl₂). ¹H-NMR (500 MHz, chloroform-d);  = 3.96 (3H, s, CH₃), 7.04 (1H, s, 3-H), 7.24 (1H, s, 6-H), 7.33 (1H, s, 6-H). Relative GC retention time, 1.096.

Instruments. GC/MS was operated on a JMS-AX505W (JEOL) connected to a JMA-DA5000 data system in the electron ionization mode. The GC separations were performed on a Hewlett-Packard 5890 instrument equipped with a DB-5MS fused silica capillary column (30 m × 0.25 mm, 0.25 µm film thickness, J&W Scientific, Inc., Folsom, CA), with helium as carrier gas. The temperature of injector was 250°C and detector was 300°C. The oven temperature was programmed from 70°C (2 min) to 220°C at 20°C/min, then to 280°C at 4°C/min and held at 280°C for 20 min. Injection was made in the splitless mode using an autoinjector (AOC-14; Shimadzu, Kyoto, Japan). ¹H-NMR spectra were obtained from samples dissolved in CDCl₃ with a GSX-500 (500 MHz, JEOL) NMR spectrometer using tetramethylsilane as internal standard. Quantification was made on a Shimadzu GC-14A gas chromatograph fitted with an electron capture (⁶³Ni) detector. The column condition was the same as for the GC/MS.

Animals. Male Wistar rats weighing an average of 200 g were housed in an air-conditioned room with free access to a commercial chow and tap water. Groups of four rats were given a single intraperitoneal dose of only one of TriCB, TetraCB, PentaCB, and HexaCB (342 µmol/kg, each), dissolved in Panacete 810, a mixture of glycerides of medium chain fatty acids (Nippon Oils and Fats Co., Tokyo, Japan). Feces were collected daily for 4 days. Urine was collected daily for 4 days from TriCB-treated rats. The rats were killed 96 hr after dosing, and the selected tissues were removed for analyses of persistent metabolites.

Extraction. Tissues were homogenized with acetone:hexane (2:1, v/v) and filtered through anhydrous Na₂SO₄, and the extracts were concentrated. The feces were extracted with acetone:hexane (2:1, v/v, 120 ml) for 18 hr in a Soxhlet apparatus. Each extract was concentrated to small volume, and 0.2 M phosphoric acid, including 0.9% NaCl was added, then reextracted twice with hexane. The combined hexane extracts were concentrated and dissolved in hexane:dichloromethane (1:1, v/v) for GPC. Urine samples were diluted with 4 volumes of 2 M HCl and heated at 80°C for 2 hr and extracted 3 times with hexane:methyl tert-butyl ether (9:1, v/v) before GPC. Blood was diluted with 4 volumes of 2 M KOH:ethanol (5:2, v/v) and heated for 2 hr at 70°C. After acidification with HCl, the aqueous solution was extracted with hexane for GPC purification.

GPC. A slurry of Bio-Beads S-X3 (50 g, 200-400 mesh, Bio-Rad Lab., Richmond, CA) in hexane:dichloromethane (1:1, v/v) was packed in a Pharmacia SR column (450 × 25 mm, i.d.). The same solvent was pumped through the column at a flow rate of 4 ml/min with a UV detector at 254 nm on line (LC-6A, SPD-6A, SCL-6A system; Shimadzu). Samples were injected via a Rheodyne model 7725 injector equipped with a 2-ml loop, and the fraction of xenobiotics between 36 and 48 min was collected by a fraction collector (FCV-100B, Shimadzu).

Partition Procedure. The metabolite fraction from GPC was partitioned into OH, MeS, and MeSO₂ fractions by the following procedure. All samples, dissolved in hexane, were extracted with 2 M KOH:ethanol (5:2, v/v). The aqueous solution was acidified by 6 M HCl, and then extracted with hexane:tert-butyl methyl ether (9:1, v/v) for acidic compounds (OH fraction). This fraction was further concentrated to dryness and derivatized by diazomethane in diethylether that was prepared from 1-methyl-3-nitro-1-nitrosoguanidine (Aldrich Chemical Co., Milwaukee, WI) and 6 M NaOH. After the reactant was allowed to stand for 30 min, the solvent was evaporated, and the residue was dissolved in hexane for GC or GC/MS analysis. The neutral metabolites in hexane extracts were separated on aluminium oxide (1 g, neutral, activity I; Merck, Darmstadt, Germany), eluted with hexane (30 ml) for nonpolar metabolites (MeS fraction), and successively with dichloromethane (30 ml) for middle-polar metabolites (MeSO₂ fraction).

Identification and Quantification. The identification of metabolites was performed on GC/MS before or after the partition procedure. The quantification of metabolites was done on GC using two internal standards, 2,3,3,4,4,5,5-heptachlorobiphenyl for OH and MeS fractions and 3-MeSO₂-4-methyl-2,3,4,5,5-pentachlorobiphenyl for MeSO₂ fraction. The former internal standard was added to OH fraction just after partition procedure, and the OH metabolites were quantified as MeO derivatives. The latter internal standard was added to samples before the Soxhlet extraction or homogenation procedure.

	Results

Abstract Introduction Results Discussion References

Identification of Fecal Metabolites. TriCB. Figure 2A shows the GC profile of fecal extracts from TriCB-treated rats after GPC purification. Components Ia, Ib, and Ic were transferred to the OH fraction by partition procedure, and the Ia and Ib exhibited mass spectra consistent with monohydroxy TriCB. The methylated Ia and Ib gave rise to an abundant fragment at [M-CH₃CO]⁺. In addition, only the methylated Ib gave a significant fragment at [M-CH₃]⁺ that was characteristic for para-substituted MeO derivatives (23). Thus, the methylated Ia and Ib components were identified as 3- and 4-MeO-TriCBs, respectively. A minor metabolite Ic was identified as dihydroxy-triBC (M⁺ at m/z 316 as dimethoxy derivative) in this fraction (24). Components IIa and IIb were present in the MeS fraction, together with unchanged TriCB. Both compounds exhibited mass spectra with an intense M⁺ at m/z 302, and less abundant ions of [M-SH]⁺ and [M-CH₃Cl]⁺. Only metabolite IIb gave an abundant ion [M-CH₃]⁺. Thus, the IIa and IIb were identified as 3- and 4-MeS-TriCBs, respectively. Two minor components IIIa and IIIb were observed in the MeSO₂ fraction. The fragmentation of IIIb showed an intense [M-SOCH₃]⁺ that was characteristic for para-substituted MeSO₂-CBs (20, 21). Therefore, the IIIa and IIIb were identified as 3- and 4-MeSO₂-TriCBs, respectively. This fraction also contained intermediate products, 3- and 4-MeSO-TriCBs, which gave mass spectra dominated by fragment ions [M-CH₃]⁺ and [M-O]⁺, instead of a less abundant M⁺ at m/z 318 (24). The 4-MeSO-TriCB had additional fragments [M-OCH₃]⁺ and [M-SCH₃]⁺ that could be distinguished from the 3-MeSO-TriCB. They were rarely present in feces, but it should be noted that they spontaneously occur by long-term standing of the MeS fraction at room temperature and can in part be converted to the corresponding 3- and 4-MeSO₂-TriCBs.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   GC profiles of fecal extracts from rats dosed with (A) TriCB, (B) TetraCB, (C) PentaCB, and (D) HexaCB. The feces after intraperitoneal injection of individual congeners (342 µmol/kg, each) were collected for 4 days. After Soxhlet extraction of feces with acetone:hexane (2:1, v/v), the materials were purified by GPC. IS-1, 2,3,3,4,4,5,5-heptachlorobiphenyl; IS-2, 3-MeSO2-4-methyl-PentaCB.

TetraCB. The metabolite profile of TetraCB (fig. 2B) resembled that of TriCB (fig. 2A). The OH fraction contained components IVa, IVb, and IVc. The methylated IVa and IVb exhibited M⁺ at m/z 320, and only the methylated IVb gave a significant fragment at [M-CH₃]⁺. Consequently, the methylated IVa and IVb were identified as 3- and 4-MeO-TetraCBs, respectively. Metabolite IVc was identified as dihydroxy tetraCB (M⁺ at m/z 350 as dimethoxy derivative) and present at a lower level. Metabolites Va and Vb were present in the MeS fraction and identified as 3- and 4-MeS-TetraCBs, respectively, whereas VIa and VIb were in the MeSO₂ fraction as 3- and 4-MeSO₂-TetraCBs, respectively.

PentaCB. Figure 2C shows the GC profile of fecal extract from PentaCB-treated rats after GPC purification. The OH fraction was dominated by components VIIa and VIIb. The methylated VIIa exhibited a mass spectrum with M⁺ at m/z 354 and an abundant fragment [M-COCH₃]⁺, but showed no [M-CH₃]⁺, which was identical with that of authentic 3-MeO-PentaCB. The methylated VIb was identified as 4-MeO-PentaCB. On the other hand, the metabolites VIIIa and VIIIb were observed in the MeS fraction and identified as 3- and 4-MeS-PentaCBs, respectively. Metabolite IX in the MeSO₂ fraction was identified as 4-MeSO₂-PentaCB.

HexaCB. The metabolite profile of HexaCB (fig. 2D) was similar to that of PentaCB (fig. 2C). The OH fraction contained components Xa and Xb. The methylated VXa and VXb were not distinguished to each other by mass fragmentation, because the diagnostic [M-CH₃]⁺ was less abundant. However, the GC retention times of the methylated Xa and Xb were identical to those of authentic 3-MeO- and 4-MeO-HexaCB, respectively. Metabolite XIa and XIb were present in the MeS fraction and identified as 3-MeS- and 4-MeS-HexaCB, respectively. The MeSO₂ fraction contained 3- and 4-MeSO₂-HexaCBs in minor amounts.

Excretion Ratio of Metabolites. Figure 3 shows the fecal excretion of OH and MeS metabolites during the first 4 days after injection of each isomer. As the degree of chlorination increased, the excreted amounts of OH products increased, whereas those of MeS metabolites greatly decreased. As a result, the MeS/OH excretion ratios in feces were 2.8 for TriCB, 1.3 for TetraCB, 0.04 for PentaCB, and 0.02 for HexaCB. The excreted amounts of both OH and MeS metabolites accounted for 11%, 7.3%, 4.0%, and 3.8% of the dose within 96 hr after administration of TriCB, TetraCB, PentaCB, and HexaCB, respectively. The MeSO₂-CBs were present in 10-fold lower levels than the precursor MeS metabolites in all cases. The 3-/4-OH substitution ratios were 0.6 for TriCB, 1.4 for TetraCB, 21 for PentaCB, and 35 for HexaCB, whereas the 3-/4-MeS substitution ratios were 1.2 for TriCB, 0.8 for TetraCB, 0.18 for PentaCB, and 0.12 for HexaCB. The 3-/4-MeSO₂ ratios were similar to the corresponding 3-/4-MeS ratios in all cases.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Fecal excretion of OH and MeS metabolites during first 4 days after injection of individual congeners to rats.

Urinary Metabolites. Urine samples from TriCB-treated rats were analyzed and 3- and 4-OH-TriCBs were detected at less than one-tenth the level of OH metabolites in feces. This result was in accordance with the result from ¹⁴C-labeled TriCB-treated rats (7). The levels of MeS- and MeSO₂-CBs were below detection limit (<0.01 nmol/ml).

MeSO₂ Metabolites in Tissues from TetraCB-Treated Rats. Several tissues from TetraCB-treated rats were analyzed for persistent metabolites, and 3- and 4-MeSO₂-TetraCBs were identified. 3- and 4-OH-TetraCBs were also detected in tissues at levels lower than MeSO₂-TetraCB (data not shown), but 3- and 4-MeS-TetraCBs were not detected in any tissues. Table 1 shows the tissue concentrations of unchanged TetraCB and its 3- and 4-MeSO₂ metabolites at 4 days after injection of TetraCB. On a wet weight basis, the highest concentration of MeSO₂-TetraCBs was found in adipose tissue, followed by liver and lung, whereas the lowest was in the blood, spleen, and testis. The 3-/4-MeSO₂-TetraCB concentration ratio was highest in the spleen (1.89), adipose tissue (1.85), and adrenal (1.84), whereas the lowest was in the lung (0.18). The metabolites/unchanged CB ratio was highest in the liver (11.9), followed by the blood (8.57), whereas the lowest was in the adipose tissue (0.15) and intestine (0.22).

Comparison of MeSO₂-CB Metabolites of Each Congeners. Adipose tissue, kidney, liver, and lung were selected for the comparison of distribution pattern of MeSO₂-CBs. The results are shown in fig. 4, A-D. In general, as the degree of chlorination increased, the formation of MeSO₂-CBs decreased largely in all tissues. The decline of MeSO₂-CBs in tissues was correlated to that of the precursor MeS-CBs in feces (cf. figs. 3 and 4). In the adipose tissue (fig. 4A), the unchanged CB was more heavily retained with the degree of chlorination increased. For example, the concentration (0.07 µmol/g) of unchanged TriCB accounted for only 0.1% of the dose, whereas the concentration (2.3 µmol/g) of unchanged HexaCB accounted for 4.5% of the dose 96 hr after administration. However, in TriCB-treated rats, the MeSO₂-CBs were distributed in higher levels than the unchanged CB. In the kidney (fig. 4B), the MeSO₂-CB/unchanged CB distribution ratio was the highest in TriCB-treated rats, but the ratios greatly decreased with increasing the chlorination degree. In the liver (fig. 4C), the levels of MeSO₂-CBs from TriCB- and TetraCB-treated rats exceeded 8.5- and 12-fold the unchanged CB, respectively, although the unchanged TriCB was present in high levels unexpectedly. The levels of MeSO₂-CB from PentaCB- and HexaCB-treated rats were much lower than those of the unchanged CB. In the lung (fig. 4D), the levels of unchanged CBs were not dependent on the chlorination degree. The metabolites in the lung were dominated by 4-MeSO₂-CB in all cases, which exceeded the unchanged CB, with the ratios of 17.5 for TriCB, 8.4 for TetraCB, 2.2 for PentaCB, and 1.1 for HexaCB.

View larger version (40K): [in this window] [in a new window]   Fig. 4.   Tissue distribution of unchanged congener and its MeSO2 metabolites from rats dosed with four individual congeners. Rats were treated intraperitoneally with four individual congeners (342 µmol/kg, each). Four days after injection, adipose tissue (A), kidney (B), liver (C), and lung (D) were analyzed for the unchanged congener and its 3- and 4-MeSO2 metabolites.

	Discussion

Abstract Introduction Results Discussion References

The present study shows that the metabolism of CBs with 2,5-chlorine substitution in rats involves methylthiolation, as well as hydroxylation on the 3- or 4-position of the phenyl ring, with different substitution patterns depending on the degree of chlorination. As the chlorination degree increased, the OH metabolites were more efficiently excreted in feces, whereas the MeS metabolites were less excreted. Thus, the excretion ratio of MeS and OH metabolites greatly decreased from 2.8 for TriCB to 0.02 for HexaCB. This finding indicates that the metabolism of the lower CBs involves preferentially methylthiolation, whereas the higher CBs undergo predominantly hydroxylation. In fact, data from TriCB-treated rats is in accordance with the findings by Bakke et al. (6), who reported that >20% of the radioactivity in oral doses of ¹⁴C-labeled TriCB given to rats was excreted as mercapturic acid pathway metabolites, and present in feces as mainly MeS metabolites rather than OH metabolites (7). Several tetraCBs have been also shown to produce MeS metabolites in rats and mice (4, 25, 26), although the formation ratios of MeS-/OH-CB have not been discussed. Berlin et al. (27) have reported that 75% of an oral dose of PentaCB to mice is excreted in feces as mainly OH products, both free and conjugated. It is thus obvious that the metabolism of the higher CBs proceeds in different pathways from the lower CBs.

In fecal excretion, there was a significant difference in the 3-/4-OH substitution pattern between the lower and higher CB congeners. TriCB and TetraCBs were metabolized almost equally to 3- and 4-OH metabolites, but PentaCB and HexaCBs gave exclusively their 3-OH metabolites. The selective formation of 3-OH metabolites was explained by Preston et al. (5), who reported that the major route for the formation of the 3-OH metabolite from 2,2,5,5-tetraCB involves a nonarene oxide direct hydroxylation mechanism by phenobarbital-induced rat liver microsomes. Although in this study tetraCB did not selectively form the 3-OH metabolites, hydroxylation of the higher CBs may involve the similar mechanism as that of 2,2,5,5-tetraCB. Bergman et al. (17) reported that CBs with 3,4-substitution were metabolized to NIH shift metabolites in rats (e.g. 4-OH-3,3,4,5-tetraCB formed from 3,3,4,4-tetraCB) that were subsequently retained in the blood and bound to a thyroxin-transporting protein (18). In the present study, however, 4-OH-CB via a 1,2 shift of 4-chlorine in the 4-, 3,4-, 2,4,5-, or 2,3,4,5-chlorinated ring was not detected in rat feces. Recently, mink (Mustela vison) treated with ¹⁴C-labeled PentaCB has been shown to produce 4-OH-2,2,3,5,5-pentaCB (28), although in the same condition, mice produced no such metabolites (28).

It is in general considered that the MeS metabolites are formed via nucleophilic reaction of 3,4-arene oxide intermediate of PCB with glutathione. The glutathione adduct is excreted in the bile as a mercapturic acid and is degraded to the corresponding cysteine conjugate. The conjugates are cleaved by an intestinal mucosal -lyase to thiols and then methylated to give 3- or 4-substituted MeS product. In this pathway, the pair is likely produced at similar amounts, but the formation ratio of 3- and 4-MeS metabolites is apparently dependent on animal species. For example, 2,2,5,5-tetraCB gives rise to the 3- and 4-MeS metabolites with a ratio of 0.13 in rats and 1.6 in mice (4). This may be explained by involvement of the regioselective attack of glutathione to the 3,4-epoxide in mediation of the enzyme in liver cytosol (4). However, the present study showed that PentaCB and HexaCBs yield selectively the 4-MeS metabolites, whereas TriCB and TetraCBs yield almost equally the 3- and 4-MeS metabolites. Although the mechanism for the selectivity for 4-MeS formation is unclear, the higher CBs may involve an alternate metabolic pathway via nonarene oxide intermediate.

The formed MeS derivatives are in part reabsorbed in the intestinal tract and transferred to liver and oxidized by P450 to the corresponding MeSO₂-CBs, which are eventually accumulated in tissues. In fact, the excreted amounts of the MeS metabolites were correlated to the tissue concentrations of the corresponding MeSO₂-CBs (figs. 2 and 3). The 3-/4-MeS substitution ratios in feces also reflected to some extent the 3-/4-MeSO₂-CBs ratios in tissues (table 2). For example, the distribution ratios of 3-/4-MeSO₂-CBs in liver decreased, ranging from 1.5 for TriCB to 0.23 for HexaCB, correlating to the 3-/4-MeS excretion ratio ranging from 1.2 for TriCB to 0.14 for HexaCB. However, the affinity of 3- and 4-MeSO₂-CBs were different between tissues, as shown by tissue/blood concentration ratios. The liver/blood concentration ratios showed that the 3-MeSO₂-CBs derived from PentaCB and HexaCB had a slightly higher affinity for liver than the 4-MeSO₂-CBs had. Recently, a strong and selective retention of 3-MeSO₂-CBs were observed in the liver of seals (Halichoerus grypus) and otters (Lutra lutra) from the Swedish environment.² The mechanism that accounts for the affinity is yet unknown. Our recent study showed that only 3-MeSO₂-CBs strongly induced phenobarbital-inducible forms of P450 (CYP2B1, CYP2B2, CYP3A2, and CYP2C6), cytochrome b₅, aminopyrine N-demethylase, 7-ethoxycoumarin O-deethylase, and benzo[a]pyrene hydroxylase in rats (16). We also found that the 3-MeSO₂-PentaCB plays an important role in the induction of the drug-metabolizing enzymes by the parent CB (29). On the other hand, the lung/blood concentration ratios indicated that the 4-MeSO₂-CB derived from all CBs tested had a pronounced affinity for the lung (table 2). This is in agreement with the previous results from mice dosed with several ¹⁴C-labeled CBs (30). The binding mechanism for the 4-MeSO₂-CBs detected in the lung cytosol has been investigated and found to be due to a strong and selective binding to a uteroglobin-like protein (13-15). It was also reported that the 4-MeSO₂-CBs form a complex with _2u-globin in kidney and urine in male rats (31), and with liver fatty acid binding protein isolated from rat intestinal mucosa (32).

In conclusion, the formation rate and substitution patterns of OH- and MeSO₂-CBs derived from CBs with a 2,5-substitution are largely influenced by the chlorination degree in the other ring. The lower CBs give rise to similar amounts of 3- and 4-OH, and 3- and 4-MeS metabolites via 3,4-arene oxide, whereas the higher CBs give predominantly 3-OH products, with minor 4-MeS derivatives, probably via nonarene oxide pathway. Based on the formation rate of MeS metabolites, the corresponding MeSO₂-CBs are distributed in all tissues, but the selective uptake of 4-MeSO₂-CBs in the lung are observed in all cases. It is not likely that the 4-chlorine substituted ring undergoes metabolism in this case. However, CBs with 4-substitution in both rings would undergo methylthiolation in rats, because MeSO₂ metabolites have been detected in mice dosed with 2,2,4,4-tetraCB (33) and 3,3,4,4-tetraCB (34). Further studies on the tissue distribution of OH and MeSO₂ metabolites derived from different CB congeners are necessary for understanding the metabolite-related toxicity of PCBs in wildlife species, including humans.