Determinants of oral bioavailability of soil-borne contaminants

Abstract

Children ingest soil, either accidentally via hand-to-mouth behavior or deliberately. In this manner, a child ingests on average between 50 and 200 mg soil/day, although amounts of as much as 60 g/day have also been observed. Hence, soil ingestion can be a main route of exposure to soil-borne contaminants to children. To estimate the health risk associated to this exposure route, and to assess intervention values for contaminants in soils, one needs to know the oral bioavailability of the soil-borne contaminant. The objective of this thesis was to gain insight into the determinants of oral bioavailability of soil-borne polychlorinated biphenyls (PCBs), lindane and lead. Oral bioavailability of soil-borne contaminants is defined as the contaminant fraction that reaches the systemic circulation. Before a soil-borne contaminant becomes bioavailable it has to go through four steps: 1) soil ingestion, 2) mobilization from soil during digestion, i.e. bioaccessibility, 3) intestinal absorption of the bioaccessible contaminant, and 4) first-pass effect. In the present thesis, mobilization from soil, distribution among different physicochemical contaminant forms in small intestinal fluid, i.e. chyme, and intestinal absorption, have been studied. To that end, we employed a physiologically based in vitro digestion model that mimicked the gastro-intestinal digestion. In vitro differentiated intestinal cells were employed to similate intestinal absorption. After ingestion of soil-borne PCBs, lindane and lead, the freely dissolved concentration in chyme can assumed to be small. The freely dissolved fraction is the contaminant fraction that is at least available for intestinal absorption. Expected main fractions for the hydrophobic organic compounds (HOCs) are PCBs and lindane sorbed to bile salt micelles, digestive proteins and soil, whereas lead phosphate, lead bile and lead soil complexes are main fractions for lead. The distribution of the presently used HOCs among sorbing constituents in chyme was based on partitioning. The in vitro experiments indicate that not all ingested and soil-borne PCBs, lindane and lead are mobilized from soil during digestion, i.e. become bioaccessible, and therefore do not reach the systemic circulation and become bioavailable. Furthermore, not all bioaccessible lead is expected to be absorbed, causing a further reduction of the lead fraction that becomes bioavailable relative to the ingested amount. Nevertheless, a considerable fraction of the contaminants (several tens of %) appears to be able to become bioavailable, stressing the importance of exposure to contaminants via soil ingestion. Physicochemical conditions in the gastro-intestinal tract are expected to affect the bioaccessibility and thereby the oral bioavailability of the soil-borne contaminants. The physiological worst case situation for the PCBs and lindane is most likely the fed state, since that results in high concentrations of sorbing constituents in the gastro-intestinal tract. The worst case situation for lead is most likely the fasted state so that the gastric pH is low, inducing a high mobilization from soil. The PCBs, lindane and lead have in common that more than the freely dissolved fraction is transported across the intestinal membrane. The freely dissolved concentration can be considered the driving force for the contaminant flux towards the intestinal cells, while the labile contaminants represent the pool of contaminants that may dissociate and contribute to the flux and thus to oral bioavailability. Chemical sampling techniques such as Solid Phase Microextraction (SPME) and voltammetric techniques offer possibilities to investigate mass transfer of contaminants in a complex medium towards biological phases like Caco-2 cells.