Size effect and mucus role on the intestinal toxicity of the E551 food additive and engineered silica nanoparticles

Abstract

The E551 food additive is composed of synthetic amorphous silica particles. The current regulation does not mention any specifications regarding their size and granulometric distribution, thus allowing the presence of silica nanoparticles despite their potential toxicity. The digestion process could modify their physicochemical properties and then influence their toxicological profile. After physicochemical characterization, subacute toxicity of engineered silica nanoparticles from 20 to 200 nm, native and digested E551 additives were evaluated from in vitro models of the intestinal barrier. Single cultures and a co-culture of enterocytes and mucus-secreting cells were established to investigate the mucus role. Toxicological endpoints including cytotoxicity, ROS production, intestinal permeability increase, and actin filament disruption were addressed after a 7-day exposure. The results showed a size-dependent effect of silica nanoparticles on cytotoxicity and intestinal permeability. A time-dependent disruption of actin filaments was observed in Caco-2 cells. The mucus layer spread on the HT29-MTX single culture acted as an efficient protective barrier while in the co-culture, small nanoparticles were able to cross it to reach the cells. From a hydrodynamic diameter of 70 nm, nanoparticles were not internalized in the intestinal cells, even in mucus-free models. Digestion did not affect the physicochemical properties of the additive. Due to a mean hydrodynamic diameter close to 200 nm, both native and digested E551 additives did not induce any toxic effect in intestinal barrier models. This study emphasized a cutoff size of 70 nm from which the interactions of the E551 additive with intestinal cells would be limited.

Keywords:

• Silica nanoparticles
• E551
• mucus
• intestinal barrier
• toxicity

Acknowledgments

T. Zaiter was supported by a fellowship from the “Centre Islamique d’Orientation et de l’Enseignement Supérieur (CIOES)”. The authors thank the cell imaging core facility, DImacell, for the confocal microscopy experiments (Plateforme DImaCell, Univ. Bourgogne Franche-Comté, F-25000 Besançon, France). NM and MH thank the European Union through the PO FEDER-FSE Bourgogne 2014/2020 programs and the EIPHI Graduate School (contract ANR-17-EURE-0002).

Disclosure statement

There are no conflicts of interest to declare.