An ex vivo intestinal absorption model is more effective than an in vitro cell model to characterise absorption of dietary carotenoids following simulated gastrointestinal digestion

To get the most accurate food digestion-related data, and how this affects nutrient absorption, it is critical to carefully simulate human digestion systems using model settings. In this study, the uptake and transepithelial transportation of dietary carotenoids was compared using two different models that have previously been used to assess nutrient availability. The permeability of differentiated Caco-2 cells and murine intestinal tissue were tested using all-trans-β-carotene and lutein prepared in artificial mixed micelles and micellar fraction from orange-fleshed sweet potato (OFSP) gastrointestinal digestion. Transepithelial transport and absorption efficiency were then determined using liquid chromatography tandem-mass spectrometry (LCMS-MS). Results showed that the mean uptake for all-trans-β-carotene in the mouse mucosal tissue was 60.2 ± 3.2% compared to 36.7 ± 2.6% in the Caco-2 cells with the mixed micelles as the test sample. Similarly, the mean uptake was higher in OFSP with 49.4 ± 4.1% following mouse tissue uptake compared to 28.9 ± 4.3% using Caco-2 cells for the same concentration. In relation to the uptake efficiency, the mean percentage uptake for all-trans-β-carotene from artificial mixed micelles was 1.8-fold greater in mouse tissue compared to Caco-2 cells (35.4 ± 1.8% against 19.9 ± 2.6%). Carotenoid uptake reached saturation at 5 µM when assessed with the mouse intestinal cells. These results demonstrate the practicality of employing physiologically relevant models simulating human intestinal absorption processes that compares well with published human in vivo data. When used in combination with the Infogest digestion model, the Ussing chamber model, using murine intestinal tissue, may thus be an efficient predictor of carotenoid bioavailability in simulating human postprandial absorption ex vivo.