Artificial Sweeteners Disrupt Tight Junctions and Barrier Function in the Intestinal Epithelium through Activation of the Sweet Taste Receptor, T1R3

Abstract

The breakdown of the intestinal epithelial barrier and subsequent increase in intestinal permeability can lead to systemic inflammatory diseases and multiple-organ failure. Nutrition impacts the intestinal barrier, with dietary components such as gluten increasing permeability. Artificial sweeteners are increasingly consumed by the general public in a range of foods and drinks. The sweet taste receptor (T1R3) is activated by artificial sweeteners and has been identified in the intestine to play a role in incretin release and glucose transport; however, T1R3 has not been previously linked to intestinal permeability. Here, the intestinal epithelial cell line, Caco-2, was used to study the effect of commonly-consumed artificial sweeteners, sucralose, aspartame and saccharin, on permeability. At high concentrations, aspartame and saccharin were found to induce apoptosis and cell death in intestinal epithelial cells, while at low concentrations, sucralose and aspartame increased epithelial barrier permeability and down-regulated claudin 3 at the cell surface. T1R3 knockdown was found to attenuate these effects of artificial sweeteners. Aspartame induced reactive oxygen species (ROS) production to cause permeability and claudin 3 internalization, while sweetener-induced permeability and oxidative stress was rescued by the overexpression of claudin 3. Taken together, our findings demonstrate that the artificial sweeteners sucralose, aspartame, and saccharin exert a range of negative effects on the intestinal epithelium through the sweet taste receptor T1R3.

Keywords: Caco-2; claudin; intestinal epithelium; permeability; sweeteners.

Figure 1

High physiological concentrations of artificial sweeteners decreased the viability and increase apoptosis of Caco2 cells through the sweet taste receptor (T1R3). (a) Caco-2 cell viability was measured using the Cell Counting Kit-8 (CCK8) assay following 24 h exposure to the artificial sweeteners sucralose, saccharin, and aspartame at concentrations ranging from 0.01 to 10,000 µM. Absorbance was normalized to a 0 µM control and expressed as percentage viability. n = 8. (b,c) Caco-2 cell viability was measured as for (a) following the siRNA knockdown of T1R3 for 24 h and exposure to sucralose, saccharin, and aspartame (10 mM) for a further 24 h. A representative blot of T1R3 and load control actin were shown to confirm siRNA knockdown using 50 µg of protein (b). Caco-2 cells were collected following exposure to sucralose, saccharin, and aspartame at concentrations ranging from 0.1 to 1000 µM for 24 h and analyzed by flow cytometry. Cell death (d) and apoptosis (e) were measured as propidium iodide and annexin V-positive and annexin V-positive cells, respectively. n = 5–6. Data are expressed as mean ± S.E.M. * p < 0.05 versus vehicle (0 µM).

Figure 2

Low physiological concentrations of artificial sweeteners sucralose and aspartame disrupt the intestinal epithelial barrier through the sweet taste receptor. (a) The permeability of the epithelial monolayer was measured with an FITC-dextran assay, following exposure to sucralose, saccharin, and aspartame (0.1 mM) for 24 h using lipopolysaccharide (LPS) (1 µg/mL) as a positive control. (b) The permeability of the Caco-2 cell monolayer was measured with an FITC-dextran assay following the siRNA knockdown of T1R3 for 24 h and exposure to sucralose, saccharin, and aspartame (0.1 mM) for a further 24 h. % permeability was calculated normalized to vehicle treatment. n = 6. Data are expressed as mean ± S.E.M. * p < 0.05 versus vehicle (0 µM).

Figure 3

Claudin mRNA expression profile in the intestinal epithelium in vitro and in vivo. Claudin mRNA transcripts were profiled in the small and large intestine and in cultured Caco-2 cells. (a) Three segments of the murine small intestine, the duodenum (stomach to ligament of Treitz), the jejunum (ligament of Treitz until mid-small intestine), the ileum (remaining half of small intestine until the caecum), and two segments of the large intestine (proximal (first half of the colon) and distal colon (second half of the colon)) were collected for RT-PCR analysis. (b) Untreated Caco-2 cells were collected for RT-PCR analysis. The relative ratio was calculated as claudin compared to actin mRNA expression levels. Data are expressed as the mean ± S.E.M of PCR reactions. n = 6.

Figure 4

Sucralose and aspartame modulate claudin 3 and 15 expression in intestinal epithelial cells through T1R3. (a–e) The cell surface expression of claudins 2, 3, 4, 7, and 15, respectively, was determined, with whole-cell indirect ELISA using chemiluminescence, in Caco-2 cells exposed to sucralose or aspartame (0.1 mM) or LPS (1 µg/mL) for 24 h. (f,g) The cell surface expression of claudins 3 and 15, respectively, was determined in Caco-2 cells following siRNA knockdown of T1R3 for 24 h and exposure to sucralose or aspartame (0.1 mM) for a further 24 h. n = 6. Data are expressed as mean ± S.E.M. * p < 0.05 versus vehicle (0 µM).

Figure 5

Overexpression of claudin 3 rescues sweetener-induced barrier leakage across the intestinal epithelium. Caco-2 cells were transiently transfected with cDNA encoding wild-type claudin 3 (CLDN3-DYK) or the vector control (DYK), or untransfected (UT) for 48 h. (a) The total protein levels of claudin 3 in cells were measured by the Western blot analysis of cell lysates using 50 µg protein. (b) The cell surface expression of claudin 3 was determined by whole-cell indirect ELISA using chemiluminescence in transfected Caco-2 cells. (c) The viability of transfected Caco-2 cells was assessed by a CCK8 assay. Absorbance values were normalized to untransfected control and expressed as percentage viability. (d) The permeability of the transfected Caco-2 cell monolayer was measured by an FITC-dextran assay following exposure to sucralose and aspartame (0.1 mM) for 24 h. % permeability was calculated normalized to vehicle treatment. n = 6. Data are expressed as mean ± S.E.M. * p < 0.05 versus untransfected cells or vehicle (0 µM).

Figure 6

Aspartame, but not sucralose, increases oxidative stress in intestinal epithelial cells linked to barrier leakage. Reactive oxygen species (ROS) production in Caco-2 cells was measured by fluorescence of DCFDA following exposure to sucralose or aspartame (0.1 mM) or LPS (1 µg/mL) as a positive control (a) in the presence and absence of the anti-oxidant N-acetyl cysteine (NAC; 1 mM) (b) or following siRNA knockdown of T1R3 (c). (d) The permeability of the Caco-2 cell monolayer was measured by an FITC-dextran assay, following exposure to sucralose and aspartame (0.1 mM) for 24 h in the presence and absence of NAC. % permeability was calculated normalized to vehicle treatment. (e) The cell surface expression of claudin 3 was determined by whole-cell indirect ELISA using chemiluminescence in Caco-2 cells exposed to sucralose or aspartame (0.1 mM) in the presence and absence of NAC. (f) ROS production was measured by fluorescence of DCFDA in Caco-2 cells transiently transfected with CLDN3-DYK or control vector, exposed to aspartame (0.1 mM). % permeability was calculated normalized to vehicle treatment. ROS production was calculated as % normalized to 0 µM. n = 5–6. Data are expressed as mean ± S.E.M. * p < 0.05 versus vehicle (0 µM); # p < 0.05 versus vehicle for NAC.