Food Chemistry 138 (2013) 101–106

Contents lists available at SciVerse ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Transepithelial transport of 6-O-caffeoylsophorose across Caco-2 cell monolayers

Hoang Lan Phuong a, Ju Qiu a, Takashi Kuwahara a, Keiichi Fukui b, Kayo Yoshiyama b,Kazusato Matsugano b, Norihiko Terahara c, Toshiro Matsui a,⇑a Department of Bioscience and Biotechnology, Division of Bioresource and Bioenviromental Sciences, Faculty of Agriculture, Graduated School of Kyushu University, 6-10-1 Hakozaki,Fukuoka 812-8581, Japanb Miyazaki JA Food Research & Development Inc., Ikimedai, Miyazaki 880-0943, Japanc Department of Food Science for Health, Faculty of Health and Nutrition, Minami-Kyushu University, 5-1-2 Kirishima, Miyazaki 880-0032, Japan

a r t i c l e i n f o

Article history:Received 21 July 2012Received in revised form 19 September 2012Accepted 22 October 2012Available online 12 November 2012

Keywords:CaffeoylsophoroseCaco-2 cellsMonocarboxylic acid transport

0308-8146/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.foodchem.2012.10.137

⇑ Corresponding author. Tel./Fax: +81 92 642 3011E-mail address: tmatsui@agr.kyushu-u.ac.jp (T. Ma

a b s t r a c t

The aim of this study was to clarify the transport behaviour and mechanism of caffeic acid analogue bear-ing a sugar-moiety, 6-O-caffeoylsophorose (CS), in Caco-2 cells. The absorption of CS was investigated byits transport across Caco-2 cell monolayers using a high-performance liquid chromatography-time-of-flight-mass spectrometry (LC–TOF-MS). The permeation of CS was concentration-dependent and reachedthe plateau at >6 mM. The apparent permeability (Papp) of CS in the apical-to-basolateral directionwas 5.4 � 10�7 cm/s, while in the reversed direction the Papp value was significantly reduced(1.9 � 10�7 cm/s). CS transport was competitively inhibited by phloretin, an inhibitor of monocarboxylicacid transporter (MCT). Benzoic acid, an MCT substrate, also reduced CS transport. A less significant changeof CS transport was observed across Caco-2 cell monolayers pretreated with quercetin, a suppressor oftight-junction. These findings strongly indicate that CS, a caffeic acid analogue bearing sophorose moiety,can be transported across Caco-2 cell monolayers via the MCT pathway.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction the first finding that a mono-phenolic acid bearing a sugar moiety

Recently, dietary phenolic acid and flavonoids have receivedmuch attention because of their potential health benefits such asanti-obesity (Tsuda, 2008), anti-diabetic mellitus (Iwai, 2008), car-diovascular disease prevention (Babu & Liu, 2008), and anti-cancer(Fresco, Borges, Marques, & Diniz, 2010). In contrast, studies onabsorption, distribution, metabolism and excretion (ADME) analy-ses are being required for demonstrating in vivo the functionality ofbioactive compounds. Many studies on the absorption of bioactivecompounds have been reported so far (Del Rio, Borges, & Crozier,2010; Suda et al., 2002) including 6-O-caffeoylsophorose (CS;6-O-(E)-caffeoyl-(2-O-b-D-glucopyranosyl)-a-D-glucopyranoside)that is a natural compound (CAS no. 845724–39-2) isolated from ared vinegar fermented with the storage root paste of purple-fleshed sweet potato (Terahara et al., 2003).

In our previous studies, we have demonstrated that CS that pos-sesses some bioactivities such as intestinal a-glucosidase inhibi-tion (Matsui et al., 2004) and anti-oxidative activity (Teraharaet al., 2003, 2009) was absorbed in intact form into blood after asingle oral administration to Sprague–Dawley (SD) rats (Qiuet al., 2011). Although the administered CS was to some extentmetabolised by methylation, glucuronidation or sulfatation, ordegraded to caffeic acid and ferulic acid (Qiu et al., 2011), it was

ll rights reserved.

.tsui).

can be absorbed in intact form into rat blood system.It has been reported that mono-phenolic acids, such as p-cou-

maric acid and ferulic acid, were absorbed by monocarboxylic acidtransporter (MCT) in Caco-2 cells or human colon adenocarcinomacells (Konishi, Kobayashi, & Shimizu, 2003; Konishi & Shimizu,2003). In contrast, caffeic acid that is a mono-phenolic acid was re-ported to be absorbed by the paracellular transport pathway aswell as the MCT transport in Caco-2 cell monolayers (Konishi &Kobayashi, 2004). These diverse transport mechanisms of mono-phenolic acids strongly permit us to investigate the mechanism(s)underlying CS transport, owing to its intact absorption in SD rats(Qiu et al., 2011) despite the mono-phenolic structure bearing su-gar moiety (or sophorose). In the present study, thus, we tried toclarify the transport pathway involved in CS absorption usingCaco-2 cell monolayers, in which expressions of some transporters,e.g., peptide-, glucose-, and MCT-transporters, have already beenclarified (Shah, Jogani, Bagchi, & Misra, 2006; Walgren, Lin, Kinne,& Walle, 2000; Zhu et al., 2008).

2. Materials and methods

2.1. Materials

Caffeic acid, ferulic acid, phloretin and benzoic acid were pur-chased from Wako Pure Chemical Industry (Osaka, Japan). Fluores-cein was purchased from Sigma Chemical Co. (St. Louis, MO, USA).

102 H.L. Phuong et al. / Food Chemistry 138 (2013) 101–106

Quercetin was purchased from Tokyo Chemical Industry (Tokyo,Japan). Other reagents were purchased from Nacalai Tesque Inc.(Kyoto, Japan) and used without further purification.

2.2. Preparation of CS

CS was prepared according to the method in our previous report(Terahara et al., 2003). Briefly, purple-flesh sweet potato (I. batatascv. Ayamurasaki) was fermented to produce a red vinegar. Next, 3-Lof red vinegar was evaporated and freeze-dried (4.9 g powder).Then, the powder was dissolved in methanol and applied to high-performance liquid chromatography (HPLC) with conditions asfollows: preparative column, Inertsil ODS 5 (Ø 20 mm � 250 mm);solvent, 30% acetonitrile containing 15% acetic acid; flow rate,7.0 ml/min; detector, L-4200 UV–vis at 310 nm. The peakcorresponding to CS elution with the retention time of 12.9 minwas collected and evaporated to dryness (purity > 95%).

2.3. Cell culture

Caco-2 cells were cultured in Dulbecco’s modified Eagle’s med-ium (DMEM) supplemented with 10% foetal bovine serum (FBS)(Gibco Life Technologies, Grand Island, NY, USA), 1% nonessentialamino acid (Tormerly ICN Biomedical, Inc., Germany), 2 mML-glutamine (Nacalai Tesque), 100 U/ml penicillin (Meiji Seika,Tokyo, Japan), 100 lg/ml streptomycin (Nacalai Tesque), and1.7 lM insulin (Sigma Chemical). The cells were incubated at37 �C, in a humidified atmosphere of 5% CO2 in air. The monolayersbecame confluent 4–5 days after seeding at a number of 1.0 � 106

cells per 100 mm dish and the cells were passaged at a split ratio of4–8 by trypsinisation with 0.25% trypsin and 0.02% di-sodium saltof ethylenediamine tetraacetic acid (EDTA) in phosphate bufferedsaline (PBS, Cosmo Bio Co., Tokyo, Japan). Caco-2 cells used in thisstudy were between passages 50 and 60. For the transport study,the cells seeded at a density of 4.0 � 105 cells/ml were grown ina transwell cell culture insert (polycarbonate membrane, 0.4 lmpore size; 12 mm, Costar, Bodenheim, Germany) with a membranecoated with type I collagen (collagen gel culturing kit, Cellmatrixtype I-A, Nitta Gelatin, Osaka, Japan). The cells were cultured usinga BD BioCoatTM intestinal epithelium differentiation kit (BD Biosci-ences, Bedford, MA, USA) and the medium was changed everyday.Monolayers were formed after 6 days. The integrity of the cell layerwas evaluated by measuring the transepithelial electrical resis-tance (TEER) with a Multi-channel Voltage/Current EVC-4000 sys-tem (World Precision Instruments, FL, USA). Monolayers with TEERof >100 O cm2 were used for the transport experiments.

2.4. Transport experiments of CS across Caco-2 cell monolayers

A transport experiment was conducted with Caco-2 cell mono-layers in an Using chamber system (Model U-2500, Warner Instru-ment Corporation, Hamden, CT, USA) as described previously (Zhuet al., 2008). Caco-2 cell monolayers grown in the transwell insertswere gently rinsed with Hank’s balanced salt solution (HBSS, pH6.0) before the experiments. An aliquot (6.0 ml) of HBSS buffer(pH 6.0, adjusted with 10 mM 2-(N-morpholino)ethanesulfonicacid (MES)) was added to apical side of the chamber, and 6.0 mlof HBSS buffer (pH 7.4, adjusted with 10 mM 2-[4-(2-hydroxy-ethyl)-1-piperazinyl]ethanesulfonic acid (HEPES)) were added tobasolateral side. After equilibration for 15 min at 37 �C, the trans-port experiments were started by replacing the HBSS buffer with6.0 ml of CS solution in the presence or absence of inhibitors tothe apical side (pH 6.0). During the transport experiments at37 �C, the solutions in both sides were bubbled continuously witha mixture of O2:CO2 (95:5) through air vents in the chamber. At

time intervals up to 30 min, a solution (1.0 ml) was drawn fromthe basolateral side and replaced with the same volume of thefresh buffer. For the transport experiments in the basolateral-to-apical direction, 6.0 ml of HBSS buffer (pH 6.0 or 7.4) was addedto apical side, and 6.0 ml of CS solution (pH 7.4) were added tothe basolateral side. The solutions were, then, subjected to aSep-Pak Plus C18 Cartridge (Waters, Milford, MA, USA) treatment;i.e., a fraction containing CS was eluted with 3 ml of 40% acetoni-trile in 0.1% trifluoroacetic acid and was evaporated to dryness.

2.5. LC–TOF-MS analysis

The amount of CS in transported solution was determined withan LC–time-of-flight-mass spectrometry (LC–TOF-MS). Namely, analiquot (20 ll) of CS fraction dissolved in 0.1% formic acid (FA) wasapplied to the system. LC separation was performed with an Agi-lent 1200 series (Agilent, Waldbronn, Germany) on a Cosmosil5C18-MS-II column (Ø 2.0 mm � 150 mm, Nacalai Tesque) at40 �C. The mobile phase consisted of 0.1% FA (solvent A) and meth-anol with 0.1% FA (solvent B) using a 20 min-linear gradient from0% to 100% of solvent B at a flow rate of 0.2 ml/min (Qiu et al.,2011). MS analysis was performed by a micrOTOF II (BrukerDaltonics, Bremen, Germany). The amount of CS (m/z 504.1406)was analysed in negative ion mode of electrospray ionisation.The MS conditions were as follows: drying gas, N2; flow rate,8.0 l/min; drying gas temperature, 200 �C; drying gas pressure,1.6 bar; and capillary voltage, 3800 V. The calibration solution of10 mM sodium formate in 50% acetonitrile was injected at thebeginning of the run. The data were analysed and acquired by aBruker Data Analysis 3.2 software.

2.6. Inhibition studies of CS transport

Inhibition experiments using quercetin, with a tight-junctionclosing effect (Suzuki & Hara, 2009), were performed in Caco-2cells treated with 200 lM quercetin for 24 h before transportexperiments. Other inhibition studies were also performed using300 lM phloretin as MCT inhibitor (Halestrap & Meredith, 2004)and 10 or 20 mM benzoic acid as MCT substrate (Tsuji, Takanaga,Tamai, & Terasaki, 1994). Lineweaver–Burk plots assayed the CStransport, from 0.5 to 4 mM in the presence or absence of300 lM phloretin.

2.7. Fluorescein transport assay

Fluorescein transport experiments were performed by 200 lMfluorescein in the presence or absence of either 1 mM CS or200 lM quercetin at pH 7.4 for both apical and basolateral sides.At time intervals of 5, 10, 15, 20, 25 and 30 min, the amounts oftransported fluorescein were determined with a fluorescence spec-trophotometer (Wallac ARVO SX 1420 Multilabel Counter, PerkinElmer Life Sciences, Tokyo, Japan) at excitation wavelength of490 nm and emission wavelength of 514 nm.

2.8. Analysis of apparent permeability coefficient

Permeability was evaluated by the apparent permeability coef-ficient (Papp).

Pappðcm=sÞ ¼ ðdC=dtÞ � V=ðA� C0Þ

where dC/dt is calculated with plotting the amount of CS trans-ported to the basolateral side versus time and determining theslope of the line, V and C0 are volume (6.0 ml) and initial concentra-tion in the donor compartment, respectively, and A is the surfacearea of monolayer (0.2826 cm2).

H.L. Phuong et al. / Food Chemistry 138 (2013) 101–106 103

2.9. Data analysis

Statistical analysis between groups was performed with one-way analysis of variance (ANOVA), followed by Tukey–Kramer’st-test. Difference between two groups was analysed by unpairedStudent’s t-test. Differences with p < 0.05 were considered to bestatistically significant. Results are expressed as the mean ± SEM.

3. Results

3.1. LC–MS analysis of CS transport across Caco-2 cell monolayers

Under the present LC–TOF-MS conditions, the transport of CS inthe basolateral side was successfully detected at 18.7 min on a5C18-MS-II column (Fig. 1A). The sensitivity of CS was estimatedto be 20 pmol/ml in negative mode. No peak corresponding to CSwas detected in the basolateral solution before transport experi-ments. Though data were not shown, caffeic acid (m/z 179.0350)and ferulic acid (m/z 193.0506) standards were also detected byLC–TOF-MS with retention times of 19.0 and 21.1 min on thecolumn, respectively. As a result of the Caco-2 cell transport

Fig. 1. Typical traces of CS in basolateral solutions at 0 and 30 min after Caco-2 cell tconcentration across Caco-2 cell monolayers with inwardly directed proton gradient (negative mode was performed at m/z 503.1406 ± 0.005. The arrow shows a peak correConcentrations of CS were ranged from 0.5 to 10 mM. Results are expressed as the mea

experiments of CS and its related compounds (caffeic and ferulicacids), the magnitude of CS transport (or Papp value) across theCaco-2 cells was 8- and 100-fold lower than those of caffeic andferulic acids, respectively (Table 1). In addition, the transport rateof CS increased with the initial concentrations of CS, up to 6 mMin the apical side, and reached the plateau at >6 mM (Fig. 1B), spec-ulating the involvement of carrier-mediated transport in CSabsorption.

3.2. Vectorial characteristics of transepithelial transport of CS

Fig. 2 shows the transport of CS across Caco-2 cell monolayerswith or without inwardly directed proton gradient (pH 7.4 or pH6.0–7.4 or vice versa). The transport of CS in the basolateral-to-apical direction was ca. 2.5-fold lower than that in the apical-to- basolateral direction. CS transport without proton gradientresulted in a lower Papp value than that with a proton gradient inthe apical-to-basolateral direction and no significant differencebetween both directions: Papp of CS in the apical-to-basolateraldirection, 2.8 ± 0.1 � 10�7 cm/s; in the basolateral-to-apical direc-tion, 2.3 ± 0.1 � 10�7 cm/s. These results indicate that CS may be

ransport experiments by LC–TOF-MS (A) and change in transport rate of CS withB). CS standard (54 ng/ml) was injected to LC–TOF-MS system. Detection of CS insponding to CS. pH for apical side was set at 6.0 and for basolateral side at 7.4.

n ± SEM (n = 3–6).

Table 1Apparent permeability coefficients (Papp) of 6-O-caffeoylsophorose and its relatedcompounds in Caco-2 cell monolayers.

Compound Papp � 10�6 (cm/s)

6-O- Caffeoylsophorose 0.54 ± 0.001a

Caffeic acid 3.9 ± 0.2b

Ferulic acid 57.5 ± 3.4c

Results are expressed as the mean ± SEM (n = 3–6). Different letters indicate sig-nificant difference at p < 0.05.

Fig. 2. Characteristic of transepithelial transport of CS across Caco-2 cell monolay-ers. Transport of CS (1 mM) from apical side to basolateral side (A–B) or frombasolateral side to apical side (B–A) was measured in the presence and absence ofproton gradient (apical pH, 6.0 or 7.4; basolateral pH, 7.4). Results are expressed asthe mean ± SEM (n = 3–6). Statistical difference between two groups was analysedby Student’s t-test. N.S., no significance at p > 0.05.

Fig. 3. Effect of phloretin (A) and benzoic acid (B) on CS transport across Caco-2 cellmonolayers. Transport of CS (1 mM) from apical side to basolateral side (apical pH,6.0; basolateral pH, 7.4) was measured in the presence and absence of phloretin(300 lM) or benzoic acid (10 or 20 mM). Results are expressed as the mean ± SEM(n = 3–6). Statistical difference was analysed by Student’s t-test. ⁄p < 0.05, ⁄⁄p < 0.01vs. CS-group.

Fig. 4. Lineweaver–Burk plots of CS transport in the presence or absence of 300 lMphloretin across Caco-2 cell monolayers. Transport rate of CS was measured atconcentrations ranging from 0.5 to 4 mM of CS. Results are expressed as themean ± SEM (n = 3–6).

104 H.L. Phuong et al. / Food Chemistry 138 (2013) 101–106

transported across the Caco-2 cell monolayers through protongradient-aided carrier-mediated transport pathways.

3.3. Inhibition studies of CS transport across Caco-2 cell monolayers

Fig. 3A shows the effect of phloretin (300 lM), an MCT inhibitor,on CS transport. A significant (p < 0.01) 42% reduction in Papp of CSwas observed in the presence of phloretin. Benzoic acid, an MCTsubstrate, also affected the CS transport with significant reductionsof 22% and 47% at 10 and 20 mM benzoic acid, respectively(Fig. 3B). In addition to these findings, the Lineweaver–Burk plotof CS (0.5–4.0 mM) in the presence of 300 lM phloretin (Fig. 4)strongly revealed that CS transport was competitively inhibitedby the MCT inhibitor or CS was transported by the monocarboxylicacid transporter MCT.

3.4. Paracellular transport of CS across Caco-2 cell monolayers

Transport experiments using fluorescein, a marker of paracellu-lar (or tight-junction) transport without proton gradient (pH 7.4for both apical and basolateral sides) (Konishi, Hagiwara, &Shimizu, 2002) were conducted to evaluate a possible transportof CS in diverse transport pathways. As shown in Fig. 5A, CSsignificantly reduced the fluorescein transport, but the reductionpower of 17% by 1 mM CS was much low, suggesting that CSslightly affected fluorescein transport via tight-junction. A smallerinvolvement of tight-junction in CS transport also revealed that theCS transport across Caco-2 cells treated by 200 lM quercetin,which is a closer of tight-junction (Suzuki & Hara, 2009) was notsignificantly reduced (Fig. 5C), as compared to the great reductionof fluorescein transport by quercetin (Fig. 5B). These results clearlydemonstrate that a possible transport pathway of CS may be a

MCT-mediated route in Caco-2 cells, while less consideration mustbe paid for other routes including tight-junction.

4. Discussion

Recent studies on functional foods have investigated theabsorption of food compounds because of the lack of ADMEanalyses. In this study, we investigated the characteristics of thetransport mechanism of CS in Caco-2 cells, since CS was found tobe an absorbable bioactive natural compound in intact form (Qiuet al., 2011) regardless of its unique structure bearing sugar

Fig. 5. Effect of CS (A) and quercetin (B) on fluorescein transport, and effect ofquercetin on CS transport (C). Fluorescein transport in the presence or absence of CS(1 mM) or quercetin (200 lM, a closer of tight-junction) was measured at pH 7.4 forapical and basolateral sides by a fluorescence spectroscopy at excitation wave-length of 490 nm and emission wavelength of 514 nm. Effect of quercetin (200 lM)on CS transport was measured by LC–TOF-MS. Results are expressed as themean ± SEM (n = 5–6). ⁄p < 0.05, ⁄⁄p < 0.01 vs. control.

H.L. Phuong et al. / Food Chemistry 138 (2013) 101–106 105

moiety. In the Caco-2 cell transport experiments, we demonstratedthat CS was transported across the Caco-2 cell monolayers throughproton gradient-aided carrier-mediated transport pathways. Theresult that phloretin competitively inhibited CS transport (Figs. 3and 4) strongly suggested that the transport route of CS in theCaco-2 cells was a MCT-mediated pathway. Less significant reduc-tion of CS transport by quercetin (Fig. 5) also supported the MCT-mediated transport of CS, but not paracellular transport.

For the absorption of caffeic acid that is a skeleton of CS, manyresearchers have already clarified the details using animal and cellexperiments. Spencer et al. (1999) revealed that less caffeic acidwas absorbed through the jejunum of SD rats, while its derivatisedcompound, ferulic acid, was predominantly detected in blood, sug-gesting that O-methylation of caffeic acid increased its transferthrough the intestinal barrier. In the present Caco-2 cell transportexperiments, we also demonstrated a preferable transport offerulic acid rather than caffeic acid and CS (Table 1), similar to

the report by Konishi and Kobayashi (2004). In addition, a 10-foldlower permeability of CS than that of caffeic acid (Table 1) specu-lated the structural interference of the sugar moiety in CS withtransport.

It has been reported that caffeic acid and its related compoundssuch as ferulic acid and p-coumaric acid were transported via theMCT pathway (Konishi & Kobayashi, 2004; Konishi & Shimizu,2003; Konishi et al., 2003). It would be applicable that MCT canrecognise the carboxyl group, such as phenolic acid, as well asthe non-polar side chain or aromatic moiety to be absorbed bythe carrier-mediated transport system. Contrary to the prevalenceof phenolic acids via the MCT transport, it has been reported thatepicatechin gallate, a penetrant having no carboxyl group, was alsotransported via the MCT transport pathway, like CS (Vaidyanathan& Walle, 2003). In this regard, the structural factors responsible forMCT recognition were controversial, since Caco-2 cells expresssome MCT subtypes (Vaidyanathan & Walle, 2003). Our presentfindings indicate that CS, probably caffeoyl moiety, may be recog-nised by MCT, regardless of the fact that there is no carboxyl groupin the CS structure. Thus, further studies must be needed to clarifythe characteristics of MCT recognition for CS structure.

In conclusion, this study provided the first finding that CS, 6-O-caffeoylsophorose, can be transported across Caco-2 cell monolay-ers through an MCT-carrier mediated transport pathway, whileless involvement of the paracellular pathway in CS transport wasobtained. This remarkably contributes to studies related to thehealth benefits of CS, and also provides structural information ofMCT substrates that have no carboxyl groups.

References

Babu, P. V., & Liu, D. (2008). Green tea catechins and cardiovascular health: Anupdate. Current Medicinal Chemistry, 15, 1840–1850.

Del Rio, D., Borges, G., & Crozier, A. (2010). Berry flavonoids and phenolics:Bioavailability and evidence of protective effects. British Journal of Nutrition, 104,S67–S90.

Fresco, P., Borges, F., Marques, M. P., & Diniz, C. (2010). The anticancer properties ofdietary polyphenols and its relation with apoptosis. Current PharmaceuticalDesign, 16, 114–134.

Halestrap, A. P., & Meredith, D. (2004). The SLC16 gene family-frommonocarboxylate transporters (MCTs) to aromatic amino acid transportersand beyond. Pflugers Archiv: European Journal of Physiology, 447, 619–628.

Iwai, K. (2008). Antidiabetic and antioxidant effects of polyphenols in brown algaEcklonia stolonifera in genetically diabetic KK-Ay mice. Plant Foods for HumanNutrition, 63, 163–169.

Konishi, Y., Hagiwara, K., & Shimizu, M. (2002). Transepithelial transport offluorescein in Caco-2 cell monolayers and use of such transport in in vitroevaluation of phenolic acid availability. Bioscience, Biotechnology, andBiochemistry, 66, 2449–2457.

Konishi, Y., & Kobayashi, S. (2004). Transepithelial transport of chlorogenic acid,caffeic acid, and their colonic metabolites in intestinal Caco-2 cell monolayers.Journal of Agricultural and Food Chemistry, 52, 2518–2526.

Konishi, Y., Kobayashi, S., & Shimizu, M. (2003). Transepithelial transport of p-coumaric acid and gallic acid in Caco-2 cell monolayers. Bioscience,Biotechnology, and Biochemistry, 67, 2317–2324.

Konishi, Y., & Shimizu, M. (2003). Transepithelial transport of ferulic acid bymonocarboxylic acid transporter in Caco-2 cell monolayers. Bioscience,Biotechnology, and Biochemistry, 67, 856–862.

Matsui, T., Ebuchi, S., Fukui, K., Matsugano, K., Terahara, N., & Matsumoto, K. (2004).Caffeoylsophorose, a new natural a-glucosidase inhibitor, from red vinegar byfermented purple-fleshed sweet potato. Bioscience, Biotechnology, andBiochemistry, 68, 2239–2246.

Qiu, J., Saito, N., Noguchi, M., Fukui, K., Yoshiyama, K., Matsugano, K., et al. (2011).Absorption of 6-O-caffeoylsophorose and its metabolites in Sprague–Dawleyrats detected by electrochemical detector-high-performance liquidchromatography and electrospray ionization-time-of-flight-massspectrometry methods. Journal of Agricultural and Food Chemistry, 59,6299–6304.

Shah, P., Jogani, V., Bagchi, T., & Misra, A. (2006). Role of Caco-2 cell monolayers inprediction of intestinal drug absorption. Biotechnolgy Progress, 22, 186–198.

Spencer, J. P. E., Chowrimootoo, G., Choudhury, R., Debnam, E. S., Srai, S. K., & Rice-Evans, C. (1999). The small intestine can both absorb and glucuronidate luminalflavonoids. FEBS Letters, 458, 224–230.

Suda, I., Oki, T., Masuda, M., Nishiba, Y., Furuta, S., Matsugano, K., et al. (2002). Directabsorption of acylated anthocyanin in purple-fleshed sweet potato into rats.Journal of Agricultural and Food Chemistry, 50, 1672–1676.

106 H.L. Phuong et al. / Food Chemistry 138 (2013) 101–106

Suzuki, T., & Hara, H. (2009). Quercetin enhances intestinal barrier function throughthe assembly of zonnula occludens-2, occludin, and claudin-1 and theexpression of claudin-4 in Caco-2 cells. Journal of Nutrition, 139, 965–974.

Terahara, N., Matsui, T., Fukui, K., Matsugano, K., Sugita, K., & Matsumoto, K. (2003).Caffeoylsophorose in a red vinegar produced through fermentation with purplesweetpotato. Journal of Agricultural and Food Chemistry, 51, 2539–2543.

Terahara, N., Matsui, T., Minoda, K., Nasu, K., Kikuchi, R., Fukui, K., et al. (2009).Functional new acylated sophoroses and deglucosylated anthocyanins in afermented red vinegar. Journal of Agricultural and Food Chemistry, 57,8331–8338.

Tsuda, T. (2008). Regulation of adipocyte function by anthocyanins; possibility ofpreventing the metabolic syndrome. Journal of Agricultural and Food Chemistry,56, 642–646.

Tsuji, A., Takanaga, H., Tamai, I., & Terasaki, T. (1994). Transcellular transport ofbenzoic acid across Caco-2 cells by a pH-dependent and carrier-mediatedtransport mechanism. Pharmaceutical Research, 11, 30–37.

Vaidyanathan, J. B., & Walle, T. (2003). Cellular uptake and efflux of the tea flavonoid(�)-epicatechin-3-gallate in the human intestinal cell line Caco-2. Journal ofPharmacology and Experimental Therapeutics, 307, 745–752.

Walgren, R. A., Lin, J. T., Kinne, R. K., & Walle, T. (2000). Cellular uptake of dietaryflavonoid quercetin 40-b-glucoside by sodium-dependent glucose transporterSGLT1. Journal of Pharmacology and Experimental Therapeutics, 294, 837–843.

Zhu, X. L., Watanabe, K., Shiraishi, K., Ueki, T., Noda, Y., Matsui, T., et al. (2008).Identification of ACE-inhibitory peptides in salt-free soy sauce that aretransportable across Caco-2 cell monolayers. Peptides, 29, 338–344.