Microplastics, small pieces of plastic derived from polystyrene, have recently become an ecological hazard due to their toxicity and widespread occurrence in aquatic ecosystems. In this study, we exposed zebrafish larvae to two types of fluorescent polystyrene nanoparticles (PS-NPs) to identify their size-dependent effects. PS-NPs of 50 nm, unlike 100 nm PS-NPs, were found to circulate in the blood vessels and accumulate in the brains of zebrafish larvae. Behavioral and electroencephalogram (EEG) analysis showed that 50 nm PS-NPs induce abnormal behavioral patterns and changes in EEG power spectral densities in zebrafish larvae. In addition, the quantification of endogenous neurochemicals in zebrafish larvae showed that 50 nm PS-NPs disturb dopaminergic metabolites, whereas 100 nm PS-NPs do not. Finally, we assessed the effect of PS-NPs on the permeability of the blood–brain barrier (BBB) using a microfluidic system. The results revealed that 50 nm PS-NPs have high BBB penetration compared with 100 nm PS-NPs. Taken together, we concluded that small nanoparticles disturb the nervous system, especially dopaminergic metabolites.

The blood-brain-barrier (BBB) is made up of blood vessels whose permeability enables the passage of some compounds. A predictive model of BBB permeability is important in the early stages of drug development. The predicted BBB permeabilities of drugs have been confirmed using a variety of in vitro methods to reduce the quantities of drug candidates needed in preclinical and clinical trials. Most prior studies have relied on animal or cell-culture models, which do not fully recapitulate the human BBB. The development of microfluidic models of human-derived BBB cells could address this issue. We analyzed a model for predicting BBB permeability using the Emulate BBB-on-a-chip machine. Ten compounds were evaluated, and their permeabilities were estimated. Our study demonstrated that the permeability trends of ten compounds in our microfluidic-based system resembled those observed in previous animal and cell-based experiments. Furthermore, we established a general correlation between the partition coefficient ($K_p$) and the apparent permeability ($P_{app}$). In conclusion, we introduced a new paradigm for predicting BBB permeability using microfluidic-based systems.