Comparison of in vitro strategies for predicting Dichlorodiphenyltrichloroethane (DDT) and its metabolites bioavailability from soils

Abstract

In vitro strategies have widely been used to assess bioaccessibility of organic pollutants in soils. However, studies for comparing in vitro models with in vivo data are still limited. In this study, Dichlorodiphenyltrichloroethane (DDT) and its metabolites (called as DDTr) bioaccessibility in nine contaminated soils were measured using physiologically based extraction test (PBET), in vitro digestion model (IVD), and Deutsches Institut für Normung (DIN) with/without Tenax as an absorptive sink, and DDTr bioavailability was assessed using an in vivo mouse model. Whether or not Tenax was added, DDTr bioaccessibility significantly varied among three methods, suggesting that DDTr bioaccessibility depended on the in vitro method employed. Multiple linear regression analysis indicated that sink, intestinal incubation time and bile content are identified to be the dominant factors in controlling DDTr bioaccessibility. Comparison of in vitro and in vivo results demonstrated that DIN assay with Tenax (TI-DIN) provided the best prediction for DDTr bioavailability (r2 = 0.66, slope=0.78). After extending intestinal incubation time to 6 h or increasing bile content to 4.5 g/L (same to DIN assay) of the TI-PBET and TI-IVD assays, the in vivo-in vitro correlation will improved significantly, with r2 = 0.76 and slope= 1.4 for TI-PBET and r2 = 0.84 and slope= 1.9 for TI-IVD under 6 h intestinal incubation, and r2 = 0.59 and slope= 0.96 for TI-PBET and r2 = 0.51 and slope= 1.0 for TI-IVD under 4.5 g/L of bile content. The results suggest that it is essential to understand these key factors influencing bioaccessibility for the development of standardized in vitro methods, which helps to refine the risk assessment of human exposure to contaminants via soil ingestion.