Polybrominated diphenyl ethers induce developmental neurotoxicity in a human in vitro model: evidence for endocrine disruption

Abstract

Background: Polybrominated diphenyl ethers (PBDEs) are persistent and bioaccumulative flame retardants, which are found in rising concentrations in human tissues. They are of concern for human health because animal studies have shown that they possess the potential to be developmentally neurotoxic.

Objective: Because there is little knowledge of the effects of PBDEs on human brain cells, we investigated their toxic potential for human neural development in vitro. Moreover, we studied the involvement of thyroid hormone (TH) disruption in the effects caused by PBDEs.

Methods: We used the two PBDE congeners BDE-47 and BDE-99 (0.1-10 microM), which are most prominent in human tissues. As a model of neural development, we employed primary fetal human neural progenitor cells (hNPCs), which are cultured as neurospheres and mimic basic processes of brain development in vitro: proliferation, migration, and differentiation.

Results: PBDEs do not disturb hNPC proliferation but decrease migration distance of hNPCs. Moreover, they cause a reduction of differentiation into neurons and oligodendrocytes. Simultaneous exposure with the TH receptor (THR) agonist triiodothyronine rescues these effects on migration and differentiation, whereas the THR antagonist NH-3 does not exert an additive effect.

Conclusion: PBDEs disturb development of hNPCs in vitro via endocrine disruption of cellular TH signaling at concentrations that might be of relevance for human exposure.

Figure 1

PBDEs inhibit migration of neural progenitor cells: phase-contrast images (A) and quantification (B) of cell migration. Migration distance was measured at four defined spots from the edge of the sphere to the farthest migrated cell after 48 hr. All data are mean ± SE of three independent experiments (five spheres/experiment). Bars, 200 μm. *p ≤ 0.05.

Figure 2

PBDEs inhibit differentiation of hNPCs. (A) Representative photomicrographs of hNPCs after 7 days of differentiation. Cells were stained with antibodies against β(III)tubulin [β(III)Tub⁺] for neurons and O4⁺ for oligodendrocytes. Cell nuclei were counterstained with Hoechst. Bars, 50 μm. (B) Quantification of immunostaining after PBDE treatment. All data are mean ± SE of three independent experiments (five spheres/experiment). *p ≤ 0.05.

Figure 3

PBDEs disrupt cellular TH signaling. (A) hNPC migrated for 48 hr in the presence of the indicated substances, and migration distance was quantified. (B) hNPCs differentiated for 7 days in the presence of the indicated substances. Proliferating neurospheres (Prolif nsph) were used as positive control. Real-time PCR analyses for nestin were quantified with a product-specific copy number standard and normalized for β-actin expression. All data (% DMSO control) are shown as mean ± SE of three independent experiments (five spheres/experiment). *p ≤ 0.05 versus control; ^#p ≤ 0.05 versus respective PBDE treatment.

Figure 4

Long-term PBDE exposure does not interfere with calcium signaling. Neurospheres were incubated with 10 μM BDE-47 or BDE-99 for 7 days under proliferating conditions and for an additional day during differentiation. Afterward, hNPCs were loaded with the fura-2 dye and puff-exposed to 1 mM ATP (A,B) or 500 μM ACh (C,D). After excitation, the ratio of fluorescence emission (F₃₅₇/F₃₈₀) in regions of interest positioned around cell somata was calculated. Any change in normalized ratios (F₃₅₇/F₃₈₀) ≥ 1.2 was considered as an increase and used for further analysis. All data are mean ± SE of three independent experiments (five spheres/experiment).