Development of an in vitro digestion model for estimating the bioaccessibility of soil contaminants.

Soil ingestion can be a major route of human exposure to many immobile soil contaminants. The present risk assessment is based on toxicity studies in which contaminants are typically ingested in liquid or food matrices. The difference in bioavailability of contaminants ingested in a soil matrix is not taken into account. To become bioavailable, contaminants first need to become bioaccessible, i.e., they must be mobilized from the soil during digestion. Soil contaminants may be less bioaccessible than contaminants from liquid or food, so that the risks can be overestimated. This article describes the development of an in vitro human digestion model that is physiologically based. It can be used as a tool to assess bioaccessibility. We explain the rationale behind the experimental design of the model. We address the aspects of the simulated compartments of the gastrointestinal tract, temperature, soil-to-fluid ratio, ratio of digestive juices, transit times, centrifugation, pH values, mixing, constituents and their concentrations, and bile. The optimized in vitro digestion model was applied in a case study. The bioaccessibility of lead in pottery flakes with glazing was determined and compared to the bioaccessibility of lead in the soil from which the pottery flakes were removed. The data indicate that pottery flake lead is considerably less bioaccessible (0.3 +/- 0.2%) than lead in soil without pottery flakes (42-66% at the same site, and 28-73% at other sites in the same town). Furthermore, bioaccessibility values of lead in soil appear to be less than calculated bioaccessibility values for dietary lead (which are based on the criterion used by the Dutch risk assessment and on literature absorption data). This indicates that accounting for the matrix of ingestion can affect the exposure assessment for lead. The in vitro digestion model is a promising tool for studying the effect of the ingestion matrix on bioaccessibility.