OBJECTIVETo develop and validate an in vitro digestion model for assessing the bioaccessibilities of some important mycotoxins of aflatoxin B group (aflatoxin B(1) and aflatoxin B(2), AFB(1) and AFB(2)).

METHODSUsing simulating gastrointestinal physiological digestion process, the effects of digestion time (long, medium and short), the fasting and feeding status (fasting, between fasting and semi-feeding, semi-feeding, between semi-feeding and feeding, feeding states), the volume and pH (high, medium and low) of digestive solution, as well as other food ingredients ingested along with aflatoxin B group from mixed foods on bioaccessiblities of AFB(1) and AFB(2) in the mouth, stomach and small intestine were studied. The optimal technical parameters of the model were identified and the model was validated with mycotoxin adsorbents.

RESULTSThe optimal conditions of AFB(1) releasing from the ingested foods at the highest concentration in gastrointestinal tract were as follows: digestion time of 6 min, 1.5 h and 2.5 h in mouth, stomach and duodenum, respectively; the optimal pH values of 1.1 and 7.5 for gastric juice and duodenal fluid; the volume of 7, 13, 12 and 6 ml for saliva, gastric juice, intestinal fluid and bile, respectively; the optimal conditions of AFB(2) releasing from the ingested foods at the highest concentration in gastrointestinal tract were as follows: digestion time of 6 min, 2.5 h and 2.5 h in mouth, stomach and duodenum, respectively; the optimal pH values of 1.1 and 7.8 for gastric juice and duodenal fluid; the volume of 5, 12, 13 and 6 ml for saliva, gastric juice, intestinal fluid and bile, respectively. The bioaccessibilities of both AFB(1) and AFB(2) were highest at the fasting state (83.1% and 89.3% respectively). The bioaccessibilities decreased with the increasing of stomach contents, but the changes in bioaccessibility were not significant when the stomach contents reached the semi-feeding state or more. From semi-feeding to feeding state, the biocessibilities of AFB(1) decreased from 72.8% to 71.5% and AFB(2) decreased from 78.3% to 76.9%. Chlorophyll and activated charcoal were the strongest absorbent in reducing the bioaccessibilities of AFB(1) and AFB(2), and the bioaccessibilities decreased to 0.8% and 1.3% respectively.

CONCLUSIONThe in vitro digestion model developed in the present study is stable and reproducible, and meets the requirements for assessing the bioaccessibilities of AFB(1) and AFB(2) in foods.