Inhibition of Glucose Transport by Tomatoside A, a Tomato Seed

Jan 22, 2018 - For 13C6-glucose transport experiments, Caco-2 cell monolayers were treated with tomatoside A (5, 10, or 20 μM) in DMEM supplemented w...
0 downloads 11 Views 977KB Size
Download PDF
Article Cite This: J. Agric. Food Chem. 2018, 66, 1428−1434

pubs.acs.org/JAFC

Inhibition of Glucose Transport by Tomatoside A, a Tomato Seed Steroidal Saponin, through the Suppression of GLUT2 Expression in Caco‑2 Cells Baorui Li,† Yusuke Terazono,† Naoto Hirasaki,† Yuki Tatemichi,‡ Emiko Kinoshita,‡ Akio Obata,‡ and Toshiro Matsui*,† †

Department of Bioscience and Biotechnology, Division of Bioresource and Bioenvironmental Sciences, Faculty of Agriculture, Graduated School of Kyushu University, 6-10-1 Hakozaki, Fukuoka 812-8581, Japan ‡ Research & Development Division, Kikkoman Co., 399 Noda, Chiba 278-0037, Japan S Supporting Information *

ABSTRACT: We investigated whether tomatoside A (5α-furostane-3β,22,26-triol-3-[O-β-D-glucopyranosyl (1→2)-β-Dglucopyranosyl (1→4)-β-D-galactopyranoside] 26-O-β-D-glucopyranoside), a tomato seed saponin, may play a role in the regulation of intestinal glucose transport in human intestinal Caco-2 cells. Tomatoside A could not penetrate through Caco-2 cell monolayers, as observed in the transport experiments using liquid chromatography−mass spectrometry. The treatment of cells with 10 μM tomatoside A for 3 h resulted in a 46.0% reduction in glucose transport as compared to untreated cells. Western blotting analyses revealed that tomatoside A significantly (p < 0.05) suppressed the expression of glucose transporter 2 (GLUT2) in Caco-2 cells, while no change in the expression of sodium-dependent glucose transporter 1 was observed. In glucose transport experiments, the reduced glucose transport by tomatoside A was ameliorated by a protein kinase C (PKC) inhibitor and a multidrug resistance-associated protein 2 (MRP2) inhibitor. The tomatoside A-induced reduction in glucose transport was restored in cells treated with apical sodium-dependent bile acid transporter (ASBT) siRNA or an ASBT antagonist. These findings demonstrated for the first time that the nontransportable tomato seed steroidal saponin, tomatoside A, suppressed GLUT2 expression via PKC signaling pathway during the ASBT-influx/MRP2-efflux process in Caco-2 cells. KEYWORDS: tomatoside A, ASBT, glucose transport, GLUT2, MRP2

■

[GIP]) secretion by chlorogenic acid.9 Kwon et al.10 also showed that quercetin inhibited glucose transport in cells overexpressing GLUT2 by mechanisms other than competitive transport inhibition, since quercetin was not transported by glucose transporters. In a series of studies, we have demonstrated that theaflavins, nonabsorbable condensed catechins, suppressed the expression of intestinal peptide transporter 1 (PepT1) via intracellular AMP-dependent kinase (AMPK) activation.11,12 Although the underlying mechanism is questionable, nonabsorbable compounds are thought to exert physiological actions in the intestinal system. Here, we focus on natural saponins bearing steroidal triterpenoid skeleton with sugar moieties as potential nonabsorbable compounds, as no study has ever reported the absorption of saponin-related compounds displaying antidiabetic effect.13,14 As tomatoside A is the most abundant in tomato seeds, we used the saponin (5α-furostane-3β,22,26-triol-3-[O-β-D-glucopyranosyl (1→2)β-D-glucopyranosyl (1→4)-β-D-galactopyranoside] 26-O-β-Dglucopyranoside, Figure 1) in this study. The transportability and antihyperglycemic potential of tomatoside A was investigated using the intestinal transport model cell line Caco-2 monolayers. In addition, we evaluated the signaling

INTRODUCTION Type II diabetes is a noninsulin dependent glycemic disorder associated with impaired insulin response; excessive energy intake due to unhealthy food choice is considered to be one of the reasons for the development of Type II diabetes among other factors, such as genetic predisposition and physical inactivity.1 Chronic hyperglycemic results in a number of complications, including cardiovascular disease, nephropathy, neuropathy, and retinopathy.2 Thus, an appropriate control over the postprandial hyperglycemic level may serve as a possible alternative for the treatment of diabetes or prevention of diabetes-related complications.3 Glucose is incorporated through the apical sodium-dependent glucose transporter 1 (SGLT1), which couples two Na+ with one glucose.4 The subsequent passive epithelial transport of glucose involves a facilitated diffusion by glucose transporter 2 (GLUT2) at the basolateral enterocytes.5 Thus, retardation of intestinal glucose absorption through inhibition of either or both transporters is a rationale approach to modulate postprandial hyperglycemia, along with the inhibition of carbohydrate digestion in the gut through α-amylase or αglucosidase suppression.6,7 Studies have been performed to examine antihyperglycemic effects through the inhibition of carbohydrate digestion by anthocyanins,6 blockade of SGLT1 transport route by green tea polyphenols,8 or promotion of incretin (glucagon-like peptide-1 [GLP-1] and glucose-dependent insulinotropic polypeptide © 2018 American Chemical Society

Received: December 25, 2017 Accepted: January 22, 2018 Published: January 22, 2018 1428

DOI: 10.1021/acs.jafc.7b06078 J. Agric. Food Chem. 2018, 66, 1428−1434

Article

Journal of Agricultural and Food Chemistry

were obtained from BD Biosciences (Franklin Lakes, NJ, USA). Nonessential amino acid mixture was supplied by MP Biomedicals (Irvine, CA, USA) and penicillin by Meiji Seika Co. (Tokyo, Japan). Valspodar, a P-glycoprotein (P-gp) inhibitor, and 13C6-glucose were procured from Cambridge Isotope Laboratories, Inc. (Andover, MA, USA). MK571, a selective multidrug resistance-associated protein 2 (MRP2) inhibitor was purchased from Adipogen Co. (San Diego, CA, USA). Compound C, an AMPK inhibitor, and GF109203X, a nonspecific protein kinase C (PKC) inhibitor, were obtained from Merck (Darmstadt, Germany). sc-106906 (apical sodium-dependent bile acid transporter [ASBT] siRNA) was supplied by Santa Cruz Biotechnology Inc. (Dallas, TX, USA). Other reagents purchased were of analytical grade and used without further purification. Cell Culture. Caco-2 cells were cultured in DMEM containing 10% FBS, 100 U/mL of penicillin, and 100 μg/L streptomycin. Cells were incubated at 37 °C in humidified atmosphere (5% CO2 and 95% O2) and the medium was exchanged every other day until the cells reached 80−90% confluence (5−7 days). Caco-2 cells used in this study were between passages 50 and 60. For transport study, cell culture conditions were maintained as previously described.11 Briefly, Caco-2 cells were seeded at a density of 4 × 105 cells/mL in DMEM supplemented with 0.1% MITO+ serum extender into a BD Falcon cell culture insert (PET membrane, 0.9 cm2, 1.0 μm pore size; BD Bioscience, Tokyo, Japan) coated with type I collagen (collagen gel culturing kit, cell matrix type I-A, Nitta Gelatin, Osaka, Japan) and were mounted in 12-well plates (BD Bioscience). After being seeded, cells were cultured in a DMEM containing 0.1% of MITO+ serum extender for 48 h. Next, the medium was replaced with fresh ENTERO-STIM enterocyte differentiation medium (BD Bioscience) containing 0.1% of MITO+ serum extender every 24 h for 72 h for cell differentiation and for formation of monolayers according to the manufacturer’s protocol. The integrity of the monolayers was evaluated by measuring transepithelial electrical resistance (TEER) using a Millipore Millicell ERS-2 (Miilipore, Billerica, MA, USA) . Caco-2 cell monolayers with TEER values of >400 Ω·cm2 were used for transport experiments as differentiated monolayers. Treatment of Caco-2 Cells with Tomatoside A. For 13C6glucose transport experiments, Caco-2 cell monolayers were treated with tomatoside A (5, 10, or 20 μM) in DMEM supplemented with 10% FBS for 1, 3, 6, or 24 h at 37 °C. Tomatoside A dissolved in 0.5% DMSO (final concentration) was used for this study. Cells incubated in tomatoside A-free medium containing 0.5% DMSO were used as control. For inhibition experiments, Caco-2 cells were treated with or without tomatoside A in the presence or absence of inhibitors (10 μM compound C, 10 μM GF109203X, 20 μM valspodar, 10 μM MK571, or 100 μM fluvastatin) for 3 h before glucose transport experiments. The toxicity of tomatoside A in Caco-2 cells was examined by a cell counting Kit-8 (CCK-8) assay (Dojindo Co., Kumamoto, Japan). Briefly, Caco-2 cells were seeded in 96-well plates at a density of 1 × 105 cells/mL. After 24 h, cells were treated with tomatoside A up to 1 mM and cultured for an additional 3 h. All experiments were conducted in triplicate. Viability determination was evaluated at 450 nm using a Wallac 1420-microplate reader (PerkinElmer Life Science, Tokyo, Japan). Assay for Tomatoside A Transport across Caco-2 Cell Monolayers. Transport of tomatoside A across Caco-2 cell monolayers was evaluated as previously described with minor modifications.11 Briefly, Caco-2 cell monolayers were cultured in BD Falcon cell culture inserts that were mounted in 12-well plates (BD Bioscience). An aliquot (0.5 mL) of Hanks’ balanced salt solution (HBSS) containing 10 mM 2-(N-morpholino)ethanesulfonic acid (MES) (pH 6.0) was added to the apical side. HBSS containing 10 mM N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES) (pH 7.4) was added to the basolateral side. After equilibration at 37 °C with 5% CO2 for 15 min, either the apical or basolateral side was replaced with fresh HBSS containing 1 mM tomatoside A. After 60 min incubation, 300 μL of solution was withdrawn from the apical or basolateral side and subjected to an LC-time-of-flight (TOF)-mass spectrometry (MS).

Figure 1. LC-TOF-MS chromatograms of tomatoside A transported across Caco-2 cell monolayers in apical to basolateral and basolateral to apical directions. The peak of tomatoside A standard (1 mM) corresponds to negative-ESI detection at [M-H]− of 1081.5940 m/z. Transport experiments of 1 mM tomatoside A were performed for 60 min at 37 °C. After the transport experiments, 300 μL of solution was used for LC-TOF-MS analysis from apical or basolateral side. LC-MS conditions are described in the Materials and Methods Section.

pathways involved in tomatoside A-mediated suppression of glucose transport across the monolayers.

■

MATERIALS AND METHODS

Chemicals. Tomatoside A was obtained according to the method of Yamanaka et al.15 Briefly, 100 g tomato seeds were homogenized in 8 L of 70% ethanol containing 0.5% acetic acid and incubated at 30 °C for 2 h. After filtration, the filtrate was applied to a Waters Sep-Pak C18 cartridge (Milford, MA, USA), followed by purification with a refractive index high-performance liquid chromatography (HPLC) (column: Waters XTerra RP18, 19 × 150 mm, 5 μm; elution: 0−100% acetonitrile containing 0.1% acetic acid over 45 min) (yield: 17−22 mg/g of dry weight). The structure of tomatoside A was confirmed by 1 H- and 13C-nuclear magnetic resonance spectrometry (NMR) (AVANCE 500, Bruker BioSpin CmbH, Karlsruhe, Germany). Dulbecco’s modified Eagle’s medium (DMEM) and fetal bovine serum (FBS) were purchased from Gibco Life Technologies (Grand Island, NY, USA). Dimethyl sulfoxide (DMSO) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Seeding basal medium, enterocyte differentiation medium, and MITO+ serum extender 1429

DOI: 10.1021/acs.jafc.7b06078 J. Agric. Food Chem. 2018, 66, 1428−1434

Article

Journal of Agricultural and Food Chemistry

were presented as the ratio of their band intensity to β-actin band intensity in the same blot. siRNA Transfection of Caco-2 Cells. Caco-2 cells were transiently transfected with ASBT or negative control siRNA, according to the manufacturer’s instructions. Briefly, Caco-2 cells were seeded in six-well tissue culture plates (2 × 105 cells/well) and transfected with siRNA (25 nM) targeting ASBT (target sequences, 5′GACAAUAGCAGUCACAACA-3′, 5′-CAUCAUAGAUUCCACGC UG -3′, 5′-GG GG AA AG CUU CA CC GGU A- 3′, a nd 5′GGGGAAAGCUUCACCGGUA-3′: sc-106906 [Santa Cruz Biotechnology] or negative control siRNA [AllStars Negative Control siRNA, QIAGEN, Valencia, CA, USA]). siRNAs were diluted with opti-MEM (Life Technologies, Gaithersburg, MD, USA) and HiPerFect transfection reagent (QIAGEN), and they were incubated for 10 min at room temperature. The mixture was added to the cells and incubated for 3 h, followed by the addition of 10% FBS/DMEM. After 48-h incubation, the medium was changed to fresh antibiotic-free differentiation medium. The cells were cultured for an additional 4 days, and retransfected with siRNA for 24 h. The ASBT−knockdown Caco-2 cells were used for 13C6-glucose transport experiments. Statistical Analyses. Results are expressed as the mean ± standard error of the mean (SEM). Statistical differences among groups were evaluated by one-factor analysis of variance (ANOVA), followed by Tukey-Kramer’s t-test for post hoc analysis. Other statistical evaluations were performed by the Student’s t-test. A p value of 1.3 pmol/mL in negative electrospray ionization mode ([M-H]−: 1081.5940 m/z) were conducted for the transported solutions of apical-to-basolateral (A to B) and basolateral-toapical (B to A) transport experiments of tomatoside A. As shown in Figure 1, no MS peak was observed at 1081.5940 m/z corresponding to tomatoside A at a retention time of 27.0 min in the basolateral solution of the A to B experiment. This indicates that the tomato seed steroidal saponin, tomatoside A, was not transported across Caco-2 cell monolayers. On the contrary, a significant MS detection of tomatoside A was observed in the B to A direction (Figure 1). These results suggest that nontransportable tomatoside A may be involved in intracellular efflux route(s) in Caco-2 cells. Glucose Transport in Tomatoside A-Treated Caco-2 Cell Monolayers. The effects of tomatoside A on glucose transport across Caco-2 cell monolayers were investigated as a function of treatment time (1, 3, 6, or 24 h) or tomatoside A concentration (5, 10, or 20 μM) using 13C6-glucose. Within the experimental conditions, no glycolytic metabolism or degradation of the penetrant, 13C6-glucose, such as 13C3-pyruvate and 13 C3-lactate, was observed by LC-TOF-MS analyses (data were not shown). As shown in Figure 2A, transported 13C6-glucose derivatized with PMP was successfully detected in both tomatoside A-treated and untreated Caco-2 cell monolayers (PMP-derivatized N-acetyl-D-glucosamine as IS). As shown in Figure 2B, a significant decrease of 46.0 ± 0.1% and 47.0 ± 3.6% in 13C6-glucose transport was observed in Caco-2 cell monolayers treated with 10 μM tomatoside A (p < 0.05) for 3 and 6 h, respectively. This is the first finding that tomatoside A, a natural steroidal saponin, has the physiological potential to suppress glucose transport. The reduction effect on glucose 1430

DOI: 10.1021/acs.jafc.7b06078 J. Agric. Food Chem. 2018, 66, 1428−1434

Article

Journal of Agricultural and Food Chemistry

Ω·cm2; 10 μM, 658 ± 31 Ω·cm2; 100 μM, 647 ± 35 Ω·cm2; 1 mM, 641 ± 27 Ω·cm2). Effect of Tomatoside A on the Expression of Glucose Transporters. Western blotting analyses of SGLT1 and GLUT2 expression were carried out in Caco-2 cell monolayers treated with 10 μM tomatoside A for 3 h (Figure 3). As shown in Figure 3B, tomatoside A significantly (p < 0.01) reduced GLUT2 expression, whereas no change in SGLT1 expression was observed (Figure 3A). A reduced GLUT2 mRNA expression by 10 μM tomatoside A was also obtained by realtime PCR analysis (Figure S2). Thus, the reduced glucose transport by tomatoside A may be associated with the suppression of GLUT2 transport route in Caco-2 cells. Effect of Tomatoside A on Glucose Transport-Related Signaling Pathways. Transport experiments of 13C6-glucose were evaluated in Caco-2 cell monolayers treated with tomatoside A (10 μM, 3 h) in the presence of 10 μM compound C (specific AMPK inhibitor) or 10 μM GF109203X (nonspecific PKCs inhibitor). As shown in Figure 4A, 10 μM compound C failed to alter tomatoside A-induced reduction in glucose transport across Caco-2 cells, although the concentration of compound C was confirmed to inhibit AMPK activity in Caco-2 cells in a previous theaflavin transport study.11 On the other hand, the glucose transport was significantly restored (p < 0.05) to the level observed in control cells in the presence of GF109203X (Figure 4B). GLUT2 expression was also restored following GF109203X treatment (Figure 4C), suggestive of the role of tomatoside A in PKCs/GLUT2 signaling pathway in Caco-2 cells. Indeed, PKC was significantly activated in tomatoside A (10 μM, 3 h)-treated Caco-2 cells compared to that of untreated cells (Figure S3). Given the apparent basolateral to apical directional transport of tomatoside A (Figure 1), the involvement of tomatoside A in efflux route(s) in Caco-2 cells was investigated. As shown in Figure 4D, tomatoside A-induced reduction in glucose transport was significantly (p < 0.05) ameliorated by 10 μM MK571 (MRP2-mediated efflux blocker17) but not 10 μM valspodar (P-gp-mediated efflux blocker18). This clearly indicates that tomatoside A that is taken up by Caco-2 cells may be excreted through the MRP2 efflux route.

Figure 2. Effects of tomatoside A on 13C6-glucose transport across Caco-2 cell monolayers. (A) LC-TOF-MS chromatograms of PMPderivatized 13C6-glucose at the basolateral side after 60 min transport using untreated or tomatoside A-treated Caco-2 cell monolayers. Nacetyl-D-glucosamine was used as an internal standard (IS). Caco-2 cells were treated with 10 μM tomatoside A for 3 h. Positive ESI-MS chromatograms were obtained for PMP-derivatized 13C6-glucose (m/z 517.2180) and PMP-derivatized IS (m/z 552.2473, 100 μM). (B) Time-dependent effect of tomatoside A on glucose transport. Caco-2 cells were treated with 10 μM tomatoside A for up to 24 h. (C) Concentration-dependent effect of tomatoside A on glucose transport. Caco-2 cells were treated with 5−20 μM tomatoside A for 3 h. Results are expressed as the mean ± SEM (n = 3). In (B) and (C), statistical differences among the groups were evaluated by the Tukey-Kramer’s ttest. The different letters represent the statistical differences at p < 0.05.

transport was in a concentration-dependent manner at 3-h incubation (Figure 2C). At a concentration of more than 10 μM tomatoside A, treatment resulted in a significant reduction in glucose transport (p < 0.05, Figure 2C). Hence, we chose a 10 μM concentration for 3 h for further experiments. No cytotoxicity was observed following the 3-h treatment of Caco2 cells with up to 1 mM of tomatoside A, as revealed by the CCK-8 assay (Figure S1). Moreover, the integrity of Caco-2 monolayers was not affected by tomatoside A treatment within the present experimental conditions (TEER: control, 664 ± 5

Figure 3. Effects of tomatoside A on SGLT1 and GLUT2 expression in Caco-2 cells analyzed by Western blotting. Caco-2 cells were incubated with 10 μM tomatoside A for 3 h. Results are expressed as the mean ± SEM (n = 3). The level of each protein was calculated by the ratio of SGLT1 or GLUT2 amount to β-actin amount and expressed as the percentage of total protein level in the cell lysate. Statistical differences were evaluated by the Student’s t-test. N.S. indicates no significant difference at p > 0.05. 1431

DOI: 10.1021/acs.jafc.7b06078 J. Agric. Food Chem. 2018, 66, 1428−1434

Article

Journal of Agricultural and Food Chemistry

Figure 5. Role of the ASBT incorporation route in tomatoside Ainduced reduction in 13C6-glucose transport across Caco-2 cell monolayers. Fluvastatin (a specific ASBT antagonist) or ASBT siRNA was used for the 60 min transport experiment. Caco-2 cells were pretreated with 100 μM fluvastatin for 3 h or transiently transfected either with an ASBT siRNA or a nontargeting siRNA for 24 h (each siRNA at 25 nM). Results are expressed as the mean ± SEM (n = 3). Statistical differences among the groups were evaluated by the Tukey-Kramer’s t-test. The different letters represent the statistical differences at p < 0.05.

Figure 4. Effect of AMPK or PKC inhibition on tomatoside A-induced reduction in 13C6-glucose transport across Caco-2 cell monolayers. Caco-2 cells were treated with 10 μM tomatoside A in the presence and absence of 10 μM compound C (AMPK inhibitor) (A) or 10 μM GF109203X (PKCs inhibitor) (B) for 60 min. GLUT2 expression in the presence or absence of 10 μM GF109203X was analyzed by Western blotting (C). Effect of efflux routes on tomatoside A-induced reduction in 13C6-glucose transport was investigated using valspodar (10 μM) as P-gp inhibitor or MK571 (10 μM) as MRP2 inhibitor in the 60 min Caco-2 transport experiment (D). Results are expressed as the mean ± SEM (n = 3). Statistical differences among the groups were evaluated by the Tukey-Kramer’s t-test. The different letters represent the statistical differences at p < 0.05.

Involvement of Tomatoside A in Influx Route. Given the structural similarity between tomatoside A and bile acid molecules (both being steroidal triterpenoid skeleton), we hypothesized the involvement of ASBT, a bile acid transporter, in the uptake of tomatoside A in Caco-2 cells. As shown in Figure 5, 100 μM fluvastatin, a specific ASBT antagonist, significantly (p < 0.05) restored tomatoside A-induced reduction in glucose transport. The ASBT-mediated uptake of tomatoside A was also confirmed in ASBT-knockdown Caco-2 cell monolayers. As shown in Figure 5, the reduction in glucose transport by tomatoside A was significantly (p < 0.05) ameliorated following ASBT knockdown in Caco-2 cells. Thus, ASBT may at least be associated with the influx of tomatoside A.

Figure 6. Proposed action of tomatoside A on influx/efflux-related signaling pathways in Caco-2 cells.

glucose transport upon treatment at 10 μM concentration for 3 h (Figure 2). Saponins occur naturally in plants and have been shown to display biological characteristics, such as anti-inflammatory,19 antibacterial,20 and antiparasitic activities.21 Aside from their health-promoting effects, saponins are known to act as detergents and may cause cell membrane disruption or toxic hemolysis.22,23 In this study, tomatoside A at