Food Chemistry 212 (2016) 712–721
Contents lists available at ScienceDirect
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
Analytical Methods Preparative separation of grape skin polyphenols by high-speed counter-current chromatography ⇑ Lanxin Luo a, Yan Cui b, Shuting Zhang c,d, Lingxi Li b, Yuanyuan Li c, Peiyu Zhou c, Baoshan Sun b,e, a School of Traditional Chinese Materia Medica, Shenyang Pharmaceutical University, 110016 Shenyang, China b School of Functional Food and Wine, Shenyang Pharmaceutical University, 110016 Shenyang, China c School of Pharmacy, Shenyang Pharmaceutical University, 110016 Shenyang, China d Department of Pharmacy, Jiamusi University, 154007 Jiamusi, China e Pólo Dois Portos, Instituto National de Investigação Agrária e Veterinária, I.P., Quinta da Almoinha, 2565-191 Dois Portos, Portugal article info abstract
Article history: To develop an efficient method for large preparation of various individual polyphenols from white grape Received 10 February 2016 skins (Fernão Pires; Vitis vinifera) by preparative high-speed counter-current chromatography (HSCCC) Received in revised form 24 May 2016 and preparative-HPLC, an optimized preparative HSCCC condition with two-phase solvent system com- Accepted 5 June 2016 posed of Hex-EtOAc-H O (1:50:50, v/v) was used to separate grape skin polyphenols into various frac- Available online 16 June 2016 2 tions. Both the tail-head and head-tail elution modes were used with a flow rate of 3.0 ml/min and a rotary speed of 950 rpm. Afterwards, a preparative-HPLC separation was applied to isolate individual Chemical compounds studied in this article: polyphenols in each of the fractions from HSCCC. Total of 7 fractions (Fraction A to G) were obtained from (+)-Catechin (PubChem CID: 107957) grape skin extract by HSCCC. After preparative-HPLC isolation, fifteen individual compounds were ( )-Epicatechin (PubChem CID: 72276) Astilbin (PubChem CID: 119258) obtained, most of which presented high yields and purity (all over 90%). The HSCCC method followed Procyanidin B1 3-gallate (PubChem CID: with preparative-HPLC appeared to be convenient and economical, constituting an efficient strategy 72193635) for the isolation of grape skin polyphenols. trans-Coutaric acid (PubChem CID: Ó 2016 Elsevier Ltd. All rights reserved. 57517924) Procyanidin B3 (PubChem CID: 146798) Procyanidin B4 (PubChem CID: 147299) Procyanidin B2 (PubChem CID: 122738) Procyanidin B1 (PubChem CID: 11250133) Procyanidin C1 (PubChem CID: 169853) Phloroglucinol (PubChem CID: 359) (+)-Catechin-phloroglucinol derivative (PubChem CID: 14009031)
Keywords: High-speed counter-current chromatography HPLC separation Polyphenols Grape skins
1. Introduction hydroxyl group and can in general be classified into two main groups: non-flavonoids (hydroxybenzoic and hydroxycinnamic Plant polyphenols constitute one of the most numerous and acids and their derivatives, stilbenes and phenolic alcohols) and widely distributed groups of natural products (Antoniolli, flavonoids (anthocyanins, flavanols, flavonols and dihy- Fontana, Piccoli, & Bottini, 2015). From chemical point of view, droflavonols) (Fanzone et al., 2012). Recently, there has been much polyphenols are characterized by a benzene ring with one or more interest in these phenolic compounds because of their wide range of bioactivities as antioxidants (Jara-Palacios, Hernanz, Escudero- Gilete, & Heredia, 2014), antimicrobials (Kemperman et al., ⇑ Corresponding author at: School of Functional Food and Wine, Shenyang 2013), anti-inflammatory agents (Mossalayi, Rambert, Renouf, Pharmaceutical University, 110016 Shenyang, China. E-mail address: [email protected] (B. Sun). http://dx.doi.org/10.1016/j.foodchem.2016.06.009 0308-8146/Ó 2016 Elsevier Ltd. All rights reserved. L. Luo et al. / Food Chemistry 212 (2016) 712–721 713
Micouleau, & Mérillon, 2014), and anti-cancer drugs (Sun, Chen, 2. Experimental Wu, Yao, & Sun, 2012). Grapes and red wines are among the richest sources of phenolic 2.1. Apparatus compounds and polyphenols presented in grapes are essentially in their solid parts, i.e., seed, skin and stem (Sun, Ribes, Leandro, The preparative HSCCC employed in study was a model TBE- Belchior, & Spranger, 2006). As a consequence, grape pomaces, par- 300B HSCCC (Tauto Biotechnique Company, Shanghai, China). The ticularly those from white wine-making technology, provide a rich apparatus with the maximum rotational speed of 1000 rpm was source of these compounds. Today, various polyphenols products equipped with three polytetrafluoroethylene preparative coils (ID can be found on the market but most of them are originated from 1.9 mm, total volume is 300 ml) and a 20 ml sample loop. The grape seeds, i.e., OPCs (oligomeric procyanidins) and few of them HSCCC system consists of a TBP-5002 pump, a UV2000D detector from grape skins. Grape skins contain not only procyanidins, as (Shanghai Sanotac Scientific Instrument Co., Ltd., Shanghai, China) grape seeds, but also other phenolic compounds such as flavonols, and a DC-0506 low constant temperature bath (Tauto Biotechnique flavanonols, flavones, derivatives of cinnamic acids and tartaric Company, Shanghai, China) which was used to control the separa- acid (Cheynier, Moutounet, & Sarni-Manchado, 1998); for grape tion temperature. EasyChrom-1000 chromatography workstation skins from red varieties, they contain also large amount of antho- (Shanghai Sanotac Scientific Instrument Co., Ltd., Shanghai, China) cyanins (Ribéreau-Gayon, 1982). Enological tannins from grape was employed to record the chromatograms. skins have in general better quality than those from grape seeds and grape stems, due essentially to their polyphenols composition 2.2. Reagents and materials (Vivas et al., 2004). From quantitative point of view, grape pomaces are mainly composed of grape skins and the number of polyphe- Sufficient amount of grape skins were manually isolated from nols in grape skins is higher than in grape seeds. Because the dis- non-macerated white grape pomace (Fernão Pires; Vitis vinifera, posal of grape skins is one of environmental problems during cv.), air-dried at <40 °C dark for one day, sealed with N2 and stored wine-making process, the recovery of phenolic compounds from at 20 °C until used. Oligomer proanthocyanidin fraction (OPC) these waste materials is of undoubtedly economical and social and polymer proanthocyanidin fraction (PPC) used for determina- significance. tion of K values of HSCCC were prepared from grape skins by col- The selection of suitable and effective methods for large separa- umn chromatography on Lichroprep RP-18 column according to tion of bioactive phenolic compounds from plant tissues is always Sun et al. (1998); both OPC and PPC presented high purity (92% a critical problem in the scientific community. Traditional separa- and 93%, respectively) (Spranger, Sun, Mateus, Freitas, & Ricardo- tion methods for polyphenols such as liquid-liquid extraction (Sun, da-Silva, 2008). ( )-Epicatechin was purchased from Chengdu Ferrão, & Spranger, 2003; Sun et al., 2006), thin-layer chromatogra- Must Bio-Technology Co., Ltd. (Chengdu, China). Phloroglucinol phy (Jesionek, Majer-Dziedzic, & Choma, 2015; Sun, Leandro, was purchased from Aladdin reagent (Shanghai, China). All organic Ricardo-da-Silva, & Spranger, 1998), column chromatography (Lu solvents used for prep-HSCCC (analytical grade) and for prep-HPLC & Foo, 1999; Sun, Belchior, Ricardo-da-Silva, & Spranger, 1999), separation and HPLC/UPLC analysis (chromatographic grade) were semi-preparative and preparative HPLC (Makhotkina & Kilmartin, purchased from Chemical Branch of Shandong Yuwang industrial 2012; Sun, Belchior, et al., 1999) have disadvantages like time- Co., Ltd. (Shandong, China). consuming, secondary-pollution, complex-process, low-yield and high-cost. High-speed counter-current chromatography (HSCCC), a supporting free liquid-liquid partition system where solutes are 2.3. Preparation of crude grape skin phenolic extract partitioned between two immiscible phases, offers various advan- tages such as high solute loading capability, high recovery, high The skins isolated from the pomace will be first immersed in repeatability and low solvent consumption (Zhang et al., 2015). liquid nitrogen and then ground finely by a miller. Each powder Furthermore, this technique has a great advantage over the tradi- obtained will be immediately used for the extraction of total phe- tional methods by eliminating the complications resulting from nolic compounds using the method described previously (Sun, the solid support matrix, such as irreversible adsorptive sample Belchior, et al., 1999). Furthermore, a 200 g portion of the powder loss and deactivation, tailing of solute peaks and contamination is extracted using 3.0 L of methanol-water (80/20; v/v) followed by (Wang et al., 2015). 3.0 L of acetone-water (75/25; v/v). Each solvent extraction is per- HSCCC is now recognized as an efficient preparative technique, formed by stirring for 3 h under a nitrogen atmosphere at room and widely used for separation and purification of catechins (Wang temperature. The combined supernatants will be evaporated at ° et al., 2008; Xia, Hong, & Liu, 2014), oligomeric proanthocyanidins <30 C to remove organic solvents, followed by extraction with hexane (3 600 ml) to eliminate fatty materials, and then filtered (Esatbeyoglu & Winterhalter, 2010) and other phenolic compounds l from natural plants (Shu et al., 2014). Degenhardt, Hofmann, through a membrane filter (0.45 m). The aqueous phenolic solu- Knapp, and Winterhalter (2000) used firstly HSCCC technique for tion will be lyophilized. The powder thus obtained as crude grape ° preparative isolation of anthocyanins from grape skins of red skin phenolic extract (GSE), was stored at 20 C until used. varieties. In practice, grape skins can be largely obtained from pomaces of 2.4. Selection of the two-phase solvent systems vinification, but only grape skins from white winemaking preserve nearly all their polyphenols while those from red winemaking are The following six solvent systems were selected based on the less interesting because they suffered maceration process during partition coefficient values (K)of( )-epicatechin, OPC, PPC and alcoholic fermentation and thus their important part of polyphe- GSE: (1) n-butanol: ethyl acetate: water (1:20:20); (2) methyl alco- nols are transferred into wine (Sun, Pinto, Leandro, Ricardo-Da- hol: ethyl acetate: water (1:25:25); (3) ethyl acetate: water (1:1); Silva, & Spranger, 1999). (4) n-hexane: ethyl acetate: water (1:20:20); (5) n-hexane: ethyl The purpose of this work was to develop an efficient method acetate: water (1:50:50); (6) n-hexane: ethyl acetate: methyl alco- using preparative HSCCC combined with preparative HPLC for large hol: water (1:50:1:50). Each solvent mixture was thoroughly equi- separation of the main polyphenols in grape skins from white librated in a separator funnel at room temperature and the two winemaking pomace. phases were separated shortly before use. 714 L. Luo et al. / Food Chemistry 212 (2016) 712–721
The K values were determined spectrophotometrically by the Two elution solvent A (water:formic acid; 99.8:0.2, v/v) and B (ace- following procedure using the above two-phase solvent systems. tonitrile:formic acid; 99.8:0.2, v/v) were used with the gradient 3.0 ml of each phase was delivered into a test tube to which elution program. Each fraction has its own chromatographic condi- 1 mg of PPC, GSE and 0.5 mg of ( )-epicatechin, OPC were added tions. Fraction A: 0 min (A98%:B2%), 1 min (A94%:B6%), 2 min respectively. The contents were thoroughly mixed and allowed to (A88%:12%), 3 min (A86%:B14%), 4 min (A85%:B15%), 6 min settle at room temperature. After two clear layers were formed, (A84%:B16%), 7 min (A73%:B27%), 8 min (A68%:B32%), 9 min 2 ml of each phase was removed to evaporate to dryness and dis- (A0%:B100%), 10 min (A0%:B100%), 10.1 min (A98%:B2%); Fraction solved with 5 ml methanol to determine the absorbance at C: 0 min (A98%:B2%), 1 min (A96%:B4%), 2 min (A90%:B10%), 280 nm using a UV-2000 spectrophotometer (UNIC, Shandong). 4 min (A88%:B12%), 6 min (A87%:B13%), 7.5 min (A69%:B31%), The K values were expressed as the absorbance of the ( )- 8 min (A68%:B32%), 9 min (A0%:B100%), 10 min (A0%:B100%), epicatechin, OPC, PPC and GSE in the lower phase divided by that 10.1 min (A98%:B2%); Fraction D: 0 min (A98%:B2%), 1 min of the upper phase. (A90%:B10%), 7 min (A90%:B10%), 8 min (A68%:B32%), 9 min (A0%:B100%), 10 min (A0%:B100%), 10.1 min (A98%:B2%); Fraction 2.5. Preparation of two-phase solvent system and sample solution E: 0 min (A98%:B2%), 1 min (A88%:B12%), 2 min (A91%:B9%), 3 min (A91%:B9%), 6 min (A77%:B23%), 7 min (A75%:B25%), 8 min The two-phase solvent system composed of n-hexane: ethyl (A68%:B32%), 9 min (A0%:B100%), 10 min (A0%:B100%), 10.1 min acetate: water (1:50:50; v:v:v) was used for HSCCC separation. (A98%:B2%); Fraction B, F, G: 0 min (A98%:B2%), 1 min (A96%: The solvent system was thoroughly equilibrated by shaking repeat- B4%), 2 min (A88%:B12%), 4 min (A88%:B12%), 8 min (A68%: edly in a separation funnel at room temperature. After being equi- B32%), 9 min (A0%:B100%), 10 min (A0%:B100%), 10.1 min (A98%: librated, the upper phase and the lower phase were separated and B2%); Fraction H: 0 min (A98%:B2%), 1 min (A96%:B4%), 2 min degassed by sonication for 10 min prior to use. The sample solution (A92%:B8%), 4 min (A92%:B8%), 5 min (A85%:B15%), 6 min (A85%: for the HSCCC separation was prepared by dissolving 300 mg of the B15%), 8 min (A68%:B32%), 9 min (A0%:B100%), 10 min (A0%: grape skin extracts in 20 ml lower phase. B100%), 10.1 min (A98%:B2%).
2.6. HSCCC separation procedure 2.8. Preparative-HPLC separation
Both tail-head and head-tail elution of semi-preparative HSCCC Preparative-HPLC separation was carried out using waters Alli- were performed for the separation of grape skin extracts. Firstly, ance e2695 Separation Module equipped with a 2998 Photo Diode the multilayer coil column was entirely filled with the lower phase Array (PDA) Detector coupled to a data processing computer (stationary phase) at a flow rate of 35.0 ml/min. Then the upper (EmpowerTM 2 chromatography data software). The detection ran- organic phase (mobile phase) was pumped into the column at a ged from 190 to 790 nm, being 280 nm for detection of grape skin flow rate of 3.0 ml/min in tail-head elution mode, while the appa- polyphenols. The column was YMC-Pack ODS-A (250 10 mm, ratus was run at a revolution speed of 950 rpm. After the mobile 5 lm) and the temperature was set at 30 °C. The flow rate of the phase front emerged at the head outlet and hydrodynamic equilib- mobile phase was fixed at 3.0 ml/min for fraction A, 3.5 ml/min rium was established throughout the column, about 20 ml sample for fraction B, D and E, and 4.0 ml/min for fraction C. Two elution solution with 300 mg GSE was injected into the tail of the column solvents, the elution solvents A (water) and B (methanol) were through the injection valve. Retention of stationary phase was cal- used for fraction A and D separation, and fraction B, C and E used culated. After the separation was carried out in tail-head elution solvent A (water:formic acid; 99.8:0.2, v/v) and B (acetonitrile:- mode for 300 min, the mode was switched to the head-tail elution formic acid; 99.8:0.2, v/v) as mobile phase. Other chromatographic mode with the lower phase as the mobile phase. The separation conditions were preformed as follow: catechin, epicatechin and temperature was controlled at 25 °C during the whole experiment. astilbin from fraction A, A57%:B43%; gallic acid from fraction B,
The effluents were monitored with a UV detector at 280 nm and A95%:B5%; procyanidin B2-3-O-gallate, procyanidin B1-3-O- 0 the fractions were collected manually according to the chro- gallate and procyanidin B2-3 -O-gallate from fraction B, A85%: matogram. Collected fractions were lyophilized and the powders B15%; trans-coutaric and quercetin-3-O-glucuronide from fraction thus obtained were stored at 20 °C under N2 and darkness. Five C, A80%:B20%; Procyanidin B3, Procyanidin B4 and B2 from fraction replicates of the same amount of samples were carried out by D, A80%:B20%; Procyanidin B1, Procyanidin T2 and C1 from fraction HSCCC in order to confirm the repeatability. Phenolic composition E, A85%:B15%. in these powdery products was verified by UPLC analysis, isolation of individual polyphenols was performed by preparative HPLC, 2.9. Q-TOF MS analysis while structural identification of each phenolic compound was car- ried out using Q-TOF MS, phloroglucinolysis-UPLC and 1H NMR The Xevo G2 Q-TOF mass spectrometer was used in negative ESI analysis, as described below. mode for data acquisition, which allows both precursor and pro- duct ion data to be acquired in one injection. Typical source condi- 2.7. UPLC analysis tions for maximum intensity of precursor ions were as follows: capillary voltage, 3.0 kV; sample cone, 30 V; extraction cone, Each fraction from the preparative HSCCC separation was ana- 4.0 V; source temperature, 120 °C; desolvation temperature lyzed using the ACQUITY H-class system composed of a Quaternary 450 °C; cone gas flow rate 50 L/h; desolvation gas (N2) flow rate Solvent Manager (QSM), a Sample Manager with Flow through 700 L h 1. All analyses were performed using the lockspray, which Needle (SM-FTN), and a Photodiode Array (PDA) Detector coupled ensured accuracy and reproducibility. Data acquisition was to a data processing computer (EmpowerTM 2 chromatography data achieved using MS, which the second function programmed with software). The column was a ACQUITY UPLC BEH C18 independent collision energies was used. Function 2 (high energy): (50 2.1 mm, 1.7 lm) and the temperature was set at 30 °C. The 0–1500 mass-scan range; 0.2 s scan time; 0.02 s inter-scan time; detect wavelength was set at 280 nm. The flow rate of the mobile collision energy ramp of 30–50 V. Acquiring data in this manner phase was fixed at 0.3 ml/min. The chromatographic conditions provided for the collection of information of intact precursor ions were optimized according to the method of Sun et al. (2011). as well as fragment ions. L. Luo et al. / Food Chemistry 212 (2016) 712–721 715
2.10. Phloroglucinolysis-UPLC analysis these solvent systems were all feasible. According to Ito (2005), the ideal K value is approximately 1. As shown in Table 1, the K
Phloroglucinolysis followed by UPLC analysis was applied to value of OPC in the n-Butanol-EtOAc-H2O (1-20-20, v/v) system estimate the proanthocyanidins composition of the fractions iso- was 1.0 while the K value of the other solvent systems are more lated from HSCCC. 200 ll of fraction in methanol (6 mg/ml) was than 1.7, which indicated that the solvent system of n-Butanol- added to 200 ll of phloroglucinol (50 mg/ml) in methanol acidified EtOAc-H2O (1-20-20, v/v) was the best to select. However, the sta- by 0.2 M HCl in a stoppered test tube. The tube was incubated for bility and the retention rate of this solvent system were low. As a
20 min at 50 °C for the complete hydrolysis reaction and then put it result, the solvent system of Hex-EtOAc-H2O (1-50-50, v/v) was in cooling water to stop the reaction. The reaction mixture was fil- selected for the study. According to the K value in this solvent sys- tered through 0.22 lm filter, followed by analysis of the hydrolysis tem, PPC required at least about another 300 min for elution in tail- products using ACQUITY H-class UPLC system. The detect wave- head elution mode with the upper mobile phase. Therefore, the length was set at 280 nm and the temperature was set at 30 °C. head-tail elution mode with the lower mobile phase was applied The chromatographic conditions were performed according to for the separation of PPC. the method of Sun et al. (2011) with a few modifications. The flow Other factors such as the flow rate of the mobile phase and the rate of the mobile phase was fixed at 0.3 ml/min. Two elution sol- amount of sample loading were also investigated. The flow rate of vent A (water:formic acid; 99.8:0.2, v/v) and B (acetonitrile:formic the mobile phase determines the separation time, the amount of acid; 99.8:0.2, v/v) were used with the gradient elution program: stationary phase retained in the column, and therefore the peak 0 min (A98%:B2%), 1 min (A96%:B4%), 2 min (A88%:B12%), 4 min resolution (Ito, 2005). In the present study, the flow rate was inves- (A88%:B12%), 8 min (A68%:B32%), 9 min (A0%:B100%), 12 min tigated at 2.0 ml/min and 3.0 ml/min. Both the 2.0 ml/min and the (A98%:B2%). 3.0 ml/min were fit for the separation resolution and the retention rate of stationary phase. But the detection time was prolonged and the peak-shape was ugly when the 2.0 ml/min was used. Therefore, 2.11. NMR analysis the best flow rate of mobile phase was selected at 3.0 ml/min. Besides, sample loading was also investigated from 100 to 1H NMR spectra were generally obtained in Me CO-d with a 2 6 500 mg of the GSE in 20 ml of the lower phase. A large amount small amount of D O by Bruker ARX-600 MHz NMR Spectrometer. 2 of sample would lead to a great loss of stationary phase from the Chemical shift (d) are expressed in parts per million (ppm) and column and low resolution, whereas a small amount of sample coupling constant (J) in Hz. would decrease the yield of products. As a result, 300 mg of GSE was the optimum with lower loss of stationary phase and higher 3. Results and discussion yield. Based on these results, the HSCCC separation was performed 3.1. Optimization of HSCCC conditions using the two-phase solvent system composed of Hex-EtOAC- H2O (1:50:50, v/v) in tail-head and head-tail elution modes. The It has been clarified that the most important step in the design sample loading was 300 mg and the flow rate was 3.0 ml/min. of an HSCCC procedure is the selection of the solvent system. The partition coefficient values (K)of( )-epicatechin, OPC, PPC and 3.2. Preparative HSCCC fractionation of grape skin extract GSE are listed in Table 1 together with separation factor (a). The K value is the most important parameter to determine the resolu- Under the optimized HSCCC conditions, grape skin extract (GSE) tion in HSCCC (Shibusawa, Yanagida, Isozaki, Shindo, & Ito, 2001). could be successfully separated into 8 fractions during 400 min of The suitable K values for HSCCC are 0.5 5 K 5 2.0 and the separa- elution. The HSCCC chromatogram was shown in Fig. 1. The results tion factor between two components (a = K2/K1, K2>K1) should showed that during a working day, 300 mg of sample loading aver- be greater than 1.5 (Ito, 2005). age gave 16.7 mg fraction A, 6.7 mg fraction B, 5.3 mg fraction C, In the six solvent systems tested in the study, all the a values 7.4 mg fraction D, 12.5 mg fraction E, 144.1 mg fraction F, were more than 1.5, indicating that the separation resolution of 61.8 mg fraction G and 6.1 mg fraction H. The total yield was
Table 1 Partition coefficient values (K) and separation factors (a)of( )-epicatechin, oligomers, polymers and GSE.
Solvent system (v/v) ( )-epicatechin(1) Oligomers(2) Polymers(3) GSE(4)
n-Butanol–EtOAc–H2O 1-20-20 K 0.257 1.014 10.415 2.561 a 3.946(2/1) 9.965(4/1)
Hex-EtOAc-H2O 1-20-20 K 0.756 3.323 20.727 5.398 a 4.396(2/1) 7.140(4/1) 1-50-50 K 0.517 2.304 18.329 3.313 a 4.456(2/1) 6.408(4/1)
EtOAc-H2O 1-1 K 0.407 1.858 16.159 3.467 a 4.565(2/1) 8.518(4/1)
MeOH-EtOAc-H2O 1-25-25 K 0.371 1.766 15.382 3.397 a 4.760(2/1) 9.156(4/1)
Hex–EtOAc–MeOH–H2O 1-50-1-50 K 0.648 2.305 18.394 3.773 a 3.557(2/1) 5.823(4/1)
Abbreviations: MeOH, methanol; Hex, n-hexane; EtOAc, ethyl acetate. 716 L. Luo et al. / Food Chemistry 212 (2016) 712–721
HPLC-DAD and HPLC-ESI-MS were used for checking the purity of each compound. Each of these individual compounds was lar- gely isolated using prep-HPLC under the optimized elution condi- tions (Fig. 2, Fraction A, right side). From 16.7 mg of Fraction A obtained by one step HSCCC separation of GSE, about 6.1 mg (2.0% in 300 mg GSE) of catechin at the purity of over 94%, 4.5 mg (1.5% in 300 mg GSE) of epicatechin at the purity of over 95%, and 1.1 mg (0.4% in 300 mg GSE) of astilbin at the purity of over 96%, could be obtained. These results indicated that white grape skins could be used as a source of catechin, epicatechin and astilbin. Although catechin, epicatechin and astilbin from other origin were commercially available, the present work developed a new procedure for rapid and large isolation of these compounds by HSCCC combined with prep-HPLC. These three compounds from white grape skin extract using one-step HSCCC followed by preparative HPLC showed high yield and purity, besides the new method was time-saving, eco- nomical and with greatly improved isolation efficiency. Fig. 1. HSCCC chromatogram on the separation of grape skin extract using tail to head (0–300 min) and head to tail (300 min-end) elution modes. 3.5.2. Fraction B Fig. 2 (Fraction B) presented the chromatograms of UPLC analy- sis and prep-HPLC isolation of individual phenolic compounds of 260.6 mg, losing only about 40 mg of GSE. The recovery reached up Fraction B isolated from HSCCC of grape skin extract. to 86.9%. UPLC analysis (Fig. 2, Fraction B, left side) showed that, for Frac- tion B, the major individual phenolic compounds included gallic 3.3. Method repeatability of HSCCC acid, galloylated procyanidin dimers B2-3-O-gallate, B1-3-O- gallate and procyanidin B2-30-O-gallate. Under the optimized HSCCC conditions, the repeatability of For the similar structures of procyanidin gallate dimers, the elu- HSCCC method for separation of grape skin polyphenols was tested tion behavior of these compounds showed consistent in HSCCC. by repetitive sample injection in five days. The relative standard Additionally, the polarity of these procyanidin gallate dimers is (RSD) values of repeatability of the Fraction A to H were 5.2%, higher than monomer units because of the increased hydroxy. And 6.1%, 16.8%, 4.5%, 16.8%, 5.1%, 4.1% and 13.7% respectively. the polarity of gallic acid is also higher than the compounds in Frac- tion A due to the carboxyl. As a result, these four polyphenols were together eluted as the second fraction by HSCCC presented in Fig. 1. 3.4. Optimization of UPLC method HPLC-DAD and HPLC-ESI-MS were used for checking the purity of each compound. Each of these individual compounds was lar- The optimization of a new method for the determination of gely isolated by prep-HPLC under the optimized elution conditions phenolic compounds has been carried out. To perform this opti- (Fig. 2, Fraction B, right side). From 6.7 mg of Fraction B obtained mization, the selection of the flow rate of the mobile phase and by one step HSCCC separation of GSE, about 3.4 mg (1.1% in the mobile phase gradient were done on the basis of previous 300 mg GSE) of gallic acid at the purity of over 96%, 1.1 mg (0.4% experiment according to Sun et al. (2011). Several trial-and-error in 300 mg GSE) of B2-3-O-gallate at the purity of over 91%, experiments were performed to obtain the optimum gradient. 0.7 mg (0.2% in 300 mg GSE) of B1-3-O-gallate at the purity of over Once an optimum separation of the compounds was obtained, 90% and 0.4 mg (0.1% in 300 mg GSE) of B2-30-O-gallate at the pur- the attention was focused on reducing the analysis time. Therefore, ity of over 91% could be obtained. the final results showed that the best resolution, shortest analysis From the above, although the yields of procyanidin dimers B2- time, and lowest pressure variations were achieved when the flow 3-O-gallate, B1-3-O-gallate and procyanidin B2-30-O-gallate is rate of mobile phase was 0.3 ml/min and a gradient elution mode lower than those phenolic compounds from Fraction A, these gal- composed of 0.2% formic acid in water (solvent A) and 0.2% formic loylated procyanidins were not commercially available presently. acid in acetonitrile (solvent B) was programmed as above men- Because galloylated procyanidins were confirmed to have much tioned (2.7 UPLC analysis). higher antioxidant activities than the non-galloylated ones (Phansalkar et al., 2015; Zhang et al., 2015), the results obtained 3.5. UPLC analysis and prep-HPLC isolation of individual polyphenols by this part of work were of utmost importance, providing a new method for preparation of these bioactive compounds. 3.5.1. Fraction A Fig. 2 (Fraction A) presented the chromatograms of UPLC anal- 3.5.3. Fraction C ysis and prep-HPLC isolation of individual phenolic compounds Fig. 3 (Fraction C) presented the chromatograms of UPLC analy- of Fraction A isolated from HSCCC of grape skin extract. sis and prep-HPLC isolation of individual phenolic compounds of UPLC analysis (Fig. 2, Fraction A, left side) showed that, for Frac- Fraction C isolated from HSCCC of grape skin extract. tion A, the major individual phenolic compounds were catechin, UPLC analysis (Fig. 3, Fraction C, left side) showed that, for Frac- epicatechin and astilbin, with the first two quantitatively tion C, the major individual phenolic compounds were trans- predominant. coutaric acid and quercetin-3-glucuronide. Catechin and epicatechin are flavan-3-ol monomer units mainly Quercetin-3-glucuronide belongs to flavonoids that are mainly through C4 ? C8 or C4 ? C6 bonds that constitute proanthocyani- localized in the skins of the grape (Yilmaz et al., 2015). trans- dins. The polarity of these three compounds is similar and low so Coutaric acid belongs to phenolic acids that possess at least one they were together eluted first by HSCCC in tail-head elution mode carboxylic acid functional group. These two compounds do not with the upper mobile phase. have much more numbers of carbons than the compounds in Frac- L. Luo et al. / Food Chemistry 212 (2016) 712–721 717
Fig. 2. Fraction A UPLC analysis (Fraction A, left side) and prep-HPLC isolation (Fraction A, right side) of individual phenolic compounds isolated from HSCCC of grape skin extract; peak 1 = catechin, peak 2 = epicatechin, peak 3 = astilbin. Fraction B UPLC analysis (Fraction B, left side) and prep-HPLC isolation (Fraction B, right side) of individual phenolic compounds isolated from HSCCC of grape skin extract; peak 4 = gallic acid, peak 5 = procyanidin dimer B2-3-O-gallate, peak 6 = procyanidin dimer B1-3-O-gallate, peak 7 = procyanidin B2-30-O-gallate. tion B. But the structure of quercetin-3-glucuronide contains lots of UPLC analysis (Fig. 3, Fraction D, left side) showed that, for Frac- hydroxyls and a carboxyl due to the glucuronide, and trans- tion D, the major individual phenolic compounds included pro- coutaric acid has two carboxyl groups, so that the polarity of cyanidin dimers B3, B4 and B2. trans-coutaric acid and quercetin-3-glucuronide is higher than These three dimers are B-type procyanidins which are linked the compounds in Fraction B. As a result, these two compounds through C4 ? C8 bond according to the 1H NMR spectroscopic were together eluted thirdly by HSCCC presented in Fig. 1. data, Ms and MS2 analysis, as well as the UPLC retention time com- HPLC-DAD and HPLC-ESI-MS were used for checking the purity pared with the standards. The similar structure of these dimeric of each compound. Each of these individual compounds was lar- procyanidins led to the same elution behavior in HSCCC. However, gely isolated by prep-HPLC under the optimized elution conditions the polarity of these three compounds is higher than the com- (Fig. 3, Fraction C, right side). From 5.3 mg of Fraction C obtained pounds in Fraction B and A because of the increase of hydroxyls by one step HSCCC separation of GSE, about 0.5 mg (0.2% in and the poor planarity. As a result, these three dimeric procyani- 300 mg GSE) of trans-coutaric acid at the purity of over 92% and dins were together eluted fourthly by HSCCC presented in Fig. 1. 1.6 mg (0.5% in 300 mg GSE) of quercetin-3-glucuronide at the pur- HPLC-DAD and HPLC-ESI-MS were used for checking the purity ity of over 94% could be obtained. Quercetin-3-glucuronide was of each compound. Each of these individual compounds was lar- one of the major phenolic compounds presented in grape skins gely isolated by prep-HPLC under the optimized elution conditions (Cheynier et al., 1998) and was reported to be antioxidative (Fig. 3, Fraction D, right side). From 7.4 mg of Fraction D obtained (Moon, Tsushida, Nakahara, & Terao, 2001), anti-inflammatory by one step HSCCC separation of GSE, about 0.2 mg (0.1% in (Suri et al., 2008) and anti-cancer (Yamazaki, Miyoshi, Kawabata, 300 mg GSE) of procyanidin B3 at the purity of over 93%, 0.9 mg Yasuda, & Shimoi, 2014). The proposed method permitted large (0.3% in 300 mg GSE) of procyanidin B4 at the purity of over 95% separation of this bioactive compound from GSE. and 4.2 mg (1.4% in 300 mg GSE) of procyanidin B2 at the purity of over 95% could be obtained. 3.5.4. Fraction D These dimers are the simplest procyanidins and have a strong Fig. 3 (Fraction D) presented the chromatograms of UPLC anal- activity of antioxidant and to be a promising property at the car- ysis and prep-HPLC isolation of individual phenolic compounds of diovascular level (Kruger, Davies, Myburgh, & Lecour, 2014). More- Fraction D isolated from HSCCC of grape skin extract. over, although these three procyanidin dimers can be easily found 718 L. Luo et al. / Food Chemistry 212 (2016) 712–721
Fig. 3. Fraction C UPLC analysis (Fraction C, left side) and prep-HPLC isolation (Fraction C, right side) of individual phenolic compounds isolated from HSCCC of grape skin extract; peak 8 = trans-coutaric acid, peak 9 = quercetin-3-glucuronide. Fraction D UPLC analysis (Fraction D, left side) and prep-HPLC isolation (Fraction D, right side) of individual phenolic compounds isolated from HSCCC of grape skin extract; peak 10 = procyanidin dimer B3, peak 11 = procyanidin dimer B4, peak 12 = procyanidin dimer B2. Fraction E UPLC analysis (Fraction E, left side) and prep-HPLC isolation (Fraction E, right side) of individual phenolic compounds isolated from HSCCC of grape skin extract; peak 13 = procyanidin dimer B1, peak 14 = procyanidin trimer T2, peak 15 = procyanidin trimer C1. today as chemical reagents, they are very expensive. The present gely isolated using prep-HPLC under the optimized elution condi- work proposed an alternative, feasible and economical method tions (Fig. 3, Fraction E, right side). From 12.5 mg of Fraction E for large preparation of these compounds. obtained by one step HSCCC separation of GSE, about 5.8 mg (1.9% in 300 mg GSE) of procyanidin B1 at the purity of over 90%, 3.5.5. Fraction E 1.7 mg (0.6% in 300 mg GSE) of procyanidin T2 at the purity of over Fig. 3 (Fraction E) presented the chromatograms of UPLC analy- 91% and 1.7 mg (0.6% in 300 mg GSE) of procyanidin C1 at the pur- sis and prep-HPLC isolation of individual phenolic compounds of ity of over 95% could be obtained. Fraction E isolated from HSCCC of grape skin extract. Today, procyanidin B1, T2 and C1 can be found in the market UPLC analysis (Fig. 3, Fraction E, left side) showed that, for Frac- but the price is very expensive. The present work proposed an tion E, the major individual phenolic compounds included pro- alternative method, convenient and practical, for large preparation cyanidin dimer B1, procyanidin trimer T2 and C1. of these compounds. With the increase of the polymerization degree of procyanidins, the hydrophobicity decreased. In the three main compounds of this 3.5.6. Fractions F, G and H fraction, two are trimeric procyanidins whose polarity is higher Fig. 4 presented the chromatograms of UPLC analysis of Fraction than the dimeric procyanidins. Due to the different linkage in pro- F, G and H isolated from HSCCC of grape skin extract. Although we cyanidin dimers, the polarity of procyanidin B1 is higher than pro- made great efforts to separate the phenolic compounds by opti- cyanidin B3 and B4. Therefore, procyanidin dimer B1, trimer T2 and mizing the elution conditions, no satisfactory resolutions of indi- C1 were together eluted fifthly in HSCCC presented in Fig. 1. vidual peaks were achieved, suggesting that the polyphenols in HPLC-DAD and HPLC-ESI-MS were used for checking the purity these fractions would be composed essentially of polymeric forms. of each compound. Each of these individual compounds was lar- For this reason, no further tentative fractionation and isolation of L. Luo et al. / Food Chemistry 212 (2016) 712–721 719
Fig. 4. UPLC analysis of polyphenols in Fractions F, G and H isolated from HSCCC of grape skin extract. Individual peaks could not be identified due to high impurity and bad resolution. individual polyphenols in these fractions were done by prep-HPLC, ing order, from 1.3 to 3.2, which was in good agreement with the but structural characterization of their polymeric polyphenols results from direct UPLC analysis. were performed by phloroglucinolysis-UPLC analysis. It was worth mentioning that for all the three fractions, F, G and H, catechin was the predominant terminal unit and epicatechin 3.6. Phloroglucinolysis was predominant extension unit, which were in coincidence with those from polymeric procyanidins of grape seeds. In other words, Phloroglucinolysis or thiolysis method has been a useful tool for the fact that the fraction F, G and H were composed of essentially structural characterization of polymeric polyphenols, i.e., poly- proanthocyanidins suggested that they may be used as new meric proanthocyanidins (Cheynier, 2002). Phloroglucinolysis or polyphenols products (OPCs or PPCs). thiolysis followed by HPLC/UPLC analysis permitted access to aver- age compositional data of proanthocyanidin molecules including 3.7. Structural identification percentage in mole of terminal and intermediate units as well as mean degree of polymerization (mDP) (Spranger et al., 2008; Sun Compounds were identified according to their mass characteris- et al., 1998). As compared with thiolysis, phloroglucinolysis tics and also phloroglucinolysis-UPLC and retention time in com- appeared more practical and efficient for grape proanthocyanidins parison with available standards or our library data, as well as 1H (Vidal, Cartalade, Souquet, Fulcrand, & Cheynier, 2002). NMR when necessary. Table 2 presented compositional data of proanthocyanidins in Compound 1 (catechin). ESI-QTOF-MS (negative mode): m/z Fraction F, G and H isolated from HSCCC of grape skin extract. It 289.08([M H] ) 245.09. The retention time and the MS data were showed that Fraction F, G and H were composed of higher oligo- in accordance with reference standard. meric and polymeric proanthocyanidins (mDP > 3) than Fraction Compound 2 (epicatechin). ESI-QTOF-MS (negative mode): m/z A to E. The mDP values of fractions F, G and H were slightly in 289.08([M H] ) 245.09. The retention time and the MS data were decreasing order, being 6.35, 4.13 and 3.95, respectively. By con- in accordance with reference standard. trast, the mDP values of Fraction A to Fraction E varied in increas- Compound 3 (astilbin). ESI-QTOF-MS (negative mode): m/z 449.11([M H] ) 303.05 285.04. The retention time and the MS Table 2 Mean polymerization degree (mDP) of HSCCC fractions. data were in accordance with reference standard. Compound 4 (gallic acid). ESI-QTOF-MS (negative mode): m/z Fraction Terminal units (%) Extension units (%) mDP 169.02([M H] ) 125.04. The retention time and the MS data were (c=c ) * 100% (c=c ) * 100% t t in accordance with reference standard. C EC ECG CP ECP ECPG Compound 5 (procyanidin B2-3-O-gallate). ESI-QTOF-MS (nega- F 9.88 5.23 0.58 8.72 71.51 4.07 6.35 tive mode): m/z 729.05([M H] ) 577.01 425.02 407.02 289.04; G 15.19 7.59 1.27 7.59 64.98 3.38 4.13 phloroglucinolysis-UPLC: terminal unit, epicatechin; extension H 13.51 11.26 0.90 7.66 61.26 5.41 3.95 unit: epicatechin 3-O-gallate. The retention time and the MS data Abbreviations: C, (+)-catechin; CP, (+)-catechin-phloroglucinol derivative; EC, ( )- were in accordance with our previous work (Sun, Belchior, et al., epicatechin; ECP, ( )-epicatechin-phloroglucinol derivative; ECG, ( )-epicatechin- 1999). 3-O-gallate; ECGP, ( )-epicatechin-3-O-galloyl-phloroglucinol derivative; c, the Compound 6 (procyanidin B1-3-O-gallate). ESI-QTOF-MS (nega- concentration of each terminal unit and extension unit; ct , the total concentration of each terminal unit and extension unit; mDP, the mean polymerization degree of tive mode): m/z 729.10([M H] ) 577.02 425.02 407.01 289.05; HSCCC fractions. phloroglucinolysis-UPLC: terminal unit, catechin; extension unit: 720 L. Luo et al. / Food Chemistry 212 (2016) 712–721 epicatechin 3-O-gallate. The retention time and the MS data were minal unit, catechin; extension unit, two epicatechin. 1HNMR in accordance with our previous work (Sun, Belchior, et al., 1999). (600 MHz, Methanol-d4): d 2.75–2.54 (m, H-4L), 3.96–4.20 (m, 0 Compound 7 (procyanidin B2-3 -O-gallate). ESI-QTOF-MS (neg- 2.0, H-3U,3M, H-3L), 4.70 and 4.76 (2 s, H-4U, H-4 M), 5.06 and ative mode): m/z 729.18([M H] ) 577.01 425.01 407.05 289.04; 5.26 (2 s, H-2M,2L,2U), 5.90 (s, H-6M,6L), 6.01 (s, H-6U,8U), 6.66- 0 0 0 0 0 0 0 0 0 phloroglucinolysis-UPLC: terminal unit, epicatechin-3-O-gallate; 7.12 (m, H-2 U,2M,2L,5U,5M,5L,6U,6M,6L). The retention time 1 extension units, epicatechin. H NMR (600 MHz, Acetone-d6) 2.92 was in accordance with reference standard. The spectral data of (d, J = 17.5 Hz, H-4L), 3.06 (dd, J = 17.3, 4.7 Hz, H-4L), 3.99 (s, 1H, compound 14 matched that of procyanidin T2 in the literature 1 H-3U), 4.85 (s, H-4U), 5.21 (s, H-2U,2L), 5.55 (s, H-3L), 5.98 (s, H- (Jin et al., 2012). The H NMR data compared with reference are 0 0 0 0 6U,8U,6L), 6.68–6.77 (m, H-2 U,6U,5L), 7.00 (s, H-5 U), 7.08 (s, shown in Table S4 in Supplementary material. 0 00 0 00 H-6 L,6 L), 7.17 (s, 2H, H-2 L,2 L). The retention time was in accor- Compound 15 (procyanidin C1). ESI-QTOF-MS (negative mode): dance with reference standard. The spectral data of compound 7 m/z 865.21 ([M H] ) 577.11 289.05; phloroglucinolysis-UPLC: ter- 0 1 matched that of procyanidin B2-3 -O-gallate in the literature minal unit, epicatechin; extension unit, two epicatechin. HNMR (Chu, Chen, Wu, & Hsieh, 2014). The 1H NMR data compared with (600 MHz, Methanol-d4): 2.84 (d, J = 16.5 Hz, H-4a L), 2.97 (d, reference are shown in Table S1 in Supplementary material. J = 16.8 Hz, H-4b L), 4.05 (d, J = 13.2 Hz, H-3U,3M), 4.34 (s, br s, H- Compound 8 (trans-coutaric acid). ESI-QTOF-MS (negative 3L),4.74 (s, 2H, H-4U,4M), 5.09 and 5.25 (2brs, H-2U, H-2M, H-2L), 0 0 0 mode): m/z 295.05([M H] ) 163.04. The retention time and the 5.97 and 6.06 (m, H-6U,8U,6M,6L), 6.69-6.85 (m, H-5 U,5M,5L, 0 0 0 0 0 0 MS data were in accordance with reference standard. 6 U,6M, 6 L), 6.94, 7.06 and 7.16 (3 br s, H-2 U, H-2 M, H-2 L). Compound 9 (quercetin-3-O-glucuronide). ESI-QTOF-MS (nega- The retention time was in accordance with reference standard. tive mode): m/z 477.07([M H] ) 301.04. The retention time and The spectral data of compound 15 matched that of procyanidin the MS data were in accordance with reference standard. C1 in the literature (Hor, Heinrich, & Rimple, 1996). The 1H NMR
Compound 10 (procyanidin B3). ESI-QTOF-MS (negative mode): data compared with reference are shown in Table S5 in Supple- m/z 577.16 ([M H] ) 425.05 405.03 289.06; phloroglucinolysis- mentary material. UPLC: terminal unit, catechin; extension unit, catechin. 1H NMR (600 MHz, Methanol-d4): 2.51 (dd, J = 16.2, 8.0 Hz, H-4L), 2.60 4. Conclusions (dd, J = 16.2, 7.7 Hz, H-4L), 3.82 (q, J = 7.5 Hz, H-3L), 4.28 (d, J = 9.7 Hz, H-2U), 4.43 (d, J = 7.8 Hz, H-3U,4U), 4.56 (d, J = 7.3 Hz, A new and effective method by HSCCC combined with prep- H-2L), 5.81 (s, H-8U), 5.86 (s, H-6U), 6.09 (s, H-6L), 6.27 (dd, HPLC was developed, which permitted isolating, in large scale, fif- 0 0 J = 8.2, 2.0 Hz, H-6 L), 6.49 (dd, J = 8.2, 1.9 Hz, H-6 U), 6.60 (d, teen individual polyphenols and three fractions containing 0 0 J = 8.2 Hz, H-2 L), 6.69 (d, J = 2.0 Hz, H-5 L), 6.76 (d, J = 1.9 Hz, H- polyphenols mixtures from white grape skins. Under the optimized 0 0 2 U), 6.69 (d, J = 8.2 Hz, H-5 U). The retention time was in accor- HSCCC conditions, grape skin polyphenols could be separated into dance with reference standard. The spectral data of compound 10 distinct fractions in order of increasing polarity, and all the individ- matched that of procyanidin B3 in the literature (Jin et al., 2012). ual polyphenols isolated from these fractions by prep-HPLC pre- 1 The H NMR data compared with reference are shown in sented high yield and high purity. The three polyphenols Table S2 in Supplementary material. mixtures would be used as new polyphenols products although Compound 11 (procyanidin B4). ESI-QTOF-MS (negative mode): their detailed compositions, chemical and biological properties m/z 577.16 ([M H] ) 425.05 405.03 289.06; phloroglucinolysis- should be further verified. The present investigation could promote 1 UPLC: terminal unit, epicatechin; extension unit, catechin. H the future research for white grape skins on new pharmaceutical NMR (600 MHz, Methanol-d4): 2.72 (dd, J = 17.1, 2.6 Hz, H-4L), preparations or functional products and bioassays. 2.84 (dd, J = 16.9, 2.4 Hz, H-4L), 2.88 (dd, J = 17.0, 5.1 Hz, H-4L), 2.94 (dd, J = 16.8, 4.5 Hz, H-4L), 4.07 (t, J = 3.8 Hz, H-3L), 4.24 (t, Conflict of interest J = 3.4 Hz, H-3L), 4.34 (d, J = 7.1 Hz, H-2U or 4U), 4.44 (d, J = 9.7 Hz, H-2U or 4U), 4.48 (d, J = 6.9 Hz, H-3U,), 4.62–4.57 (m, H-3U), 4.65 The authors have declared no conflict of interest. (d, J = 7.9 Hz, H-2U or 4U), 4.82 (s, H-2L), 4.95 (s, H-2L), 5.81 (s, H- 6U or 8U), 5.86 (s, H-6U or 8U), 5.91 (s, H-6U or 8U), 5.97 (s, H-6U 0 Appendix A. Supplementary data or 8U), 5.96 (s, H-6L), 6.12 (s, H-6L), 6.43 (d, J = 8.0 Hz, H-6 U or 0 0 0 0 6 L), 6.47 (d, J = 8.2 Hz, H-6 U or 6 L), 6.63 (d, J = 8.2 Hz, H-5 U or 0 0 0 0 0 Supplementary data associated with this article can be found, in 5 L), 6.69 (s, H-2 U or 2 L, 0.5H), 6.71 (s, H-2 U or 2 L,0.5H), 6.73 (d, 0 0 0 0 the online version, at http://dx.doi.org/10.1016/j.foodchem.2016. J = 8.1 Hz, H-5 U or 5 L), 6.80 (t, J = 8.2 Hz, H-5 U or 5 L), 6.88 (d, 0 0 0 0 0 06.009. J = 8.1 Hz, H-6 U and 6 L), 7.01 (s, H-2 U or 2 L), 7.11 (s, H-2 U or 20 ,0.5H). The retention time was in accordance with reference L References standard. The spectral data of compound 11 matched that of pro- cyanidin B4 in the literature (Mohri et al., 2009). The 1H NMR data Antoniolli, A., Fontana, A. R., Piccoli, P., & Bottini, R. (2015). Characterization of compared with reference are shown in Table S3 in Supplementary polyphenols and evaluation of antioxidant capacity in grape pomace of the cv. material. Malbec. Food Chemistry, 178, 172–178. Cheynier, V. (2002). Grape polyphenols and their reactions in wine. In S. Martens, D. Compound 12 (procyanidin B2). ESI-QTOF-MS (negative mode): Treutter, & Forkmann (Eds.). Polyphenols (pp. 1–14). Germany: Freising- m/z 577.15 ([M H] ) 425.04 405.03 289.06; phloroglucinolysis- Weihenstephan. 1-14. UPLC: terminal unit, epicatechin; extension unit, epicatechin. The Cheynier, V., Moutounet, M., & Sarni-Manchado, P. (1998). Les composés retention time and the MS data were in accordance with reference phénolique. In Claude London, Paris, New York (Ed.), Œnologie (pp. 123–162). New York): Lavoisier press. standard. Chu, Y. H., Chen, C. J., Wu, S. H., & Hsieh, J. F. (2014). Inhibition of xanthine oxidase Compound 13 (procyanidin B1). ESI-QTOF-MS (negative mode): BY Rhodiola crenulata extracts and their phytochemicals. Journal of Agriculture m/z 577.11 ([M H] ) 425.02 405.03 289.04; phloroglucinolysis- and Food Chemistry, 62, 3742–3749. Degenhardt, A., Hofmann, S., Knapp, H., & Winterhalter, P. (2000). Preparative UPLC: terminal unit, catechin; extension unit, epicatechin. The isolation of anthocyanins by high-speed countercurrent chromatography and retention time and the MS data were in accordance with reference application of the color activity concept to red wine. Journal of Agricultural and standard. Food Chemistry, 48, 5812–5818. Esatbeyoglu, T., & Winterhalter, P. (2010). Preparation of dimeric procyanidins B1, Compound 14 (procyanidin T2). ESI-QTOF-MS (negative mode): B2, B5, and B7 from a polymeric procyanidin fraction of Black Chokeberry m/z 865.17 ([M H] ) 577.10 289.03; phloroglucinolysis-UPLC: ter- (Aronia melanocarpa). Journal of Agricultural and Food Chemistry, 58, 5147–5153. L. Luo et al. / Food Chemistry 212 (2016) 712–721 721
Fanzone, M., Zamora, F., Jofré, V., Assof, M., Gómez-Cordovés, C., & Peña-Neira, A. Sun, B. S., Ferrão, C., & Spranger, M. I. (2003). Effect of wine style and winemaking (2012). Phenolic characterisation of red wines from different grape varieties technology on resveratrol level in wines. Ciência Téc. Vitiv, 18(2), 77–91. cultivated in Mendoza province (Argentina). Journal of the Science of Food and Sun, B. S., Leandro, M. C., Ricardo-da-Silva, J. M., & Spranger, M. I. (1998). Separation Agriculture, 92(3), 704–718. of grape and wine proanthocyanidins according to their degree of Hor, M., Heinrich, M., & Rimple, H. (1996). Proanthocyanidin polymers with polymerization. Journal of Agricultural and Food Chemistry, 46, 1390–1396. antisecretory activity and proanthocyanidin oligomers from Guazuma Ulmifolia Sun, B. S., Pinto, T., Leandro, M. C., Ricardo-Da-Silva, J. M., & Spranger, M. I. (1999b). bark. Phytochemistry, 40(1), 109–119. Transfer of catechins and proanthocyanidins from solid parts of the grape Ito, Y. (2005). Golden rules and pitfalls in selecting optimum conditions for high- cluster into wine. America Journal of Enology and Viticulture, 50(2), 179–184. speed counter-current chromatography. Journal of Chromatography A, 1065, Sun, B. S., Neves, A. C., Fernandes, T. A., Fernandes, A. L., Mateus, N., & De Freitas, V. 145–168. (2011). Evolution of phenolic composition of red wine during vinification and Jara-Palacios, M. J., Hernanz, D., Escudero-Gilete, M. L., & Heredia, J. F. (2014). storage and its contribution to wine sensory properties and antioxidant activity. Antioxidant potential of white grape pomaces: Phenolic composition and Journal of Agricultural and Food Chemistry, 59, 6550–6557. antioxidant capacity measured by spectrophotometric and cyclic voltammetry Sun, B. S., Ribes, A. M., Leandro, M. C., Belchior, A. P., & Spranger, M. I. (2006). methods. Food Research International, 66, 150–157. Stilbenes: quantitative extraction from grape skins, contribution of grape solid Jesionek, W., Majer-Dziedzic, B., & Choma, I. M. (2015). Separation, identification, to wine and variation during wine maturation. Analytica Chimica Acta, 563, and investigation of antioxidant ability of plant extract components using TLC, 382–390. LC–MS, and TLC–DPPH. Journal of Liquid Chromatography & Related Technologies, Sun, T., Chen, Q. Y., Wu, L. J., Yao, X. M., & Sun, X. J. (2012). Antitumor and 38(11), 1147–1153. antimetastatic activities of grape skin polyphenols in a murine model of breast Jin, S., Eerdunbayaer Doi, A., Kuroda, T., Zhang, G., Hatano, T., & Chen, G. (2012). cancer. Food and Chemical Toxicology, 50, 3462–3467. Polyphenolic Constituents of Cynomorium songaricum Rupr. And Antibacterial Suri, S., Taylor, M. A., Verity, A., Tribolo, S., Needs, P. W., Kroon, P. A., ... Hughes, D. A. Effect of Polymeric Proanthocyanidin on MethicillinResistant Staphylococcus (2008). A comparative study of the effects of quercetin and its glucuronide and aureus. Journal of Agriculture and Food Chemistry, 60, 7297–7305. sulfate metabolites on human neutrophil function in vitro. Biochemical Kemperman, R. A., Gross, G., Mondot, S., Possemiers, S., Marzorati, M., Wiele, T. V., ... Pharmacology, 76, 645–653. Doré, J. (2013). Impact of polyphenols from black tea and red wine/grape juice Shu, X. K., Duan, W. J., Liu, F., Shi, X. G., Geng, Y. L., Wang, X., & Yang, B. T. (2014). on a gut model microbiome. Food Research International, 53, 659–669. Preparative separation of polyphenols from the flowers of Paeonia lactiflora Pall. Kruger, M. J., Davies, N., Myburgh, K. H., & Lecour, S. (2014). Proanthocyanidins, by high-speed counter-current chromatography. Journal of Chromatography B, anthocyanins and cardiovascular diseases. Food Research International, 59, 947–948, 62–67. 41–52. Vidal, S., Cartalade, D., Souquet, J. M., Fulcrand, H., & Cheynier, V. (2002). Changes in Lu, Y., & Foo, L. Y. (1999). The polyphenol constituents of grape pomace. Food proanthocyanidin chain length in winelike model solutions. Journal of Chemistry, 65, 1–8. Agricultural and Food Chemistry, 50, 2261–2266. Makhotkina, O., & Kilmartin, P. A. (2012). The phenolic composition of Sauvignon Vivas, N., Nonier, M., Gaulejac, N. V., Absalon, C., Bertrand, A., & Mirabel, M. (2004). blanc juice profiled by cyclic voltammetry. Electrochimica Acta, 83, 188–195. Differentiation of proanthocyanidin tannins from seeds, skins and stems of Mohri, Y., Sagehashi, M., Yamada, T., Hattori, Y., Morimura, K., Hamauzu, Y., ... Kamo, grapes (Vitis vinifera) and heartwood of Quebracho(Schinopsis balansae) by T. (2009). An efficient synthesis of procyanidins using equimolar condensation matrix-assisted laser desorption/ionization time-of-flight mass spectrometry of catechin and/or epicatechin catalyzed by ytterbium triflate. Heterocycles, 79, and thioacidolysis/liquid chromatography/electrospray ionization mass 549–563. spectrometry. Analytica Chimica Acta, 513, 247–256. Mossalayi, M. D., Rambert, J., Renouf, E., Micouleau, M., & Mérillon, J. M. (2014). Wang, K., Liu, Z., Huang, J. A., Dong, X., Song, L., & Yu, P. (2008). Preparative isolation Grape polyphenols and propolis mixture inhibits inflammatory mediator and purification of theaflavins and catechins by high-speed countercurrent release from human leukocytes and reduces clinical scores in experimental chromatography. Journal of Chromatography B, 867, 282–286. arthritis. Phytomedicine, 21, 290–297. Wang, Y., Chen, Y., Deng, L., Cai, S., Liu, J., Li, W., ... Du, L. (2015). Systematic Moon, J.-H., Tsushida, T., Nakahara, K., & Terao, J. (2001). Identification of quercetin separation and purification of iridoid glycosides and crocetin derivatives from 3-O-b-D-glucuronide as an antioxidative metabolite in rat plasma after oral Gardenia jasminoides Ellis by high-speed counter-current chromatography. administration of quercetin. Free Radical Biology and Medicine, 30, 1274–1285. Phytochemical analysis, 26, 202–208. Phansalkar, R. S., Nam, J., Chen, S., McAlpine, J. B., Napolitano, J. G., Leme, A., ... Vidal, Xia, G. B., Hong, S., & Liu, S. B. (2014). Simultaneous preparation of naturally C. M. P. (2015). A galloylated dimeric proanthocyanidin from grape seed abundant and rare catechins by tannase-mediated biotransformation combining exhibits dentin biomodification potential. Fitoterapia, 101, 169–178. high speed counter current chromatography. Food Chemistry, 151, 380–384. Ribéreau-Gayon, P. (1982). Chapter 8 – The anthocyanins of grapes and wines. Yamazaki, S., Miyoshi, N., Kawabata, K., Yasuda, M., & Shimoi, K. (2014). Quercetin- anthocyanins as food colors, 209–244 . 3-O-glucuronide inhibits noradrenaline-promoted invasion of MDA-MB-231 Spranger, I., Sun, B. S., Mateus, A. M., Freitas, V., & Ricardo-da-Silva, J. M. (2008). human breast cancer cells by blocking b2-adrenergic signaling. Archives of Chemical characterization and antioxidant activities of oligomeric and Biochemistry and Biophysics, 557, 18–27. polymeric procyanidin fractions from grape seeds. Food Chemistry, 108, Yilmaz, Y., Göksel, Z., Erdogan, S. S., Öztürk, A., Atak, A., & Özer, C. (2015). 519–532. Antioxidant activity and phenolic content of seed, skin and pulp parts of 22 Shibusawa, Y., Yanagida, A., Isozaki, M., Shindo, H., & Ito, Y. (2001). Separation of grape (Vitis Vinifera L.) cultivas (4 common and 18 registered or candidate for apple procyanidins into different degrees of polymerization by high-speed registration). Journal of Food Processing and Preservation, 39, 1682–1691. counter-current chromatography. Journal of Chromatography A, 915, 253–257. Zhang, S., Cui, Y., Li, L., Li, Y., Zhou, P., Luo, L., & Sun, B. (2015). Preparative HSCCC Sun, B. S., Belchior, G. P., Ricardo-da-Silva, J. M., & Spranger, M. I. (1999). Isolation isolation of phloroglucinolysis products from grape seed polymeric and purification of dimeric and trimeric procyanidins from grape seeds. Journal proanthocyanidins as new powerful antioxidants. Food Chemistry, 188, of Chromatography A, 841, 115–121. 422–429. 本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。
学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，
提供一站式文献检索和下载服务”的24 小时在线不限IP 图书馆。 图书馆致力于便利、促进学习与科研，提供最强文献下载服务。
图书馆导航：
图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具