The impact of synthetic amorphous silica (E 551) on differentiated Caco-2 cells, a model for the human intestinal epithelium.

For several decades, food-grade synthetic amorphous silica (SAS) have been used as a technological additive to reduce caking of food powders. Human exposure is thus inevitable and safety concerns are taken seriously. The toxicity of silica in general and SAS in particular has been studied extensively. Overall, there is little evidence that food-grade SAS pose any health risks to humans. However, from the available data it was often not clear which type of silica was used. Accordingly, the latest report of the European food safety authority requested additional toxicity data for well-characterised "real food-grade SAS". To close this gap, we screened a panel of ten well-defined, food-grade SAS for potential adverse effects on differentiated Caco-2 cells. Precipitated and fumed SAS with low, intermediate and high specific surface area were included to determine structure-activity relationships. In a physiological dose-range up to 50 µg/ml and 48 h of incubation, none of the materials induced adverse effects on differentiated Caco-2 cells. This held true for endpoints of acute cytotoxicity as well as epithelial specific measures of barrier integrity. These results showed that despite considerable differences in production routes and material characteristics, food-relevant SAS did not elicit acute toxicity responses in intestinal epithelial cells.