Research Article: The flavanone homoeriodictyol increases SGLT-1-mediated glucose uptake but decreases serotonin release in differentiated Caco-2 cells

Date Published: February 13, 2017

Publisher: Public Library of Science

Author(s): Barbara Lieder, Julia Katharina Hoi, Ann-Katrin Holik, Katrin Geissler, Joachim Hans, Barbara Friedl, Kathrin Liszt, Gerhard E. Krammer, Jakob P. Ley, Veronika Somoza, Jeong-Ho Kim.

http://doi.org/10.1371/journal.pone.0171580

Abstract

Flavanoids and related polyphenols, among them hesperitin, have been shown to modulate cellular glucose transport by targeting SGLT-1 and GLUT-2 transport proteins. We aimed to investigate whether homoeriodictyol, which is structurally related to hesperitin, affects glucose uptake in differentiated Caco-2 cells as a model for the intestinal barrier. The results revealed that, in contrast to other polyphenols, the flavanon homoeriodictyol promotes glucose uptake by 29.0 ± 3.83% at a concentration of 100 μM. The glucose uptake stimulating effect was sensitive to phloridzin, but not to phloretin, indicating an involvement of the sodium-coupled glucose transporter SGLT-1, but not of sodium-independent glucose transporters (GLUT). In addition, in contrast to the increased extracellular serotonin levels by stimulation with 500 mM D-(+)-glucose, treatment with 100 μM homoeriodictyol decreased serotonin release by –48.8 ± 7.57% in Caco-2 cells via a phloridzin-sensitive signaling pathway. Extracellular serotonin levels were also reduced by –57.1 ± 5.43% after application of 0.01 μM homoeriodictyol to human neural SH-SY5Y cells. In conclusion, we demonstrate that homoeriodictyol affects both the glucose metabolism and the serotonin system in Caco-2 cells via a SGLT-1-meditated pathway. Furthermore, the results presented here support the usage of Caco-2 cells as a model for peripheral serotonin release. Further investigations may address the value of homoeriodictyol in the treatment of anorexia and malnutrition through the targeting of SGLT-1.

Partial Text

Glucose uptake from the lumen into the epithelial cells of the small intestine is predominantly mediated by the sodium-coupled transporter SGLT-1 [1]. Further transport from the enterocytes to the blood stream is thought to be mediated by the facilitative uniporter glucose transporter 2 (GLUT-2), which exhibits, compared to SGLT-1, a low affinity, but high capacity for glucose [2]. However, in mice [3] and also in human intestinal Caco-2 cells [4], GLUT-2 has been convincingly shown to be expressed not only on the basolateral side, but in the brush-border membranes as well. In addition, more recent studies suggest that at high luminal glucose concentrations, GLUT-2 may be recruited from intracellular vesicles into the apical membrane to support a rapid transport of large quantities of glucose from the lumen to the enterocytes [5]. This recruitment to the brush-border membrane was also demonstrated in Caco-2 cells, demonstrating that this cell line is a suitable model to study intestinal glucose uptake [6, 7].

Several polyphenols have been shown to influence intestinal glucose uptake. However, results are inconsistent since a structure-associated activity has not yet been demonstrated; whereas phloretin, myricetin, and quercetin affect glucose uptake via GLUT transporters, the polyphenols phloridzin and neohesperidin have been shown to decrease sodium-dependent glucose uptake, indicating a SGLT-1-dependent mechanism [13]. Since the polyphenol hesperitin was demonstrated to decrease glucose uptake in two different cell models [14, 15], we hypothesized here that the structural analog homoeriodictyol (HED) affects glucose uptake in a similar manner. Differentiated Caco-2 cells were chosen since these cells have been demonstrated to express the relevant glucose transport systems, such as SGLT-1 at the apical surface [19] and GLUT-2 at the basolateral and apical surface [5]. Full differentiation within 21 days of cultivation applied in this study has been confirmed in earlier studies using the trans-epithelial electrical resistance as a marker [20, 21]. The test compound HED differs from hesperitin only in the position of the residues at the B-ring. However, in contrast to our hypothesis, this slight structural difference led to a major difference in the impact on glucose uptake. Both HED and its sodium salt instead increased glucose uptake in differentiated Caco-2 cells up to 135 ± 8.89% (p = 0.002) and 129 ± 3.83% (p<0.001), respectively, after application of the highest test concentration of 100 μM (Fig 1). These data demonstrate that slight structural differences may have a major impact on the bioactivity of a compound and that not all polyphenols reduce glucose uptake. Since higher concentrations could not be applied in the assay due to HED’s limited solubility in aqueous solutions, no saturation point of HED on glucose uptake is presented here. Comparison of the effects of the HED pre-dissolved in ethanol with the more water soluble sodium salt confirmed that there is no difference in the effect size between the two modes of application (p = 0.47). Therefore, an impact of the sodium ions on glucose uptake can be excluded. In addition, the use of ethanol during incubations was avoided, and the following experiments in Caco-2 cells were carried out using the water-soluble sodium salt of HED. In conclusion, we demonstrate that the polyphenol homoeriodictyol affects both glucose metabolism and the serotonin system in Caco-2 cells via a SGLT-1-meditated pathway. Since stimulation of serotonin levels by glucose is, unlike reduced serotonin levels after HED treatment, not phloridzin-sensitive, a direct link between of glucose uptake and serotonin release via SGLT-1 in Caco-2 cells is not assumed. However, the results presented here support the usage of Caco-2 cells as a model for peripheral serotonin release.   Source: http://doi.org/10.1371/journal.pone.0171580