Meta‐analysis of Bioaccumulation Data for Nondissolvable Engineered Nanomaterials in Freshwater Aquatic Organisms

Abstract

Understanding the bioaccumulation of engineered nanomaterials (ENMs) is essential for making regulatory decisions on potential environmental risks. Research in the field of ENM bioaccumulation has increased in recent years, but the compilation and statistical analysis of the available experimental data have not been updated. We therefore performed a meta-analysis of the existing literature on the bioaccumulation of eight types of nondissolvable ENMs (titanium dioxide [TiO 2 ], aluminum oxide [Al 2 O 3 ], gold [Au], fullerene [C 60 ], carbon nanotubes, iron oxide [FeO x ], graphene, and polystyrene) in nonmammalian freshwater aquatic organisms across three trophic levels including phytoplankton, zooplankton, and fish. Three typical endpoints were used to assess the bioaccumulation potential: the bioconcentration factor (BCF), the bioaccumulation factor (BAF), and the biomagnification factor (BMF). Our results suggest that zooplankton has greater mean logarithmic BCF and BAF values than phytoplankton (3.31 vs. 1.42) and fish (2.04). The ENMs are biomagnified in zooplankton, with a mean BMF of 17.4, whereas trophic transfer from primary consumers (zooplankton) to secondary consumers (fish) was not observed (mean BMF of 0.13). No clear dependency was identified between the physicochemical characteristics of ENMs (e.g., primary particle size, zeta potential, or shape) and bioaccumulation, except for coated versus uncoated particles accumulated in phytoplankton. Carbonaceous ENMs were found to be more bioaccumulated than the other ENMs we considered, except for TiO 2 . A meta-analysis of bioaccumulation data can (1) deepen the understanding of bioconcentration, bioaccumulation, and biomagnification of ENMs, (2) be used to support grouping strategies as a basis for a safer-by-design approach for ENMs, (3) be integrated into comprehensive hazard and risk assessments, (4) promote the standardization of testing guidelines, and (5) enhance future kinetic bioaccumulation modeling.

Nanoparticles in bodily tissues: predicting their equilibrium distributions

“…γ is a function of surface morphology (shape), 155,208 but applicability to non-spherical NPs remains speculation, needing further study. Long-term exposures to various NPs/coatings (PbO, TiO 2 , QDs, C 60 , citrate 61,209–214 ) show distributions similar to Fig. 2 and values, e.g.…”

“…40,80 This unifies descriptions for NPs and small organic molecules: 218 both γ for NPs and charged surface areas of small molecules drive their partitioning; both find use in risk assessment. 214,219,220…”

“…For NP surfaces with limited/uniform polarizability, we can obtain γ LW NP , γ + NP and γ -NP from partial charged surface area/density [40,80]. This unifies descriptions for NPs and small organic molecules [218]: both γ for NPs and charged surface areas of small molecules drive their partitioning; both find use in risk assessment [214,219,220] .…”

“…Our calculus for various NP types successfully yields partitioning in(to) many tissues/organs, by cells [40,81] and biomolecules. These data find use in PBPK modeling with extensions via scaling [201,214,221]. Existing PBPK models parametrize partitioning without (much) regard for mechanism [222]: higher K in nonphagocytotic organ tissues [223].…”

“…40,80 This unifies descriptions for NPs and small organic molecules: 218 both γ for NPs and charged surface areas of small molecules drive their partitioning; both find use in risk assessment. 214,219,220 Our calculus for various NP types successfully yields partitioning in(to) many tissues/organs, by cells 40,81 and biomolecules. These data find use in PBPK modeling with extensions via scaling.…”