Oxidation of polychlorinated biphenyls by liver tissue slices from phenobarbital-pretreated mice is congener-specific and atropselective

Abstract

Mouse models are powerful tools to study the developmental neurotoxicity of polychlorinated biphenyls (PCBs); however, studies of the oxidation of chiral PCB congeners to potentially neurotoxic hydroxylated metabolites (OH-PCBs) in mice have not been reported. Here, we investigate the atropselective oxidation of chiral PCB 91 (2,2',3,4',6-pentachlorobiphenyl), PCB 95 (2,2',3,5',6-pentachlorobiphenyl), PCB 132 (2,2',3,3',4,6'-hexachlorobiphenyl), PCB 136 (2,2',3,3',6,6'-hexachlorobiphenyl), and PCB 149 (2,2',3,4',5',6-hexachlorobiphenyl) to OH-PCBs in liver tissue slices prepared from female mice. The metabolite profile of PCB 136 typically followed the rank order 5-OH-PCB > 4-OH-PCB > 4,5-OH-PCB, and metabolite levels increased with PCB concentration and incubation time. A similar OH-PCB profile was observed with the other PCB congeners, with 5-OH-PCB/4-OH-PCB ratios ranging from 2 to 12. More 5-OH-PCB 136 was formed in liver tissue slices obtained from animals pretreated with phenobarbital (P450 2B inducer) or, to a lesser extent, dexamethasone (P450 2B and 3A enzyme inducer) compared to tissue slices prepared from vehicle-pretreated animals. The apparent rate of 5-OH-PCBs formation followed the approximate rank order PCB 149 > PCB 91 > PCB 132 ∼ PCB 136 > PCB 95. Atropselective gas chromatography revealed a congener-specific atropisomeric enrichment of major OH-PCB metabolites. Comparison of our results with published OH-PCB patterns and chiral signatures (i.e., the direction and extent of the atropisomeric enrichment) from rat liver microsomal revealed drastic differences between both species, especially following the induction of P450 2B enzymes. These species differences in the metabolism of chiral PCBs should be considered in developmental neurotoxicity studies of PCBs.

Figure 1

Simplified metabolism scheme showing putative hydroxylated metabolites (OH-PCBs) of chiral PCBs formed by mouse liver tissue slices and the OH-PCB abbreviations used in this study. Metabolites in parentheses were not detected or below the detection limit in the GC-ECD analysis (see Table S5). ^$ OH-PCBs were tentatively identified by GC-MS analysis.

Figure 2

Optimization of PCB metabolism experiments using liver tissue slices prepared from female mice suggests that incubations for 4 hour with 50 µM PCB provides the highest OH-PCB metabolite levels while maintaining tissue slice viability. A) Uptake of PCB 136 by liver tissue slices prepared from phenobarbital (PB)-pretreated mice over a 2 h incubation period; B) medium to slices ratio of PCB 136 in 2 h incubations with tissue slices prepared from PB-pretreated mice; C) concentration-dependent oxidation of PCB 136 in 2 h incubations with tissue slices prepared from PB-pretreated mice; D) oxidation of PCB 136 in 2 h incubations using tissue slices from saline-, PB-, corn oil (CO) and dexamethasone (DEX)-pretreated mice (50 µM PCB 136); E) rate of formation of OH-PCB 136 metabolites in slices obtained from PB-pretreated mice (50 µM PCB 136) at 2 to 6 h; F) time-dependent release of LDH from tissue slices obtained from PB-pretreated mice into medium after PCB treatment (50 µM PCB 136). As indicated by the dotted line, tissue slices were considered viable if the LDH release was below 30%. All values were mean ± standard error of mean (3 to 5 mouse livers for each treatment).

Figure 3

The levels of PCBs 91, 95, 132, 136 and 149 and their hydroxylated metabolites detected in liver slices prepared from PB-pretreated mice. The major metabolite of all PCB congeners investigated had the hydroxyl group in 5-position of the 2,3,6-trichloro-substituted phenyl ring. Incubations of the parent PCBs with liver tissue slices from PB-pretreated mice were performed at least in triplicate for 4 h at 50 µM PCB. PCBs and OH-PCBs were extracted and qualified by gas chromatography as described under Experimental Procedure. LOD: Limit of detection; NA: not applicable (no standards); ND: Not detected. The values are means ± standard error of mean.

Figure 4

The relative enantiomeric fractions (EF') of PCBs 91, 95, 132, 136 and 149 and their hydroxylated metabolites from mouse liver slice incubations. The EF' values were calculated using the formula EF' = EF_sample-EF_{racemic standard} as sescribed under Experimental Procedure. Atropisomers were separated on a BDM column for PCB 91, 4–91, 5–91, PCB 95, 5'–132, PCB 149 and 5–149; on a Chirasil-Dex column for 5–95, PCB 132, and 5–136; on a Cyclosil-B column for PCB 136 and 4–136 as described under Experimental Procedure. The values are mean ± standard deviation.