A comparison of the in vitro cyto- and neurotoxicity of brominated and halogen-free flame retardants : prioritization in search for safe(r) alternatives

Brominated flame retardants (BFRs) are abundant persistent organic pollutants with well-studied toxicity. The toxicological and ecological concerns associated with BFRs argue for replacement by safe(r) alternatives. Though previous research identified the nervous system as a sensitive target organ for BFRs, the (neuro) toxic potential of alternative halogen-free flame retardants (HFFRs) is largely unknown. We therefore investigated the in vitro (neuro) toxicity of 13 HFFRs and three BFRs in dopaminergic pheochromocytoma (PC12) and neuroblastoma (B35) cells by assessing several cytotoxic and neurotoxic endpoints. Effects on cell viability and production of reactive oxygen species (ROS) were measured using a combined Alamar Blue and Neutral Red assay and a H2-DCFDA assay, respectively, whereas effects on calcium homeostasis were measured using single-cell fluorescent Ca(2+)-imaging. The majority of the tested flame retardants induced negligible cytotoxicity, except zinc hydroxystannate (ZHS) and zinc stannate (ZS). A considerable fraction of flame retardants affected ROS production (decabromodiphenyl ether (BDE-209), triphenylphosphate (TPP), aluminium trihydroxide (ATH), ammonium polyphosphate (APP), magnesium hydroxide (MHO), ZHS, ZS and melamine polyphosphate (MPP)). Interestingly, ATH, ZHS, ZS and montmorillonite (MMT) increased the basal intracellular calcium concentration ([Ca(2+)]i), whereas tetrabromobisphenol A (TBBPA), resorcinol bis (diphenylphosphate) (RDP), TPP, 9,10-dihydro-9-oxa-10-phosphaphenanthrene-10-oxide (DOPO), ATH, ZHS, ZS and MMT reduced depolarization-evoked increases in [Ca(2+)]i as a result of inhibition of voltage-gated calcium channels. These combined data on the in vitro (neuro) toxicity of HFFRs in comparison with BFRs are essential for prioritization of safe(r) flame retardants. Though additional data are required for a complete (toxic) risk assessment, our data demonstrate that several HFFRs could be suitable substitutes for BFRs