Repository.umac.mo

Journal of Agricultural and Food Chemistry
Characterization of the Determinant Factors for Systemic Exposures of Mulberroside A and Oxyresveratrol, Two Main Bioactive Constituents in Mulberry (Morus alba L.) Journal: Journal of Agricultural and Food Chemistry Manuscript ID: Draft Manuscript Type: Article Date Submitted by the n/a Complete List of Authors: Mei, Mei; University of Macau, State Key Laboratory of Quality Research of Chinese Medicine Ruan, Jian-Qing; University of Macau, State Key Laboratory of Quality Research of Chinese Medicine Wu, Wen-Jin; University of Macau, State Key Laboratory of Quality Research of Chinese Medicine Zhou, Rui-Na; University of Macau, State Key Laboratory of Quality Research of Chinese Medicine Zhao, Hai-Yu; University of Macau, State Key Laboratory of Quality Research of Chinese Medicine Yan, Ru; University of Macau, State Key Laboratory of Quality Research of Chinese Medicine Wang, Yi-Tao; University of Macau, State Key Laboratory of Quality Research of Chinese Medicine; School of Chinese Materia Medica ACS Paragon Plus Environment
Page 1 of 33
Journal of Agricultural and Food Chemistry
Characterization of the Determinant Factors for Systemic Exposures of Mulberroside A
and Oxyresveratrol, Two Main Bioactive Constituents in Mulberry (Morus alba L.)
Mei Meia, Jian-Qing Ruana, Wen-Jin Wua, Rui-Na Zhoua, Hai-Yu Zhaoa, Ru Yana,*, and Yi- aState Key Laboratory of Quality Research of Chinese Medicine, Institute of Chinese Medical Sciences, University of Macau, Taipa, Macao, China bSchool of Chinese Materia Medica, Beijing University of Chinese Medicine, Beijing, China *To whom correspondence should be addressed Tel: 853-83974876 Fax: 853-28841358 Email: [email protected] ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 2 of 33
Abstract
Mulberroside A (MulA) and its aglycone oxyresveratrol (OXY) are two main bioactive constituents in mulberry (Morus alba L.). This study examined the determining factors for previously reported in vivo pharmacokinetic profiles of MulA and OXY on in vitro models. When incubated anaerobically with intestinal bacteria, MulA exhibited rapid deglycosylation and generated two monoglucosides and OXY sequentially. MulA exhibited a poor permeability and predominantly traversed Caco-2 cells via passive diffusion, yet the permeation of OXY across Caco-2 cells involved efflux (both p-glycoprotein and MRPs)- mediated mechanisms. Moreover, MulA was stable in liver subcellular preparations, yet OXY underwent extensive glucuronidation. The species difference was insignificant in intestinal bacterial conversion of MulA and the extent of OXY hepatic glucuronidation between humans and rats, while OXY exhibited distinct positional preference of glucuronidation in the two species. Overall, these findings warrant further investigational emphasis on OXY and its hepatic metabolites for understanding benefits of mulberry. Keywords:
Morus alba L.; mulberroside A; oxyresveratrol; deglycosylation; permeability; ACS Paragon Plus Environment
Page 3 of 33
Journal of Agricultural and Food Chemistry
Morus alba L. (mulberry) is a well known plant which is widely cultivated in many countries for different purposes, including the leaves as feed for silkworms, fruits as dietary supplements, and leaves, fruits, twigs and root bark for medicinal uses in some Asian countries.1 Modern biological evaluations have revealed health-promoting effects of mulberry, including antibacterial, hypolipidemic, antioxidant, antityrosinase, neuroprotective, nephroprotective, hypoglycemic, antidiabetic and adaptogenic activities, etc. 2-6, supporting its attractive prospects for health promotion. Mulberry is rich in polyphenols, alkaloids, flavonoids, and anthocyanins, which have been suggested to be responsible for its health benefits. Trans-hydroxystilbenes are a group of polyphenolic compounds naturally occurring in large amounts in mulberry.7 Mulberroside A (MulA, Fig. 1) and its aglycone oxyresveratrol (OXY, Fig. 1) are two main polyhydroxylated stilbenes of the water extract of the root bark of mulberry.8 Both MulA and OXY have exhibited wide beneficial profiles 9-13, and thereby are considered as the main active components of mulberry in a conventional oral route of aqueous decoctions. Recently, OXY has received more attentions because of its structural similarity with resveratrol (3,5,4'- trihydroxy-trans-stilbene), the most important stilbene related to human health, yet more potent bioactivities that have been revealed.10,11-13 MulA exhibited a poor oral bioavailability ( 1%) in the rat.14 Zhaxi and co-workers found OXY and two OXY mono-glycosides in rat gastrointestinal contents and feces after oral administration of MulA.15 The main forms detected in plasma, bile and urine were OXY and OXY conjugates (glucuronidate and/or sulfate) which accounted for 50% of the oral dose of MulA.14-15 These findings suggest rapid conversion of MulA to OXY in the rat gastrointestinal (GI) tract after oral dosing, which may enable OXY the main form that been absorbed and ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 4 of 33
subjected to extensive hepatic conversions before entering the systemic circulation to elicit activities. Thus, three steps which include intestinal bacterial conversion, trans-epithelial permeability and hepatic biotransformation might be main contributing factors for respective contribution of MulA and OXY to actions of mulberry, yet remain to be addressed. In addition, so far, the biological activities and in vivo pharmacokinetic profiles of MulA and OXY have been characterized with rats, whether and to what extent the species difference in gut microbiota16-17 and hepatic drug-metabolizing enzymes18-19 lead to distinct alterations of these compounds could not be estimated. Therefore, the present study aimed to characterize intestinal bacterial degradation, permeability across intestinal epithelia and hepatic biotransformation of MulA and OXY on in vitro models, and reveal species difference between humans and rats in their biotransformation if any. MATERIALS AND METHODS
Materials. All chemicals were from Sigma-Aldrich (St. Louis, MO) unless otherwise
claimed. Mulberroside A and oxyresveratrol (purities 98%) were supplied by Kuiqing Co., Ltd. (Tianjin, China). Kaempferol (purity 98%) was from National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China) and utilized as an internal standard (IS). BBLTM Brain Heart Infusion (BHI) medium, GasPakTM EZ anaerobe container system with indicator and GasPak™ EZ large incubation container were purchased from Becton Dickinson (Franklin Lakes, NJ). L-Cystine was from Research Organics, Inc. (Cleveland, Ohio). Trishydroxymethylaminomethane was obtained from USB corporation (Cleveland, OH). Acetic acid and trichloroacetic acid were analytical grade from UNI-CHEM Chemical Reagent Co., Ltd., and Damao Co., Ltd. (Tianjin, China), respectively. Methanol, n- ACS Paragon Plus Environment
Page 5 of 33
Journal of Agricultural and Food Chemistry
butanol and formic acid were HPLC-grade from Merck (Darmstadt, Germany). Dulbecco's Modified Eagle's medium (DMEM), fetal bovine serum (FBS), 0.25% trypsin-EDTA, penicillin-streptomycin solution and non-essential amino acids were purchased from GibcoBRL Life & Technologies (Grand Island, NY). Deionized water was obtained using a Milli-Q purification system (Millipore, Bedford, MA). GF/F glass fiber filters were obtained from Whatman (Brentford, UK). 6-Well transwell® plates (0.4 m pore size, 4.71 cm2, polycarbonate filter) were supplied by Corning Costar Co (Cambridge, MA). Caco-2 cells were obtained from the American Type Culture Collection (Rockville, MD). Pooled human liver microsomes (HLMs) and S9 were purchased from BD Biosciences (Billerica, MA). A pool of rat liver S9 and microsomes (RLMs) was prepared from 30 healthy Sprague-Dawley rats (male, 250–300 g) by differential centrifugation according to a standard procedure. The content of microsomal proteins was determined by Lowry's method.20 Both HLMs and RLMs were stored at -80°C until use. Preparation of human and rat intestinal bacteria. The culture medium was 100 mL
autoclaved BHI medium (3.7 g/100 mL) supplemented with 0.05 mg vitamin K1, 0.5 mg hemin bovine and 50 mg L-cystine. Fresh human fecal samples were collected from 6 healthy Chinese volunteers (20-30 year, 2 males and 4 females) from the Institute of Chinese Medical Sciences, University of Macau, and 1 gram of each was pooled and mixed with 60 mL of culture medium. The resultant fecal suspension was centrifuged at 200×g for 5 min. The precipitate was re-suspended and centrifuged at 5,000×g for 30 min. After decanting the supernatant, the precipitate was re-suspended with BHI medium to produce human intestinal microflora solution (0.5 mg mg⋅mL-1 bacteria). Fresh rat fecal samples were collected from 8 male Sprague-Dawley rats (200-250 g) and 0.6 gram of each was pooled together and treated as described above to obtain rat intestinal microflora solution (0.5 mg⋅mL-1 bacteria). All fecal samples were processed according to a protocol approved by the internal ethical committee of ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 6 of 33
the Instititue of Chinese Medical Sciences, University of Macau. Metabolism of MulA by human and rat intestinal bacteria. The incubation system
contains 25 L of intestinal microflora solution, 5 L of MulA in DMSO and BHI medium in a total volume of 500 L (MulA final concentration: 100 M). The reaction system was anaerobically incubated at 37°C in a GasPakTM EZ Anaerobe Pouch system for different time intervals (0, 0.17, 0.33, 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 3.5, and 4 h). Zero-minute incubations or parallel reactions without microflora or MulA served as controls. At the end of reactions, 5 L of the internal standard kaempferol (1 mg⋅mL-1 in DMSO) was added and thorough vortexed to mix, followed by immediate addition of 1 mL of ice-cold water saturated 1-butanol. The resultant mixture was centrifuged at 15,000×g for 10 min and the supernatant evaporated under N2 at 37°C. The residue was reconstituted with 100 L of methanol, then an aliquot (10 L) was subjected to HPLC analysis. Acidic hydrolysis of MulA. MulA (10 mM in DMSO) was dissolved in deionized water
containing 5% (v/v) 1 M HCl (MulA final concentration: 100 M). The mixed solution was kept at 90°C for 6 h, and then extracted once with 1mL water saturated 1-butanol. The organic layer was collected and evaporated under N2 at 37°C. The residue was processed and analyzed in the same manner as described above. Cell culture. Caco-2 cells at passages from 30 to 40 were used for the experiment. The
cells were seeded on 12-well plates and cultured under conditions as previously reported.21 The integrity of the monolayer was monitored by measuring the transepithelial electrical resistance (TEER) at 37°C with an epithelial voltohmmeter (World Precision instruments, Inc., FL) before and after the transport study. Only Caco-2 monolayers with TEER above 300 cm2 were adopted in the transport study. The functionality of Caco-2 monolayers were validated by measuring Papp values of the paracellular marker lucifer yellow, the transcellular marker propranolol and the P-glycoprotein (P-gp) probe substrate rhodamine 123. Lucifer ACS Paragon Plus Environment
Page 7 of 33
Journal of Agricultural and Food Chemistry
yellow exhibited Papp values of 0.183 ± 0.099×10-6 cm.s-1 and propranolol 12.25 ± 2.590×10-6 cm.s-1. Moreover, rhodamine 123 exhibited substantial polarized transport and the efflux ratio was 35.7. Thus, the Caco-2 monolayer possesses normal paracellular, transcellular and efflux transport functions. Sulforhodamine B toxicity assay. Sulforhodamine B assay22 was performed to determine
cytotoxicity of MulA and OXY. In view of the polyphenolic properties of MulA and OXY, their stability in HBSS buffer were examined at pH 7.4, 6.8 and 6.0 for 2 h. MulA showed high stability at pH 7.4 yet OXY was pH-sensitive and only kept intact (>95% of added amount) within 2 h when the pH of the medium was lowered to 6.0. Thus, the cytotoxicity assay and the transport study were carried out in HBSS buffer at pH7.4 for MulA and pH 6.0 for OXY, respectively. Caco-2 cells were seeded at a density of 1×104/well into 96-well plates and cultured as described above. After 24 h incubation, the medium was replaced with 200 L of HBSS buffer (control), HBSS containing different amounts of MulA (final concentrations 3-200 µM) at pH 7.4 or OXY (final concentrations 6-400 µM) at pH 6.0. After exposure to test compound or HBSS alone at 37°C for 4 h, the cells were fixed with ice-cold trichloroacetic acid (final concentration 10%, v/v) at 4°C for 1h. Then the cell plates were washed with deionized water and allowed to dry at ambient temperature. Afterwards, the cells were stained with sulforhodamine B solution (4 mg⋅mL-1, 100 L/well) for 30 min at room temperature, then rinsed with 1% acetic acid for three times to remove unbound dye and air- dried. Sulforhodamine B formed was solubilized with 200 µL of trishydroxymethyl- aminomethane (100 mM) and the absorbance measured at a wavelength of 490 nm. Transport studies of MulA and OXY. After 21-day culture, the Caco-2 monolayers were
rinsed twice with HBSS and pre-incubated in HBSS at 37°C for 30 min. In the absorptive transport study, 0.5 mL of HBSS solution containing test compound was loaded at the apical (A) side (donor chamber) and 1.5 mL of blank HBSS placed at the basolateral (B) side ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 8 of 33
(receiver chamber). In the secretory transport study, 1.5 mL of the HBSS containing test compound was added at the B side (donor chamber) and 0.5mL of blank HBSS was placed at the A side (receiver chamber). Aliquots (100 L) were taken from the receiver chamber at appropriate time intervals (MulA: 0, 1, 1.5, 2, 2.5 h; OXY: 0, 0.25, 0.5, 0.75, 1, 1.25 h). After each sampling, 100 L of HBSS was supplemented to the receiver chamber to maintain a constant volume. At the end of the experiment, samples were also withdrawn from the donor chamber to calculate the recovery. Samples collected from the MulA study were directly injected into an HPLC instrument. Due to the instability of OXY in HBSS, samples collected from the transport study of OXY were mixed with 2 L of kaempferol (0.2 mg⋅mL-1 in DMSO), extracted once with 1 mL of ice-cold 1-butanol, then centrifuged at 5,000×g for 5 min. The organic layer was collected, evaporated under N2 at 37°C and the residue reconstituted with 100 L of methanol/water (50/50, v/v) before subjecting to HPLC analysis. The involvement of efflux transporters in OXY transport was determined by adding either of the following inhibitors in both sides and incubated with the cell monolayers for 30 min at 37°C before adding OXY: verapamil (the P-gp inhibitor) 100 µM, indomethacin (the multidrug resistance-associated proteins (MRPs) inhibitor) 10 µM. To determine the extent of cell accumulation of test compound, Caco-2 monolayers were removed from transwell plates at the end of the transport study, rinsed twice with deionized water, and then sonicated in methanol for 30 min. The resultant solution was evaporated under N2 and the residue reconstituted with 100 L of HBSS prior to HPLC analysis. Calibration curves of MulA and OXY for transport studies. MulA and OXY stock
solutions were prepared and diluted to different concentrations with DMSO. The serial solutions of MulA were diluted with HBSS at pH 7.4 (DMSO 1%, v/v) and applied to HPLC analysis directly. Calibration curves of MulA were constructed by plotting the peak area of MulA as a function of MulA concentration. The serial solutions of OXY were diluted with ACS Paragon Plus Environment
Page 9 of 33
Journal of Agricultural and Food Chemistry
HBSS at pH 6.0 (DMSO 1%, v/v). These samples were then mixed with kaempferol and processed as described above. Calibration curves of OXY were constructed by plotting the peak area ratio of OXY to the internal standard as a function of concentration. Metabolic studies of MulA and OXY in human and rat hepatic subcellular fractions.
Metabolic profiles of MulA and OXY were characterized with human and rat liver subcellular fractions in a total of 200 L reaction solution containing MulA or OXY (final concentration:100 M) under the following different conditions: Phase I reaction: 0.8 mg/mL HLMs or RLMs, NADPH-regenerating system (4 mM MgCl2, 1 mM NADP+, 1 mM glucose-6-phosphate, and 1 U⋅mL-1 glucose-6-phosphate dehydrogenase (G-6-PD) in 100 mM potassium phosphate buffer (pH 7.4); Glucuronidation: 0.8 mg⋅mL-1 HLMs or RLMs, 8 mM MgCl2, 2 mM uridine diphosphate glucuronic acid (UDPGA) and 25 g alamethicin in 50 mM Tris-HCl (pH 7.4); Sulfation: 0.8 mg⋅mL-1 S9, 5 mM MgCl2, 8 mM DTT, 0.0625% BSA, and 200 M 3'- phosphoadenosine 5'-phosphosulfate (PAPS) in 50 mM Tris-HCl (pH 7.4). All reactions were pre-incubated at 37°C for 5 min and initiated by adding the cofactor (G-6-PD, UDPGA or PAPS, respectively) and kept at 37°C for 10 min (glucuronidation) or 1 h (phase I and sulfation). Reactions without the cofactors served as controls. All the experiments were performed in duplicate. Reactions were terminated by adding 100 L ice cold methanol. After centrifugation at 15,000×g for 10 min at 4°C, the supernatants were subjected to HPLC or HPLC-MS analysis. HPLC Analysis. All samples were analyzed on an Agilent series 1200 (Agilent
Technologies, Palo Alto, CA) liquid chromatography that equipped with a vacuum degasser, a binary pump, an autosampler and a diode array detector (DAD) system, and operated with an Agilent ChemStation B 3.0 software. An Alltech Alltima C18 column (250 mm × 4.6 mm, 5 m) was used for separation at 25oC. The mobile phase consisted of 0.1% aqueous formic ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 10 of 33
acid (A) and methanol (B) and eluted at 1 mL/min with a linear gradient increasing from 5% B to 100% B within 38 min. The UV spectra were obtained over 210-380 nm. MulA, OXY and other metabolites were monitored at 320 nm. The injection volume was 10 L of samples from intestinal bacterial studies and 70 L of samples from the transport study or hepatic metabolic study. HPLC-MS Analysis. Mass spectrometry was performed on an LC/MSD Ion Trap system
(Agilent Technologies, Palo Alto, CA) consisting of an HP 1100 series binary pump HPLC system and an ion-trap mass spectrometer with electrospray ionization interface, and operated with an Agilent ChemStation software. HPLC separation was conducted as described above. ESI-MS analyzes were operated in negative ion mode under following conditions: dry gas N2 8 L/min, dry temperature 350°C, nebulizer pressure 40 psi, capillary voltage 3500 V, scan range 50–600 m/z. ESI-MS/MS conditions were as follows: negative ion mode, separation width 4, fragment amplification 1.5, scan range 50–600 m/z. NMR Analysis of G4. G4 was prepared from a scale-up reaction with HLMs and isolated
using preparative HPLC. NMR spectra (1H, 13C, HSQC, HMBC, ROESY and NOESY) of G4 were recorded on a Bruker AV-600 (Bruker, Newark, Germany), using TMS as the internal standard. Chemical shifts were expressed in and coupling constants (J) were reported in Data Analysis. The apparent permeability coefficients (Papp) of test compound from
apical side to basolateral side (PappA B) or from basolateral side to apical side (PappB A) in the bidirectional transport study were calculated using the following equation: Papp = (dC/dt × V) / (C0 × A) where dC/dt is the rate of the test compound appearing in the receiver chamber, V is the volume of the solution in the receiver chamber, C0 is the initial concentration of the compound added in the donor chamber, and A is the cell monolayer surface area. An efflux ACS Paragon Plus Environment
Page 11 of 33
Journal of Agricultural and Food Chemistry
ratio (PappB A /PappA B) > 2 was adopted when determining whether efflux transporter(s) was The area under peak area ratio-time curve (AUP) of MulA metabolites formed from intestinal bacterial incubations were calculated by linear trapezoidal method using WinNonlin 5.2.1 software (Pharsight, USA) and used for comparison of the relative amounts between compounds within the same species or the same metabolite between humans and rats. All data were expressed as mean ± standard deviation (S.D.). The apparent permeability coefficient (Papp) values of test compound were compared between absorptive direction (PappA B) and secretary direction (PappB A) and between inhibitor treated and untreated groups using unpaired Student's t-test. A P < 0.05 value was deemed significant for all tests. Metabolism of MulA by human and rat intestinal bacteria. When incubated
anaerobically with human intestinal bacteria, MulA decreased rapidly with time and was not detectable at 3 h after incubation. Three additional peaks namely H1, H2, and H3 were eluted at retention times of 17.9 min (H1), 18.4 min (H2) and 21.7 min (H3), sequentially and confirmed to be metabolites of MulA when compared with controls (Fig. 2 A). The MS data of MulA and OXY standards and H1-H3 were shown in Table 2 and Fig. 3, MS1 of MulA showed a pseudo molecular ion ([M-H]-) at m/z 567, which produced product ions at m/z 405 and m/z 243 in the MS2 spectrum, corresponding to the loss of one and two glucose molecules from MulA, respectively. OXY had a [M-H]- ion at m/z 243, and generated characteristic ions at m/z 225 ([M-H-H2O]-), m/z 199 ([M-H-CH2=CH-OH]-), m/z 185, m/z 5, and m/z 97 at comparable abundance in its MS2 spectrum. MS1 spectra of both H1 and H2 showed a pseudo-molecular ion ([M-H]-) at m/z 405, which was 162 mass units less than that of MulA (m/z 567), corresponding to the loss of one ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 12 of 33
glucose molecule. The MS2 of m/z 405 of both metabolites yielded the predominant fragment ion at m/z 243, indicating the loss of the other glucose moiety. Thus, H1 and H2 should be monoglucosides of OXY formed from deglycosylation of MulA at C-3 and C-4 respectively. H3 showed identical retention time, UV and mass spectra to those of OXY, thus was unambiguously identified as OXY from sequential loss of two glucoses from MulA. Similarly, MulA also showed a rapid elimination in rat intestinal bacteria and gave three metabolites namely R1, R2, and R3. The retention times, UV and MS spectra of R1-R3 were identical to those of H1-H3, respectively. Thus, R1 and R2 were tentatively identified as oxyresveratrol-4-O- -D-glucopyranoside and oxyresveratrol-3 -O- -D-glucopyranoside and Acidic hydrolysis of MulA. After acidic treatment of MulA, three peaks (A1, A2, and A3)
were observed along with a decrease of MulA. The three products showed identical retention times, UV and MS spectra to those of H1-H3 and R1-R3, respectively (Fig. 2B, Table 2). Thus, the products after acid hydrolysis of MulA were assigned as the same 3 metabolites of MulA formed by intestinal bacteria, i.e., A1 is R1/H1, A2 is R2/H2, and A3 is OXY. Time-course of MulA metabolism by human and rat intestinal bacteria. As shown in
Fig. 4, MulA exhibited a rapid linear elimination (y = -1.36x + 3.76, r2 = 0.98) and disappeared within 3 h with an average velocity of elimination at 0.016 mol⋅h-1 when incubated with human intestinal bacteria (Fig. 4A). The two monoglucosides of OXY, H1 and H2, increased with time, reached their maximum at 1.5-2 h, and then were not detectable within 1 h. The AUP value, which was calculated on basis of peak area ratio of each metabolite to the internal standard over 4 h of incubation, of H2 was around 2 times that of H1 (1.89 ± 0.15 versus 1.00 ± 0.43). The increase of OXY was slow within the first 1.5 h, and then increased much faster before reaching a plateau in another hour. In the case of rat intestinal bacteria, metabolic conversion of MulA occurred linearly yet ACS Paragon Plus Environment
Page 13 of 33
Journal of Agricultural and Food Chemistry
slightly slower (y=-1.07x + 3.23, r2 = 0.94) than that by human intestinal bacteria (Fig. 4B). It took 4 h to deplete MulA and the overall elimination rate was 0.013 mol⋅h-1. The rates of R1 and R2 increase within first 2 h were similar to those of H1 and H2 increase, respectively, yet the elimination of R1 and R2 was slightly slower, suggesting a slower deglycosylation of the OXY monoglycosides by microbiota from rats than those from humans. As a consequence, a slower increase of OXY was observed with the maximum appeared at 4 h. The average velocity of OXY formation by rat was around half of that formed by human as judged based on the slopes of OXY formation curves over 1.5-2.5 h. Similarly, the AUP of R1 (1.52 ± 0.05) was smaller than that of R2 (2.17 ± 0.28). Transport of MulA and OXY across Caco-2 monolayer. Sulforhodamine B toxicity
assays revealed non-toxicity of MulA and OXY towards Caco-2 monolayers within 4 h over respective concentration range tested. Therefore, the transport studies were carried out over 25-200 µM (MulA) and 50-400 µM (OXY). The Papp values of MulA and OXY obtained from bidirectional transport studies are summarized in Table 1. MulA was not detectable at receiver chambers at 25 µM in both directions and OXY was not detectable at receiver chambers at 50 µM and 100 µM in the absorptive direction, thus, the efflux ratios were not available for these concentrations and data not shown in Table 1. MulA exhibited very low Papp values (×10-7 cm⋅s-1) in both directions, indicating a poor oral absorption in human body. The bidirectional Papp values of MulA obtained for different concentrations were similar, indicating a dose-independent transport. Furthermore, efflux ratios were all less than 2, suggesting that passive diffusion dominate MulA transport across Caco-2 monolayers. As has been expected, the aglycone OXY exhibited Papp values (×10-6 cm⋅s-1) that were 1 order of magnitude higher than that of MulA. Different from that observed with MulA, OXY ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 14 of 33
exhibited higher transport in the secretory direction than the absorptive direction over the concentration range tested (Table 1). Moreover, the efflux ratio of OXY was 4.22 at 200 µM and increased inversely with dose, suggesting that OXY might be the substrate of efflux transporter(s). At 200 µM of OXY, the P-gp inhibitor verapamil abolished the directional preference of OXY transport with an efflux ratio of 1.27 and the MRPs inhibitor indomethacin substantially diminished the polarized transport and the efflux ratio was 2.32. Both MulA and OXY showed high recovery from the transport study. No MulA was detected in Caco-2 cells and the amount of OXY trapped in Caco-2 cells was negligible (< 1%). No additional peak was detected during the permeation of OXY or MulA across the Caco-2 cell monolayers. In vitro hepatic metabolism of MulA and OXY. When incubated with liver microsomes
of humans or rats for 1h, the recovery of MulA was >96% in the presence of a NADPH- regenerating system or UDPGA. In incubations with hepatic S9 preparations and PAPS, MulA remained intact and the recovery reached >99%. Further, HPLC-MS analysis also revealed no metabolites of MulA. OXY exhibited high (>95%) recovery from reactions with HLMs or RLMs in absence/presence of a NADPH-regenerating system. HPLC-MS analysis also revealed no additional peaks (data not shown). In contrast, four additional peaks, namely G1, G2, G3 (minor peaks) and G4 (major peak), were observed at the end of the 1-h reaction with HLMs or RLMs as compared to control reactions (Fig. 5A & 5B). There were 90% (HLMs) and 97% (RLMs) of OXY loss over 10 min in presence of UDPGA with the average velocity of OXY elimination 11.4 nmole⋅min-1⋅mg-1 protein and 12.1 nmole⋅min-1⋅mg-1 protein, respectively, indicating an extensive glucuronidation of OXY. The formation of G4 was highest in HLMs, yet G3 was the most predominant metabolite of RLMs. When compared with the peak area values, the amount of G4 formed in HLMs was 7 times that in RLMs and G3 formation in ACS Paragon Plus Environment
Page 15 of 33
Journal of Agricultural and Food Chemistry
RLMs was 20 times that in HLMs. In hepatic S9, a minor sulfation was evidenced by 10% (human) and 15% (rat) of OXY loss over 1 h and the simultaneous appearance of one additional peak, namely S1 (Fig. 5C) in the presence of PAPS. The average velocity of OXY elimination via sulfation was 0.19 (human) versus 0.31 (rat) nmole⋅min-1⋅mg-1 protein. The mass spectra of G1-G4 and S1 were shown in Fig. 3. G1-G4 all exhibited pseudo- molecular ions [M-H]- at m/z 419 in their mass spectra, corresponding to a molecular weight of 420 Da, which was 176 mass units (equal to one glucuronic acid molecule) higher than the molecular weight of OXY ([M-H]- at m/z 243). Thus, the four metabolites were all mono- glucuronidates of OXY. Their relative amounts, as calculated based on the percentage of the peak area of each to the sum of the four, were in the following descending order: Human: G4 (92.7%), G3 (4.0%), G1 (2.3%), G2 (1.0%); Rat: G3 (85.1%), G4 (13.2%), G2 (1.3%), G2 (0.4%). The mass spectrum of S1 revealed a pseudo-molecular ion at m/z 323 and a characteristic fragment ion at m/z 243 ([M-H-SO3]-) (Fig. 3), suggesting a mono-sulfated The 1H-NMR spectrum of G4 showed signals due to trans-olefinic protons at = 7.31 (1H, d, J = 16.5 Hz, H-α) and 6.80 (1H, d, J = 16.5 Hz, H-β), as well as proton signals belonging to two independent aromatic rings (Table 2). By comparison of the NMR spectroscopic data of G4 with those of OXY15, all proton signals in ring B were unaffected, yet those in ring A showed a change of +0.22 ppm and +0.24 ppm, respectively, in the chemical shift of C3-H and C5-H with their JHH coupling of 2.3 Hz. The 13C-NMR spectrum of G4 revealed significant changes in especially the signals of C2 (+1.4 ppm), C1 (+2.6 ppm) and C3 (+0.4 ppm) in comparison with OXY. The presence of a D-glucuronosyl moiety was evidenced by carbon signal at 170.0 ppm due to carboxyl carbon. The β-configuration of the glucuronide linkage was determined from the coupling constant of its anomeric proton at = 4.95 (1H, d, J = 8.0 Hz). The NOESY spectrum showed correlation of the anomeric proton at = 4.95 to ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 16 of 33
H-3, corroborating the linkage of glucuronide at C-2 of OXY. All together these NMR data unambiguously identify G4 as oxyresveratrol-2-O- -D-glucuronosyl. DISCUSSION
In daily life, most herbs or herbal products are used as food supplements and in formats of oral preparations. In the long history of traditional medicinal practice, herbal medicines are usually prepared as decoctions and taken orally in clinical use. Thus, when study nutritional or medicinal herbs, emphasis should be put on the constituents in the water extract and their fates prior to entering systemic circulation to understand their in vivo benefits. As two main bioactive constituents in the water extract of mulberry root bark the existing pharmacokinetic data 10,14-15 indicate the occurrence of conversion of MulA to OXY via deglycosylation in gut lumen, which might be a determinant step for subsequent absorption and disposition of MulA and OXY, and consequently their respective contribution to the benefits of mulberry. Moreover, the existing in vivo pharmacokinetic data indicated occurrence of phase II conjugation of OXY when a Mori Cortex extract or MulA was administered orally to the rat 10,14,29. In consideration of the richness of MulA and OXY in mulberry and an increasing awareness of the benefits of OXY, it is of great importance to understand the determinant factors for their systemic exposure. Thus, in the present study, 3 steps that might be crucial for systematic exposure of MulA and OXY, including biotransformation by intestinal bacteria, transepithelial permeability and hepatic metabolism, were examined to estimate their respective contribution to the benefits of mulberry. To our best knowledge, this is the first report on in vitro bacterial and hepatic metabolism and permeability evaluation of MulA and OXY. The results demonstrated a rapid hydrolysis of MulA to generate OXY in both human and rat intestinal bacteria, a markedly higher ACS Paragon Plus Environment
Page 17 of 33
Journal of Agricultural and Food Chemistry
permeability of OXY than MulA, and an extensive hepatic conversions of OXY. Hepatic conjugation of OXY, not intestinal bacterial degradation of MulA, may lead to a significant species difference in their systemic exposures. As expected, MulA underwent stepwise deglycosylation in intestinal bacteria of humans and rats. Formation of the aglycone oxyresveratrol was rapid and its structure was unambiguously confirmed with standard OXY. The other two metabolites were tentatively determined based on the current MS data, and their structures were speculated as C-3´ and C- 4 mono-glycosides of OXY that were formed from removal of one glucose molecule from C-4 and C-3´of MulA, respectively. Our speculation on the structures of the two monoglycosides was supported by an acidic hydrolytic study of MulA which yielded, in addition to OXY, the same two monoglycosides as evidenced by identical UV spectra, MS spectra and retention times to those determined in in vitro intestinal bacteria incubation. The structures of the two metabolites were speculated as oxyresveratrol-3´-O-β-D-glucopyranoside and oxyresveratrol- 4-O-β-D-glucopyranoside, respectively. Zhaxi's group also reported two monoglycosides of OXY in rat feces and gut contents after an oral dose of MulA, yet the structures were determined to be oxyresveratrol-3´-O-β-D-glucopyranoside and oxyresveratrol-2-O- -D- glucopyranoside.15 The discrepancy between our speculation from in vitro incubation and the previous in vivo study warrants further study. So far, most of the biological evaluations of MulA and OXY were performed on rats. Whether the species difference in the composition and hence the metabolic activity of intestinal microflora between humans and rats16-17 significantly affect MulA conversion to OXY in the gut lumen remains to be addressed. In the present study, incubations of MulA with human and rat intestinal bacteria produced 3 identical metabolites. Time courses of the parent compound depletion revealed a rapid elimination of MulA in both species with slightly quicker elimination in humans. The insignificant species difference between human and rat in ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 18 of 33
intestinal bacterial conversion of MulA warrants the extrapolation of existing pharmacokinetic data obtained on rats to humans. The AUP values of the two monoglucoside of OXY indicated a slight regio-selectivity in intestinal hydrolysis of C-4 and C-3´ glucosides of MulA although the position preference could not be identified in the present study. The two monoglycoside intermediates disappeared from the incubation system rapidly and their maximum levels (compared based on peak area data) were all less than 25% of that of OXY. Whether and how the two monoglucosides contribute to the health benefits of mulberry are unknown and warrant further study. In concert with a quicker elimination of MulA in human intestinal bacteria, OXY reached a plateau faster in human intestinal bacteria (2.5 h vs 4 h with rat). Taken together, we can conclude that deglycosylation of MulA to form OXY by gut microbiota is rapid and may be of considerable importance in determining their systemic exposures and consequently respective contribution to benefits of mulberry. It has been reported that some phenolic aglycones permeated across the apical side of Caco-2 cell monolayers more easily than their glucosides.23 In the present study, the bidirectional transport study on the Caco-2 model revealed low Papp values (×10-7 cm⋅s-1) of MulA and high Papp (×10-6 cm⋅s-1) of OXY, predicting a poor absorption (<20%) of MulA and a medium absorption (20% 70%) of OXY.24 This much higher permeability of OXY, together with a rapid bacterial hydrolysis of MulA that demonstrated in vitro, enables OXY as the main form that been absorbed after oral dosing of MulA or mulberry. This finding is in good agreement with the previous report that oral bioavailability of MulA in the rat was about 1% and OXY and its derivatives accounted for 50% of oral dose of MulA.14 The dose- independent permeation and comparable absorptive and secretory transports of MulA indicate a predominant passive diffusion, whereas P-gp and MRP(s) were demonstrated to be involved in OXY transport across Caco-2 cells (Table 1). The efflux-mediated property of OXY transport should arouse attentions on potential interactions during dietary or medicinal uses of ACS Paragon Plus Environment
Page 19 of 33
Journal of Agricultural and Food Chemistry
OXY or mulberry-derived products. Interestingly, resveratrol, the polyhydroxystilbene that lack of the hydroxyl group at C-2 position when compared to OXY, is a substrate of MRP2,25 but not that of MRP1 and P-glycoprotein26. These findings suggest that the difference in number and position of hydroxyl substitution might be crucial for transepithelial permeability of trans-hydroxystilbenes. Moreover, although there was considerable resveratrol accumulation in Caco-2 cells,27 the accumulation of OXY was insignificant. This might explain the absence of conjugation (glucuronidation or sulfation) of OXY although conjugation of rasveratrol (main sulfation and minor glucuronidation) was observed during their transport across Caco-2 cells.28 Liver plays a crucial role in biotransformation of many xenobiotics in humans and rats. In the present study, MulA was unaltered in human liver preparations. In contrast, OXY underwent extensive phase II conversions (both glucuronidation and sulfation), predicting a significant hepatic first-pass effect. These in vitro data were in line with previous in vivo findings that glucuronidated and sulfated OXY were mainly identified in rat plasma, urine and bile samples after oral adminstration of MulA 14-15 or OXY 29, and thereby warrant further investigational emphasis on contributions of OXY and its metabolites to mulberry. More interestingly, there were four OXY mono-glucuronidates formed in both HLMs and RLMs, corresponding to the four hydroxyl groups in OXY, but only one mono-sulfated product of OXY was detected in hepatic S9 of both species. Rat hepatic subcellular preparations exhibited slightly higher activities for OXY glucuronidation and sulfation than those of human. It was noteworthy that there was apparent positional preference with hepatic glucuronidation of OXY which varied with species. The major product G4 of HLMs was firstly identified as 2-glucuronosyl of OXY and accounted for around 93% of total OXY glucuronides formed by HLMs. Similar regioselective glucuronidation in HLMs has been identified with resveratrol.30 The predominant metabolite G3 in RLMs accounted for about ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 20 of 33
85% of total glucuronides formed by RLMs, yet the absolute structure was unable to be identified in the present study due to the unavailability of the standard compound. Reaction phenotyping study and biological evaluation of G4 are ongoing in our laboratory. AUTHOR INFORMATION
Corresponding Author
*Tel : 853-83974876. Fax: 853-28841358. E-mail: [email protected]. Funding Sources
The present work is supported by the National Basic Research Program of China (973 program, 2009CB522707) and the Research Committee of the University of Macau (RG086/09-10S/YR/ICMS and MYRG162(Y1-L2)-ICMS11-YR). ABBREVIATIONS USED
MulA, mulberroside A; OXY, oxyresveratrol; UV, ultraviolet; MS, mass spectrometry; NMR, nuclear magnetic resonance; NOESY, Nuclear Overhauser effect spectroscopy.
ACS Paragon Plus Environment
Page 21 of 33
Journal of Agricultural and Food Chemistry
LITERATURE CITED
(1) The state pharmacopoeia commission of P. R. China. (2010). Pharmacopoeia of people's republic of China. Beijing, China: Chemical industry press. (2) Naowaboot, J.; Pannangpetch, P.; Kukongviriyapan, V.; Kongyingyoes, B.; Kukongviriyapan, U. Antihyperglycemic, antioxidant and antiglycation activities of mulberry leaf extract in streptozotocin-induced chronic diabetic rats. Plant. Foods. Hum. Nutr. 2009, 64,
(3) Lee, Y. J.; Choi, D. H.; Kim, E. J.; Kim, H. Y.; Kwon, T. O.; Kang, D. G.; Lee, H. S. Hypotensive, hypolipidemic, and vascular protective effects of Morus alba L. in rats fed an atherogenic diet. Am. J. Chin. Med. 2011, 39, 39-52.
(4) Chang, L. W.; Juang, L. J.; Wang, B. S.; Wang, M. Y.; Tai, H. M.; Hung, W. J.; Chen, Y. J.; Huang, M. H. Antioxidant and antityrosinase activity of mulberry (Morus alba L.) twigs and root bark. Food Chem. Toxicol. 2011, 49, 785-90.
(5) Katsube, T.; Yamasaki, M.; Shiwaku, K.; Ishijima, T.; Matsumoto, I.;Abe, K.; Yamasaki, Y. Effect of flavonol glycoside in mulberry (Morus alba L.) leaf on glucose metabolism and oxidative stress in liver in diet-induced obese mice. J. Sci. Food. Agr. 2010,
90, 2386-92. (6) Nade, V. S.; Kawale, L. A.; Yadav, A.V. Protective effect of Morus alba leaves on haloperidol-induced orofacial dyskinesia and oxidative stress. Pharm. Biol. 2010, 48(1), 17-
(7) Asano, N.; Yamashita, T.; Yasuda, K.; Ikeda, K.; Kizu, H.; Kameda, Y.; Kato, A.; Nash, R. J. Polyhydroxylated alkaloids isolated from mulberry trees (Morus alba L.) and silkworms (Bombyx mori L.). J. Agr. Food Chem. 2001, 49, 4208-13.
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 22 of 33
(8) Piao, S. J.; Qu, G. X.; Qiu, F. Chemical constituents of the water extract from Mori Cortex. Zhongguo Yaowu Huaxue Zazhi. 2006, 16, 40–45.
(9) Wang, C. P.; Wang, Y.; Wang, X.; Zhang, X.; Ye, J. F.; Hu, L.S.; & Kong, L. D. Mulberroside A possesses potent uricosuric and nephroprotective effects in hyperuricemic mice. Planta Med. 2011, 77, 786-794.
(10) Kim, J. K.; Kim, M.; Cho, S. G.; Kim, M. K.; Kim, S. W.; Lim, Y. H. Biotransformation of mulberroside A from Morus alba results in enhancement of tyrosinase inhibition. J. Ind. Microbiol. Biot. 2010, 37, 631-7.
(11) Chao, J.; Yu, M. S.; Ho, Y. S.; Wang, M.; Chang, R. C. Dietary oxyresveratrol prevents parkinsonian mimetic 6-hydroxydopamine neurotoxicity. Free Radical Bio. Med. 2008, 45, 1019-26.
(12) Lorenz, P.; Roychowdhury, S.; Engelmann, M.; Wolf, G.; Horn, T. F. Oxyresveratrol and resveratrol are potent antioxidants and free radical scavengers: effect on nitrosative and oxidative stress derived from microglial cells. Nitric. Oxide–Biol. Ch. 2003, 9, 64-76.
(13) Kim, Y. M.; Yun, J.; Lee, C. K.; Lee, H.; Min, K. R.; Kim, Y. Oxyresveratrol and hydroxystilbene compounds. Inhibitory effect on tyrosinase and mechanism of action. J. Biol. Chem. 2002, 277, 16340-4.
(14) Qiu, F.; Komatsu, K.; Saito, K.; Kawasaki, K.; Yao, X.; Kano, Y. Pharmacological properties of traditional medicines. XXII. Pharmacokinetic study of mulberroside A and its metabolites in rat. Biol. Pharm. Bull. 1996, 19, 1463-7.
(15) Zhaxi, M.; Chen, L.; Li, X.; Komatsu, K.; Yao, X.; Qiu, F. Three major metabolites of mulberroside a in rat intestinal contents and feces. Planta Med. 2010, 76, 362-4.
(16) Wang, R. F.; Cao, W. W.; Cerniglia C. E. PCR detection and quantitation of predominant anaerobic bacteria in human and animal fecal samples. Appl. Environ. Microb. 1996, 62, 1242-7.
ACS Paragon Plus Environment
Page 23 of 33
Journal of Agricultural and Food Chemistry
(17) Blaut, M.; Clavel, T. Metabolic diversity of the intestinal microbiota: implications for health and disease. J. Nutr. 2007, 137, 751S–755S.
(18) Turpeinen, M.; Ghiciuc, C.; Opritoui, M.; Tursas, L.; Pelkonen, O.; Pasanen, M. Predictive value of animal models for human cytochrome P450 (CYP) mediated metabolism: comparative study in vitro. Xenobiotica. 2007, 37, 1367-1377.
(19) Liang, S. C.; Ge, G. B.; Liu, H. X.; Shang, H. T.; Wei, H.; Fang, Z. Z.; Zhu, L. L.; Mao, Y. X.; Yang, L. Determination of propofol UDP-glucuronosyltransferase (UGT) activities in hepatic microsomes from different species by UFLC–ESI-MS. J. Pharmaceut. Biomed. 2011, 54, 236-241.
(20) Lowry, O. H,; Rosebrough, N. J.; Farr, A. L.; Randall, R. J. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 1951, 193, 265-275.
(21) Ruan, J. Q.; Leong, W. I.; Yan, R.; Wang, Y.T. Characterization of metabolism and in vitro permeability study of notoginsenoside R1from Radix Notoginseng. J. Agr. Food Chem. 2010, 58,5770-5776.
(22) Voigt, W. Sulforhodamine B assay and chemosensitivity. Methods in Molecular Medicine. 2005, 110, 39-48.
(23) Liu, Y.; Hu, M. Absorption and metabolism of flavonoids in the caco-2 cell culture model and a perused rat intestinal model. Drug Metab. Dispos. 2002, 30, 370-7.
(24) Artursson, P.; Karlsson, J. Correlation between oral drug absorption in humans and apparent drug permeability coefficients in human intestinal epithelial (Caco-2) cells. Biochem. Bioph. Res. Co. 1991, 175, 880-885.
(25) Henry, C.; Vitrac, X.; Decendit, A.; Ennamany, R.; Krisa, S.; Mérillon, J. M. Cellular uptake and efflux of trans-piceid and its aglycone trans-resveratrol on the apical membrane of human intestinal Caco-2 cells. J. Agr. Food Chem. 2005, 53, 798-803.
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 24 of 33
(26) Li, Y.; Revalde, J. L.; Reid, G.; Paxton, J. W. Interactions of dietary phytochemicals with ABC transporters: possible implications for drug disposition and multidrug resistance in cancer. Drug Metab. Rev. 2010, 42, 590-611.
(27) Maier-Salamon, A.; Hagenauer, B.; Wirth, M.; Gabor, F.; Szekeres, T.; Jäger, W. Increased transport of resveratrol across monolayers of the human intestinal Caco-2 cells is mediated by inhibition and saturation of metabolites. Pharm. Res-Dordr. 2006, 23, 2107-15.
(28) Sabolovic, N.; Humbert, A. C.; Radominska-Pandya, A.; Magdalou, J. Resveratrol is efficiently glucuronidated by UDP-glucuronosyltransferases in the human gastrointestinal tract and in Caco-2 cells. Biopharm. Drug Dispos. 2006, 27, 181-9.
(29) Huang, H.; Chen, G.; Lu, Z.; Zhang, J.; Guo, D.A. Identification of seven metabolites of oxyresveratrol in rat urine and bile using liquid chromatography/tandem mass spectrometry. Biomed. Chromatogr. 2010, 24, 426-32.
(30) Aumont, V.; Krisa, S.; Battaglia, E.; Netter, P.; Richard, T.;Mérillon, J. M.; Magdalou, J.; Sabolovic, N. Regioselective and stereospecific glucuronidation of trans- and cis-resveratrol in human. Arch. Biochem. Biophys. 2001, 393, 281-9.
ACS Paragon Plus Environment
Page 25 of 33
Journal of Agricultural and Food Chemistry
FIGURE CAPTIONS
FIGURE 1 Chemical structures of mulberroside A and oxyresveratrol, and their proposed
metabolic changes in humans. FIGURE 2 Typical HPLC chromatograms of incubations of mulberroside A with human
intestinal bacteria (A) and HCl (B). a, reaction with bacteria or HCl; b, control reaction without bacteria or HCl. FIGURE 3 MS1 spectra of mulberroside A, oxyresveratrol, metabolites (H1-H3) of
mulberroside A formed by human intestinal bacteria, and hepatic metabolites (G1-G4, S1) of FIGURE 4 Time course of mulberroside A metabolism by human (A) and rat (B) intestinal
bacteria. Values are means FIGURE 5 Typical HPLC chromatograms of incubations of oxyresveratrol with liver
microsomes of human (A) and rat (B) for 10 min and human S9 (C) for 60 min. a, reaction in presence of co-factor; b, reaction without co-factor. *: impurity in the incubation system ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 26 of 33
Table 1 Apparent pearmeability coefficients (Papp) of mulberroside A and oxyresveratrol at
different concentrations on a Caco-2 cell model. Papp A B, ×10-7 Papp B A, ×10-7 Concentration, M (verapamil 100 µM) (indomethacin 10 µM) **p < 0.01 compared with the corresponding Papp A B value; p<0.05, ## p<0.01 compared with the corresponding Papp A B or Papp B A value for 200 µM of oxyresveratrol in absence of inhibitors. ACS Paragon Plus Environment
Page 27 of 33
Journal of Agricultural and Food Chemistry
Table 2 1H and 13C NMR spectroscopic data for metabolite G4 and oxyresveratrol.
6.53 (1H, d, J=2.3 Hz) 6.31 (1H, d, J=2.3 Hz) 6.47 (1H, dd, J=8.6, 2.3 Hz) 107.2 6.23 (1H, dd, J=8.4, 2.3 Hz) 7.47 (1H, d, J=8.5 Hz) 7.33 (1H, d, J=8.4 Hz) 7.31 (1H, d, J=16.5 Hz) 7.14 (1H, d, J=16.4 Hz) 6.80 (1H, d, J=16.5 Hz) 6.76 (1H, d, J=16.4 Hz) 6.39 (1H, d, J=2.0 Hz) 6.33 (1H, d, J=2.0 Hz) 6.10 (1H, d, J=2.0 Hz) 6.06 (1H, t, J=2.0 Hz) 6.39 (1H, d, J=2.0 Hz) 6.33 (1H, d, J=2.0 Hz) 4.95 (1H, d, J=8.0 Hz) ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 28 of 33
1Measured in DMSO-d6. 2Cited from Zhaxi et al. (2010). 3Glu is glucuronosyl moiety. ACS Paragon Plus Environment
Page 29 of 33
Journal of Agricultural and Food Chemistry
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 30 of 33
ACS Paragon Plus Environment
Page 31 of 33
Journal of Agricultural and Food Chemistry
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 32 of 33
ACS Paragon Plus Environment
Page 33 of 33
Journal of Agricultural and Food Chemistry
ACS Paragon Plus Environment

Source: http://repository.umac.mo/dspace/bitstream/10692/1348/1/7257_0_jf_submitted%2022-7-2011.pdf