In vitro profiling of toxic effects of prominent environmental lower-chlorinated PCB congeners linked with endocrine disruption and tumor promotion

Abstract

The mechanisms contributing to toxic effects of airborne lower-chlorinated PCB congeners (LC-PCBs) remain poorly characterized. We evaluated in vitro toxicities of environmental LC-PCBs found in both indoor and outdoor air (PCB 4, 8, 11, 18, 28 and 31), and selected hydroxylated metabolites of PCB 8, 11 and 18, using reporter gene assays, as well as other functional cellular bioassays. We focused on processes linked with endocrine disruption, tumor promotion and/or regulation of transcription factors controlling metabolism of both endogenous compounds and xenobiotics. The tested LC-PCBs were found to be mostly efficient anti-androgenic (within nanomolar - micromolar range) and estrogenic (at micromolar concentrations) compounds, as well as inhibitors of gap junctional intercellular communication (GJIC) at micromolar concentrations. PCB 8, 28 and 31 were found to partially inhibit the aryl hydrocarbon receptor (AhR)-mediated activity. The tested LC-PCBs were also partial constitutive androstane receptor (CAR) and pregnane X receptor (PXR) agonists, with PCB 4, 8 and 18 being the most active compounds. They were inactive towards other nuclear receptors, such as vitamin D receptor, thyroid receptor α, glucocorticoid receptor or peroxisome proliferator-activated receptor γ. We found that only PCB 8 contributed to generation of oxidative stress, while all tested LC-PCBs induced arachidonic acid release (albeit without further modulations of arachidonic acid metabolism) in human lung epithelial cells. Importantly, estrogenic effects of hydroxylated (OH-PCB) metabolites of LC-PCBs (4-OH-PCB 8, 4-OH-PCB 11 and 4'-OH-PCB 18) were higher than those of the parent PCBs, while their other toxic effects were only slightly altered or suppressed. This suggested that metabolism may alter toxicity profiles of LC-PCBs in a receptor-specific manner. In summary, anti-androgenic and estrogenic activities, acute inhibition of GJIC and suppression of the AhR-mediated activity were found to be the most relevant modes of action of airborne LC-PCBs, although they partially affected also additional cellular targets.

Keywords: Airborne polychlorinated biphenyls; Endocrine disruption; HydroxyLated PCBs; Metabolism of xenobiotics; Tumor promotion.

Figure 1

Anti-AhR-, CAR- and PXR-mediated activities of LC-PCBs in human liver cell models. A) Anti-AhR activities determined in AZ-AhR cells co-exposed to test compounds and 4.5 nM TCDD for 24 h. The results, expressed relative to TCCD-induced luciferase activity, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to TCDD-treated cells. B) and C) CAR- and PXR-mediated activities determined as induction of CYP2B6 and CYP3A4 mRNAs, respectively, in HepaRG cells exposed to LC-PCBs or reference inducers, PB and RIF, for 24 h. The results, expressed relative to maximum PB-induced CYP2B6 and RIF-induced CYP3A4 mRNA levels, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to vehicle (DMSO).

Figure 2

Anti-androgenic and estrogenic effects of LC-PCBs. A) Anti-androgenic activities determined in AIZ-AR cell line co-exposed to LC-PCBs and 2 nM DHT for 24 h. CPTA was used as a control compound suppressing AR activity. The results, expressed relative to DHT-induced luciferase activity, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to DHT-treated cells. B) Estrogenic activities determined in T47D.luc cells exposed to LC-PCBs or E2, as a reference compound, for 24 h. The results, expressed relative to maximum E2-induced luciferase activity, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to DMSO-treated cells. C) Induction of cell cycle progression in MCF-7 cells by LC-PCBs. Synchronized MCF-7 cells were exposed to test compounds or vehicle (DMSO) for 24 h and cell cycle was analyzed by flow cytometry. Percentage of cells in S-phase was expressed relative to DMSO-treated (control) cells. The results represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to DMSO-treated cells.

Figure 3

Effects of LC-PCBs on induction of oxidative stress, release of AA and inhibition of GJIC. A) Generation of ROS and formation of 8-OHdG in human lung A549 cells. Cells were exposed to LC-PCBs (25 μM) for 24 h or to indicated reference oxidative stress inducers, collected and analyzed by flow cytometry and LC/MS-MS, respectively. The results, expressed relative to control, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to DMSO-treated cells. B) Levels of AA release in culture medium collected from A549 cells exposed to LC-PCBs (25 μM) or AA (10 μM) for 24 h. The results, expressed relative to control, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to DMSO-treated cells. C) Inhibition of GJIC in rat liver epithelial WB-F344 cells. Cells were exposed to LC-PCBs, fluoranthene (Fla; 100 μM; positive control) or vehicle (DMSO) for 1 h, and GJIC was determined using SL-DT assay. The results, expressed relative to DMSO, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to DMSO-treated cells.

Figure 4

Anti-AhR, anti-androgenic or estrogenic effects of hydroxylated LC-PCBs, and their impact on GJIC. A) Anti-AhR activities were determined in AZ-AhR cells as in Fig. 1A. The results, expressed relative to TCCD-induced luciferase activity, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to TCDD-treated cells. B) Anti-androgenic activities were determined in AIZ-AR cells as in Fig. 2A. The results, expressed relative to DHT-induced luciferase activity, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to DHT-treated cells. C) Estrogenic activities were determined in T47D.luc cells as in Fig. 2B. The results, expressed relative to maximum E2-induced luciferase activity, represent means ± S.D. of three independent experiments performed in triplicates. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to DMSO-treated cells. D) Inhibition of GJIC in rat liver epithelial WB-F344 cells was determined as in Fig. 3C. The results, expressed relative to DMSO, represent means ± S.D. of three independent experiments. Symbols ‘*’ and ‘**’ denote significant difference (p < 0.05 and p < 0.01, respectively), as compared to DMSO-treated cells.