Food Chemistry 129 (2011) 1331–1342
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Changes in antioxidant phytochemicals and volatile composition of Camellia sinensis by oxidation during tea fermentation Youngmok Kim a,⇑, Kevin L. Goodner a, Jong-Dae Park b, Jeong Choi b, Stephen T. Talcott c a
Sensus Research and Development Center, Sensus, LLC, 7255 Hamilton Enterprise Park Drive, Hamilton, OH 45011, United States Tea Research Institute of JARES (Jeonnam Agricultural Research and Extension Services), Jeonnam, South Korea c Department of Nutrition and Food Science, Texas A&M University, 1500 Research Parkway A, College Station, TX 77843, United States b
a r t i c l e
i n f o
Article history: Received 8 November 2010 Received in revised form 17 March 2011 Accepted 3 May 2011 Available online 8 May 2011 Keywords: Green tea Oolong tea Black tea Tea fermentation Polyphenolics Caffeine Flavonoids Catechins Total soluble phenolics Antioxidant capacity Tea volatile compounds
a b s t r a c t Monomeric flavonoids (flavan 3-ols or tea catechins) present in Camellia sinensis leaf are transformed to polymeric theaflavin and thearubigin by oxidation occurring during tea fermentation. The distinctive colour, decreased bitterness and astringency, and characteristic flavour are derived from the fermentation process giving fermented teas a marked distinction from non-fermented green tea. Even though teas are available in many different fermentation levels from green to black, the difference in phytochemicals and volatile compounds in teas with different degrees of fermentation has not been fully investigated yet within the same tea leaf. The objective of this study was to observe non-volatile phytochemicals including polyphenolic and volatile compounds changes by oxidation under strict processing control and to evaluate the degree of fermentation for the maximum antioxidant capacity with the same tea material. Harvested tea leaf was immediately processed to different degrees of oxidative fermentation (0%, 20%, 40%, 60%, and 80%). Tea infusions brewed with each processed tea leaf were analysed for polyphenolic profile, total soluble phenolics, antioxidant capacity, and volatile profile using LC–MS, HPLC, Folin–Ciocalteu assay, Oxygen Radical Absorbance Capacity (ORAC), and GC–MS analyses. The flavonoids in non-fermented green tea were significantly lessened during the oxidative fermentation process and the decreased monomeric flavonoids were transformed to polymeric theaflavin and thearubigin as the leaves were more processed. Total soluble phenolics and antioxidant capacity were significantly higher as tea leaves were less processed with a high correlation with individual polyphenolic changes. Volatile compounds present in tea leaf were analysed by GC–MS to observe changes due to processing and were utilised to create a model to differentiate fermentation based on volatile composition. Twenty-four compounds were used to build an initial model which was optimised to 16 compounds with complete separation of the groups using discriminant function analysis. The data suggested that fermentation diminished antioxidant capacity of tea and could result in lowering potential health benefits from flavonoids. This result should be considered for tea manufacturing and the development of functional foods desiring maximum potential health benefits from antioxidant flavonoids in tea. Ó 2011 Published by Elsevier Ltd.
1. Introduction Tea (Camellia sinensis) is the most consumed drink in the world next to water and the amount of consumption well exceeds coffee, beer, wine, and soft drinks (Rietveld & Wiseman, 2003). The reasons for the worldwide popularity of tea were unique aroma and characteristic flavour but recently, its popularity has increased due to its potential health benefits against cardiovascular diseases and cancer as well as pharmaceutical activities such as antihypertensive, antiateriosclerotic, hipocholesteroladmic, and hypolipi-
⇑ Corresponding author. Tel.: +1 513 892 7100; fax: +1 513 892 7111. E-mail address: ymk@sensusflavours.com (Y. Kim). 0308-8146/$ - see front matter Ó 2011 Published by Elsevier Ltd. doi:10.1016/j.foodchem.2011.05.012
demic properties mostly from activities of antioxidant flavonoids present in tea (Chan et al., 1999; Chen, Zhu, Tsang, & Huang, 2001; Cheng, 2006). Teas from the genus Camellia can be divided into three categories based on tea manufacturing (fermentation) process: green tea (unfermented), oolong tea (partially fermented), and black tea (fully fermented). Green tea has been regarded as a rich source of catechin and its derivatives called tea catechins or flavan 3-ols including (+)-catechin (C), ()-epicatechin (EC), ()-epigallocatechin (EGC), ()-epicatechin gallate (ECG), ()-epigallocatechin gallate (EGCG), and ()-gallocatechin gallate (GCG) that contribute to antioxidant capacity and organoleptic properties (Chen, Zhu, Wong, Zhang, & Chung, 1998; Fernández, Martín, González, & Pablos, 2000). For green tea manufacturing, freshly plucked tea
1332
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
leaves are immediately steamed or pan-fired to inactivate polyphenol oxidase and native microflora that initiate and catalyses the aerobic oxidation of tea catechins during tea fermentation. Whereas, fresh tea leaves are crushed and allowed to wither to induce oxidation as a part of tea fermentation process prior to drying for the black tea manufacturing process (generally more than 80% fermented). The characteristic reddish-black colour, reduced bitterness and astringency, and removal of leafy and grassy flavour are derived from this oxidation process giving black tea a marked distinction from green tea (Cheng, 2006; Wang, Provan, & Helliwell, 2000). While C. sinensis leaf is fermented, monomeric flavan-3-ols including EGC, EGCG, EC, and ECG undergo oxidative polymerisation converting into theaflavins as illustrated in Fig. 1. Phenolic flavan 3-ols in fresh tea leaf are transformed into two principal groups of phenolic pigments (red–orange coloured theaflavin and rusty-red coloured thearubigin) in black tea by fermentation, which is a natural browning reaction initiated by an oxidative enzyme (polyphenol oxidase) within the plant cell (Haslam, 2003). Semi-fermented oolong tea is generally fermented from 20% to 60% to avoid green tea’s characteristic leafy and grassy notes while obtaining black tea’s sweet and bold flavour. Oolong tea was traditionally reported to have anti-obesity and hypolipidemic effects coming from unoxidised tea catechins and oxidised theaflavins and thearubigins present in oolong tea (Han, Takaku, Limura, & Okuda, 1999). The difference of polyphenolics between non-fermented and partially or fully fermented teas has been reported by studies using tea materials procured from an unknown source that does not provide information on the degree of processing, harvested area and time. Little information is available on the polyphenolic changes
during fermentation especially when the starting tea material is the same and the tea fermentation process was strictly controlled to obtain the desired degree of the fermentation. Moreover, changes of volatile compounds present in tea were observed since only limited information is available on volatile compounds changes by tea fermentation. The objective of this study is to investigate the changes of volatile and non-volatile chemical compounds by different degrees of fermentation and to determine the oxidation degree for the maximum level of antioxidant capacity from the same tea material.
2. Materials and methods 2.1. Tea harvest and processing Fresh tea leaf was harvested in July 30th, 2009 in Boseong, Korea and immediately processed by JARES (Jeonnam Agricultural Research and Extension Services). Teas were divided into five groups for five different fermentation processes. (1) Non-fermented tea (Green tea) – 0% fermentation Fresh leaves were panned at 180 °C for 5 min prior to rolling for 10 min and then transferred to a drum dryer for four multiple stage drying processes at different temperatures and durations. The rolled leaves were first dried at 150 °C for 3 min (1st drying) followed by additional drying processes at 130 °C for 3 min (2nd drying) and 110 °C for another 3 min. The leaves were finally dried at 80 °C for 25 min (4th drying) to complete green tea production.
Fig. 1. Oxidation scheme of flavan-3-ols (tea catechins) in green tea to theaflavin in black tea. Number assignment: (1) ()-epigallocatechin (EGC), (2) ()-epigallocatechin gallate (EGCG), (3) orthoquinone derivative of ECG (R = H) and EGCG (R = galloyl group), (4) ()-epicatechin (5) ()-epicatechin gallate, (6) theaflavin. Abbreviations: PPO: polyphenol oxidase.
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
(2) Partially-fermented tea (Oolong tea) 20% fermentation Fresh leaves were first withered by solar light for 30 min at ambient temperature prior to additional indoor withering at ambient temperature for 3 h. Then, the leaves were fermented via mass breaking stage until the desired degree of fermentation (20%) was achieved. The fermented leaves were sent to four multiple drying stages as described in green tea production. (3) Partially-fermented tea (Oolong tea) – 40% fermentation This semi-fermented tea was processed as described in the 20% fermented tea, except that the mass-breaking stage was held until 40% fermentation was achieved. (4) Partially-fermented tea (Oolong tea) – 60% fermentation
This semi-fermented tea was processed as described in the 20% fermented tea, except that the mass-breaking stage was held until 60% fermentation was achieved. (5) Fully fermented tea (Black tea) – 80% fermentation
This fully-fermented tea was processed as described in the 20% fermented tea, except that the mass-breaking stage was held until 80% fermentation was achieved. The degree of fermentation (% fermentation) was determined by measuring the ratio of fermented and unfermented area of tea leaf based on colour changes by oxidative fermentation. Processed tea leaf was immediately vacuum-packed and stored at room temperature without exposure to light until analysis. 2.2. Preparation of tea infusion Five tea infusions from the leaves with different fermentation degrees were prepared by brewing 5 g of tea leaves with 95 ml of previously purified water using a Milli-Q water system (Billerica, MA) at 90 °C for 10 min with constant stirring. The brewed teas were immediately filtered through Whatman #4 filter paper followed by a 1 cm bed of diatomaceous earth under a slight vacuum to remove suspended particles. After cooling down to room temperature, tea infusions were frozen at 20 °C until analysis. 2.3. Polyphenolic identification using LC–ESI-MSn analysis Mass spectrometric analysis was used in this study to identify individual polyphenolics present in C. sinensis teas based on molecular masses and fragment ions. All tea infusions were previously filtered through a 0.45 lm PTFE syringe filter (Whatman, Clifton, NJ) prior to injection. Polyphenolics present in tea infusions were analysed using a Thermo Finnigan LCQ Deca XP Max MSn ion trap mass spectrometer system (ThermoFisher, San Jose, CA) with Finnigan Surveyor PDA plus detector (ThermoFisher, San Jose, CA), Finnigan Surveyor MS pump, and Finnigan Surveyor Autosampler plus. Polyphenolics in tea infusion were separated on the Dionex 250 4.6 mm Acclaim 120-C18 column. Two mobile phases were comprised of Phase A (100% H2O w/0.5% formic acid (5 mM ammonium formate), v/v) and Phase B (100% methanol w/0.5% formic acid, v/v) and run at 0.6 ml/min. Polyphenolics were separated by the following gradient system: 0–50% for 10 min (B); 50–70% for 45 min (B); 70–100% for 60 min (B) and then returned to the original condition for 1 min (100% A) and held for 9 min to equilibrate the column prior to the next injection (60 min total run time + 10 min equilibration time). Mass analysis was obtained in negative
1333
ion mode with an atmospheric pressure electrospray ionisation (ESI) source. The electrospray needle voltage was 3300 V. Nitrogen was used as sheath gas and auxiliary gas at flow rate of 60 and 5 units/min, respectively. Capillary voltage and gas temperature were 1.5 V and 250 °C, respectively. Mass spectrometry data was obtained in the full scan mode (m/z 200–2000) and fragment ions were additionally acquired in MS2 and MS3 mode. 2.4. Polyphenolic identification and quantification using HPLC-DAD analysis Polyphenolics present in each tea infusion were additionally analysed by HPLC as described by Lee and Talcott (2002) with a slight modification. Properly diluted and filtered tea infusion was injected into an Agilent 1200 HPLC system equipped with an Agilent Diode Array Detector (DAD) detector and polyphenolic separation was performed on a Dionex 250 4.6 mm Acclaim 120-C18 column running at 0.8 ml/min. A gradient mobile phase consisted of Phase A (100% H2O) and Phase B (60% methanol and 40% H2O) each adjusted to pH 2.4 using o-phosphoric acid. The binary gradient system started by running 0–30% Phase B over 30 min, 30–80% Phase B for 15 min, 80–100% Phase B for 15 min for a total run time of 60 min. The column was equilibrated with 100% phase A for 5 min prior to the next sample injection. Phenolic compounds besides flavonol glycosides were detected and quantified at 280 nm while flavonol glycosides were found at 360 nm. All the detected phenolic compounds were analysed against external standards of gallic acid, caffeine, theobromine, rutin, kaempferol, and quercetin procured from Sigma–Aldrich (St. Louis, MO) and C, EC, EGC, EGCG, and ECG purchased from Chromadex (Irvine, CA). 2.5. Quantification of total theaflavin and thearubigin by spectrometric methods Total theaflavin (TF) was determined using the Flavognost method as described by Obanda, Owuor and Mang’oka (2001). Briefly, 10 ml of previously prepared tea infusion was mixed with 10 ml of IBMK (isobuthylmethylketone) solution using a separatory funnel and the mixture was allowed to stand for 15 min for complete separation after shaking for 10 min. Two millilitres of IBMK layer (upper layer) was taken into a test tube which contained 4 ml of ethanol and 2 ml of Flavognost reagent. After shaking, the mixture remained at room temperature for 15 min for complete colour development and the absorbance was measured at 625 nm using a Genesys 6 spectrometer (ThermoFisher, San Jose, CA). IBMK solution was procured from Sigma–Aldrich (Sigma Chemical Co., St. Louis, MO). The Flavognost reagent was prepared by dissolving 2 g of diphenylboric acid-2-aminoethyl ester in 100 ml of ethanol. The measured absorbance was introduced into the following equation for total TF.
Total TF ¼ A625 47:9 DM=100 Total thearubigin (TR) was quantified spectrophotometrically as described by Obanda, Owuor, Mang’oka, and Kavoi (2004). For both TF and TR, green tea (0% fermentation) was assumed to contain no TF and TR without analysis since this assay was developed for the compounds in fermented tea which contains polymeric polyphenolics including theaflavins and thearubigins. The concentrations of TF and TR were considered as zero in green tea as indicated by previous studies. 2.6. Determination of antioxidant capacity and total soluble phenolics Antioxidant capacity of each tea infusion was measured using the Oxygen Radical Absorbance Capacity (ORAC) run according to Talcott and Lee (2002) adapted to work with a 96-well microplate
1334
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
and BMG Labtech FLUOSTAR Optima microplate reader (Offenburg, Germany) used to monitor fluorescence decay over time. Total soluble phenolics (TSP) concentration including contributions from ascorbic acid was measured using Folin–Ciocalteu assay (Swain & Hillis, 1959). The absorbance of each tea infusion was read using a Genesys 6 spectrometer (ThermoFisher, San Jose, CA) at 726 nm. Total soluble phenolics of tea infusions were quantified based on a linear regression against a gallic acid standard curve and data was expressed in mg/l gallic acid equivalents (GAE). 2.7. Colour analysis of tea infusions Teas for colour analysis were brewed by mixing hot water and 10 g of each tea leaf with 90 g of water. After 10 min of infusion at 90 °C, the tea was filtered through cheesecloth and then allowed to cool to 25 °C for final filtrations through Whatman #4 filter paper to remove suspended particles. The colour values (CIELAB) of each tea infusion were measured with a Minolta CR400 colourimeter (Ramsey, NJ) using standard cuvettes after proper calibration. Each colour value (L⁄, a⁄, b⁄) was measured three times. 2.8. Evaluation of volatile compounds using GC–MS analysis Volatile analysis was performed on the brewed tea used for the colour analysis. The samples were prepared for analysis by pipetting 4.95 ml of tea along with 50 ll of internal standard (phenol-D6, 1000 ppm) into a 20 ml headspace vial. A Gerstel MultiPurposeSampler (MPS-2) (Baltimore, MD) was used with a 2 cm 3-phase SPME fiber (divinylbenene, Carboxen, Polydimethylsiloxane) for sample preparation. A 10-min incubation followed by a 40 min exposure was used to capture the volatiles on the fiber for injection into the GC. The sample was stirred using a 3 12 mm stir bar in the 20 ml vial. The fiber was desorbed in the GC injector for 5 min. An Agilent 7890A gas chromatograph (Palo Alto, CA) was used for the analysis. Analysis was performed in the splitless mode with a helium flow rate of 1.25 ml/min through a 60 m 0.25 mm 0.25 lm RTX5 ms column. The initial oven temperature was 50 °C immediately followed by a 4 °C/min temperature ramp to 170 °C which was followed by a 100 °C/min ramp to 250 °C and held for 5 min in order to ensure no sample to sample contamination. The transfer line to the Leco TruTOF MS (St. Joseph, MN) was held at 240 °C. Data was collected for 20–250 m/z at an acquisition rate of 10 spectra per sec. Compounds were identified by using retention indices, spectral matches based on the NIST05 database, and by comparison to authentic standards. Peak areas were normalised to the internal standard response for each analysis for statistical analyses. 2.9. Statistical analysis Data represent the mean triplicate analysis using ANOVA (analysis of variance) with JMP 5 statistical software. Mean separation was conducted using the LSD test (P < 0.05). Correlations (r) of data was evaluated using Pearson’s correlation analysis. Volatile analysis was performed using Statistica Version 8. 3. Results and discussion 3.1. Polyphenolic identification by LC–ESI-MSn analysis A total of 16 phenolic compounds and 2 methylxanthines (theobromine and caffeine) were identified by LC–ESI-MSn analysis based on retention time, absorbance spectrum, and mass fragmentation pattern (Fig. 2). Mass fragmentation data, retention time and spectrum information were displayed in Table 1.
Peak 1 was identified as gallic acid, a common phenolic compound of tea, which was previously observed in green tea by Del Rio et al. (2004). The presence of a positively charged molecular ion at m/z 169 was revealed by MS/MS analysis and it was additionally defragmented to yield MS2 spectrum with an ion at m/z 125. Peak 2 was identified as 5-galloylquinic acid and the presence of this compound in green tea was earlier confirmed by Kuhr and Engelhardt (1991). The parent ion [MH] was observed at m/z 343 and it was additionally fragmented to ions at m/z 191 (the most intense ion and the deprotonated form of a quinic acid) and 169 (gallic acid residue) by MS2 analysis. Peak 3 and 7 were identified as theobromine and caffeine and the presence of these compounds in green tea was previously reported by Ahmad and Mukhtar (1999), Del Rio et al. (2004), and Gupta, Saha, and Giri (2002). As previously reported (Del Rio et al., 2004 and Stewart, Mullen, & Crozier, 2005), theobromine and caffeine were also poorly ionised in negative ion mode in this study. The identification of these two methylxanthines was confirmed by comparing retention time and absorbance spectra with authentic standards. Peak 4 was determined to be ()-gallocatechin which was previously observed in green tea by Del Rio et al. (2004). The parent ion was obtained at m/z 305 and MS2 fragment yielded ions at m/z 179, 221, and 261 as previously observed by Stewart et al. (2005). Peak 5 was identified as ()-epigallocatechin even though it has the same [MH] at 305 and fragmentations at m/z 179, 221, and 261 with ()-gallocatechin (Peak 4). The identification of ()-epigallocatechin was determined based on retention time and kmax of authentic ()-epigallocatechin standard. Peak 6 was identified as ()-epigallocatechin gallate which is known as the most prevalent tea catechin in green tea (Gupta, Ahmad, Nieminen, & Mukhtar, 2000; Gupta et al., 2002). The parent ion [MH] at m/z 457 was observed and ions at m/z 169 (gallic acid), 305 (epigallocatechin moiety), and 331 were additionally obtained by MS2 fragmentation. Then, fragment ion at m/z 125 (dihydroxy phenol moiety) was obtained after gallic acid fragmentation by MS3 analysis. Peak 8 was determined as ()-epicatechin based on its fragmentation pattern of [MH] at m/z 289, retention time, and similar absorbance spectra with authentic standard. It was additionally fragmented to yield MS2 fragment ions at m/z 179 and 205. Peak 9 was identified as ()-epicatechin gallate due to its similar absorbance spectrum, retention of authentic standard, and MS fragmentation pattern. The parent ion [MH] was obtained at m/z 441 and fragment ions at m/z 289 was yielded by MS2 analysis. Finally, the ions at m/z 169 (free gallic acid) and 125 (dihydroxy phenol moiety) were obtained by MS3 fragmentation. Peak 10 and 11 were tentatively identified as apigenin glycosides that have been found in various plants including green and black teas (Dubick & Omaye, 2007; Janssen et al., 1998; Engelhardt, Finger & Kuhr, 1993). Most of the aglycone of the glycosylflavonoids is either apigenin or luteolin and likewise, the compounds detected in the present study may be apigenin glycosylflavonoids based on its fragment pattern. The [MH] was obtained at m/z 563 and ions were yielded at m/z 353, 443, 473, and 503 by MS2 fragmentation. Especially, ions at m/z 443, 473, 503 represents [MH120], [MH90], and [MH60] and it indicated the presence of pentose substitution in both peaks (Ferreres, Silva, Andrade, Seabra, & Ferreira, 2003). Peak 12 was identified as myricetin 3-glycoside with [MH] at m/z 479 and MS2 fragment at m/z 317 by losing a sugar molecule as observed by Atoui, Mansouri, Boskou, and Kefalas (2005) who found the same compounds in green tea by MS analysis.
1335
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
Relative Absorbance
RT: 0.00 - 60.00 100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0
7
280nm
NL: 1.07E6 Channel A UV 1_A0_20X 6
9
2
8
1 0
5
10
15
4 5
3 20
25
30
35
40
45
50
55
60
Time (min)
Relative Absorbance
RT: 0.00 - 60.00 13
100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0
NL: 9.88E4 Channel C UV 1_A0_20X
360nm 15 14 12 16
17 18
10 11
0
5
10
15
20
25
30
35
40
45
50
55
60
Time (min) Fig. 2. Chromatogram determined by HPLC-DAD analysis at 280 and 360 nm of phenolic compounds present in non-fermented tea infusion (A0). For peak assignment see Table 1.
Table 1 LC/ESI/MSn characteristics of phenolic compounds and methylxanthins present in tea infusions. Peak 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 a b
RT 15.40 16.60 21.95 26.01 26.31 29.82 30.37 31.45 34.73 36.42 36.97 38.12 39.22 39.77 40.15 40.47 40.85 41.10
271 201, 230, 215, 215, 215, 233, 213, 215, 214, 215, 233, 215, 233, 218, 215, 215, 215.
[MH]
Compound
kmax 227, 273 235, 233, 234, 271 234, 233, 233, 233, 301, 233, 255, 233, 234, 234, 234,
274 270 281 274 279 277 272 271 358 355 293 266 265 266 265
Gallic acid 5-galloylquinic acid Theobromine ()-Gallocatechin (GC) ()-Epigallocatechin (EGC) ()-Epigallocatechin gallate (EGCG) Caffeine ()-Epicatechin (EC) ()-Epicatechin gallate (ECG) Apigenin glycoside Apigenin glycoside Myricetin 3-glycoside Quercetin 3-rutinoside (Rutin) Quercetin 3-glucosyl-rhamnosyl-galactoside Kaempferol glycoside Kaempferol glycoside Kaempferol 3-rutinoside Kaempferol 3-rutinoside
a
169 343 N/D 305 305 457 N/D 289 441 563 563 479 609 771 755 755 593 593
MS2 (m/z)
MS3 (m/z) b
125 169, 191
179, 221, 261 179, 221, 261 169, 305, 331 179, 184, 353, 443, 473, 535, 383, 473,
205 318 503 503 317 301 301 285 285 285 285
125
353, 353, 179, 271, 151,
239 383 383 316 179 162
162 162
References fragmentation patters provided in the accompanying discussion. Ions in boldface indicate the most intense product ion for MS2 fragmentation.
Peak 13 was identified as quercetin 3-rutinoside as many studies have reported its presence in green tea (Markowicz Bastos et al., 2007; Stewart et al., 2005; Yen & Chen, 1995). Its identification was confirmed by its fragmentation pattern and comparing retention time and absorbance spectra of an authentic standard. The parent ion [MH] at m/z 609 was first obtained and an ion at m/z 301 was yielded by losing a sugar molecule of glucose and rhamnose from aglycone quercetin.
Peak 14 was tentatively identified as quercetin 3-glucosylrhamnosyl-galactoside based on its fragmentation pattern. The parent ion [MH] was obtained at m/z 771 and MS2 and MS3 fragmentation ions at m/z 301 and 162 indicates loss of aglycone quercetin and presence of a hexosyl residue, respectively as reported by Dou, Lee, Tzen, and Lee (2008). Peak 15 and 16 were identified as kaempferol glycosides due to its fragmentation pattern. The parent ion [MH] at m/z 755 was
1336
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
obtained from both peaks and [MH] at m/z 285 was additionally obtained by MS2 fragmentation. It indicates that one glycoside which can be glucose, galactose, or glucose-rhamnose was lost by MS2 fragmentation from aglycone kaempferol. Peak 17 and 18 were determined as kaempferol 3-rutinosides because of its parent ion [MH] at m/z 593 and MS2 fragmentation at m/z 285 indicating loss of a hexose-rhamnose molecule. This fragmentation pattern was early reported by Atoui et al. (2005) and Kiehne and Engelhardt (1996). 3.2. Changes in phenolic compounds and methylxanthines by different tea processing During tea fermentation from green tea (0%) to black tea (80%), four major tea catechins including EGCG, EGC, EC, and ECG decreased by 74%, 91%, 51%, and 62%, respectively while gallic acid concentration increased by 1.64-fold (Fig. 3). Galloyl groups of EGCG and/or ECG were cleaved during oxidative fermentation and resulted in increase of free gallic acid, which is in agreement with Zuo, Chen, and Deng (2002). More free gallic acids were observed during tea fermentation because more gallated catechins were transformed to non-gallated catechins by releasing free gallic acid before forming theaflavin. It was observed that gallo-flavanols (EGC and EGCG) decreased more than catechol-flavanols (EC and ECG) did during oxidative fermentation. One hydrogen atom from many of the hydroxyl groups on flavonoids is lost during oxidative fermentation and it forms a semiquinone radical with an unpaired electron on the oxygen atom (Kim, 2008). Gallo-flavanols are known to be less stable to oxidation during storage because the radical is more freely formed on gallo-flavanols which contains three hydroxyl groups on the B ring (Wang et al., 2000; Yoshioka et al., 1991). Likewise, it was hypothesised that more gallo-flava-
nols were degraded by oxidative fermentation due to its higher reactivity than catechol-flavanols. Total theaflavin and thearubigin contents gradually increased as tea leaves were fermented and especially the increase of theaflavin was negatively correlated with the decrease of EGCG, EGC, EC, and ECG (r = 0.90, 0.94, 0.91, and 0.92, respectively) (Table 2). Total thearubigin which is produced by further fermentation after theaflavin is generated, was also correlated with total theaflavin (r = 0.83) (Table 3). Infusion colour (darkness) and spectrophotometric brightness (turbidity) were affected by two phenolic pigments (theaflavin and thearubigin), but theaflavin was more impactful on the darkness and turbidity compared to thearubigin. The correlation (r) between infusion colour and brightness and theaflavin was 0.97 and 0.91, respectively while it was 0.72 and 0.61 between infusion colour and brightness and thearubigin. Flavonol glycosides which account for up to 18% total phenolic compounds present in green tea were also reduced during tea fermentation (Table 4). Total flavonol glycosides were reduced by 38% during tea fermentation and it is in agreement with Wang and Helliwell (2001) who reported higher concentrations of flavonol aglycones, myricetin, quercetin, and kaempferol in green tea than black tea. Reduction in flavonol glycosides during fermentation might be due to oxidative degradation. Decline in flavonol glycosides, including rutin, which is the most prevalent flavonol glycoside (36% of total flavonol glycoside) by various processing methods was previously reported by Makris and Rossiter (2000) and Buchner, Krumbein, Rohn, and Kroh (2006) who reported this occurring during the oxidative degradation of compounds under various oxidative conditions. Both methylxanthines (caffeine and theobromine) found in tea leaf, decreased during tea fermentation (Table 2). This result was somewhat unexpected because many studies which compared caf-
500
Gallc acid
450
EGCG EGC
400 EC 350
ECG
mg/L
300 250 200 150 100 50 0 A0
A20
A40
A60
A80
Tea processing Fig. 3. Changes of EGCG, EGC, EC, ECG and gallic acid concentrations by oxidation during tea fermentation process.
Table 2 Concentrations of methylxanthins, total soluble phenolics, and antioxidant capacity of variously fermented teas.
Theobromine (mg/l)a Caffeine (mg/l) TSP (mg/l)b Antioxidant capacity (lmol TE/ml)c a b c d e
A0
A20
A40
A60
A80
60.47 ± 2.42ad,e 1357.28 ± 0.07a 5975.81 ± 56.45a 71.08 ± 0.38a
55.03 ± 9.70a 1260.85 ± 41.90ab 4766.13 ± 21.19b 66.01 ± 0.78c
42.27 ± 4.72b 1176.05 ± 68.31bc 4690.86 ± 56.64b 68.32 ± 0.48b
35.55 ± 0.34b 1097.28 ± 29.56 cd 3763.44 ± 38.11c 65.32 ± 1.27c
37.12 ± 2.23b 1030.79 ± 0.50d 3752.69 ± 56.64c 57.18 ± 1.51d
Concentrations of methylxanthins (theobromine and caffeine) were determined by HPLC-DAD analysis. Total Soluble Phenolics (TSP) was determined using Folin–Ciocalteu assay and the values were expressed as gallic acid equivalent. Antioxidant capacity was measured by Oxygen Radical Absorbance Capacity (ORAC) method and the values were expressed as Trolox Equivalent (TE). Values were expressed as mean ± standard deviation (n = 3). Mean values with similar letters within rows are not significantly different during tea fermentation (LSD test, P < 0.05).
1337
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342 Table 3 Concentrations of black tea polyphenolics in tea infusions by different degree of fermentation and brightness of each infusion. A0 a
Total Theaflavin (lmol/g) Total Thearubiginc (%) Spectrometer brightnessd (%) Infusion color (no unit) a b c d
A20 b
0 0 0 1.08 ± 0.03d
2.13 ± 0.06d 7.36 ± 0.04c 0.54 ± 0.01b 1.23 ± 0.02 cd
A40
A60
A80
2.46 ± 0.06c 7.44 ± 0.06c 0.61 ± 0.00b 1.47 ± 0.03c
4.09 ± 0.10b 7.65 ± 0.07b 1.64 ± 0.09a 1.73 ± 0.00b
5.18 ± 0.10a 7.94 ± 0.01a 1.68 ± 0.01a 2.25 ± 0.31a
Concentrations of TF (Theaflavin) was determined by Flavognost analysis. Mean values with similar letters within rows are not significantly different during tea fermentation (LSD test, P < 0.05). Total TR (Thearubigin) was measured spectrophotometrically. Spectrometer brightness and infusion color were determined based on the values measured at 460 nm using Genesys 6 spectrometer.
Table 4 Flavonol glycosides in teas determined by HPLC-DAD analysis. Polyphenolic concentration was expressed as mg/l.
a b
No.
Compounds
A0
A20
A40
A60
A80
1 2 3 4 5 6 7 8 9
Apigenin glycoside_1 Apigenin glycoside_2 Myricetin 3-glycoside Quercetin 3-rutinoside (Rutin) Quercetin 3-glucosyl-rhamnosyl-galactoside Kaempferol glycoside_1 Kaempferol glycoside_2 Kaempferol 3-rutinoside_1 Kaempferol 3-rutinoside_2
8.25 ± 0.27aa,b 5.71 ± 0.32a 43.50 ± 1.21a 103.01 ± 0.89a 41.55 ± 0.39a 42.20 ± 0.44a 26.65 ± 0.67a 8.81 ± 0.80a 6.80 ± 0.40a
8.87 ± 0.22a 6.14 ± 0.20a 37.72 ± 1.34b 89.07 ± 1.77b 34.41 ± 0.37b 30.11 ± 0.39b 24.01 ± 0.09bc 5.59 ± 0.44c 6.03 ± 0.48b
7.75 ± 0.65b 4.77 ± 0.28b 33.56 ± 4.45b 73.74 ± 8.34c 32.48 ± 2.73b 30.57 ± 3.73b 21.82 ± 2.44c 6.36 ± 0.86bc 5.43 ± 0.48b
6.97 ± 0.18c 4.53 ± 0.31b 20.82 ± 2.96c 64.71 ± 2.16d 27.65 ± 0.92c 32.01 ± 1.22b 22.24 ± 0.95bc 6.62 ± 0.78bc 5.50 ± 0.26b
6.76 ± 0.23c 4.53 ± 0.23b 13.73 ± 1.47d 56.75 ± 0.73e 26.49 ± 0.76c 33.14 ± 0.43b 24.22 ± 0.20b 7.23 ± 0.28b 5.90 ± 0.40b
Values were expressed as mean ± standard deviation (n = 3) and rutin equivalent. Mean values with similar letters within rows are not significantly different during tea fermentation (LSD test, P < 0.05).
Table 5 CIELAB colour values of the teas with different fermentation level. A0 ⁄
L a⁄ b⁄ a b
a,b
26.69 ± 0.04a 1.58 ± 0.07d 7.30 ± 0.06a
A20
A40
A60
A80
25.32 ± 0.06b 0.02 ± 0.06c 6.57 ± 0.04b
25.07 ± 0.01c 0.72 ± 0.04b 6.21 ± 0.04c
24.77 ± 0.02d 1.61 ± 0.05a 6.07 ± 0.03d
23.31 ± 0.02e 1.57 ± 0.05a 4.46 ± 0.01e
Values were expressed as mean ± standard deviation (n = 3) and rutin equivalent. Mean values with similar letters within rows are not significantly different during tea fermentation (LSD test, P < 0.05).
feine and theobromine contents in green and black teas reported higher methylxanthine concentrations in black tea than green tea (Cabrera, Giménez, & López, 2003; Khokhar & Magnusdottir, 2002; Lin, Tsai, Tsay, & Lin, 2003). However, caffeine and theobromine concentrations in green tea were higher when the starting material was the same because they were likely to be degraded during tea fermentation process and this observation is in agreement with Astill, Birch, Dacombe, Humphrey, and Martin (2001) and Del Rio et al. (2004). Thus, the reason why black tea was reported to contain more caffeine than green tea might not be because caffeine is generated during tea fermentation but because other factors such as plant variety (geographical difference), leaf age, leaf quality, and extractability may have caused this result. 3.3. Changes in total soluble phenolics and antioxidant capacity by tea processing A Folin–Ciocalteu assay was run to measure total soluble phenolics with gradually reduced during tea fermentation process. After 20, 40, 60, and 80% fermentation, total soluble phenolics were decreased by 20.24, 21.50, 37.02, and 37.20%, respectively (Table 2). The reduction rate of total soluble phenolics by tea fermentation was positively correlated with the decreased rate of EGCG, EGC, EC, ECG and total flavonol glycoside (sum of flavonol glycosides) (r = 0.84, 0.89, 0.82, 0.87, and 0.95, respectively). However, the Folin–Ciocalteu assay did not reflect the changes of black tea tannins such as total TF and TR because they were negatively correlated with total soluble phenolics (r = 0.98 and 0.88). This phenomenon was observed early by Appel, Govenor, D’Ascenzo, Siska and
Table 6 Compounds used in the analysis of volatile compounds of the tea along with retention indices. Peak #
Name
Retention index
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Dimethyl sulphide Propanal, 2-methyl2-Butanone Furan, 2-methylButanal, 3-methylButanal, 2-methyl1-Penten-3-ol 1-Penten-3-one Pentanal Furan, 2-ethyl1-Pentanol 2-Penten-1-ol, (Z)Toluene Hexanal Ethylbenzene p-Xylene Heptanal Phenol-d6- (internal standard) 5-Hepten-2-one, 6-methylFuran, 2-pentylCyclohexanone, 2,2,6-trimethylUnknown Linalool Unknown Unknown
525 558 603 608 653 663 681 689 701 705 764 769 771 802 866 874 904 981 989 994 1042 1068 1103 1108 1607
Schultz (2001) who reported that the Folin–Ciocalteu assay was a common method to determine the amount of polyphenolics by
1338
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
Fig. 4. Changes of compound abundances with fermentation level.
Fig. 5. PCA of tea volatile data showing grouping along the x-axis using 25 compounds in the analysis. Factor 1 represented 39.8% of the variance while Factor 2 represented 14.6%.
measuring their reducing capacity but it was not suitable for complex polyphenols due to selectivity of the Folin reagent for tannins. Therefore, reduction in total soluble phenolic value was due to the decline on flavonol glycosides and conversion from monomeric phenolic compounds to polymeric tannins during the fermentation process. Antioxidant capacity, determined using the ORAC assay, was also reduced by tea fermentation. Antioxidant capacity of each tea infusion decreased by 7%, 4%, 8%, and 20% by 20%, 40%, 60%,
and 80% tea fermentation processes, respectively (Table 2). Antioxidant capacity of tea infusions was correlated with changes of total soluble phenolics (r = 0.79) and EGCG, EGC, EC, ECG, total flavonol glycoside (r = 0.81, 0.81, 0.87, 0.82, and 0.79, respectively). Many studies reported that green tea showed more potential health benefits from significantly higher antioxidant capacity when measured using many different in vitro methods for antioxidant activity including VCEAC (Vitamin C Equivalent Antioxidant Capacity), TEAC (Trolox Equivalent Antioxidant Capacity), FRAP (Ferric
1339
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
Fig. 6. DFA of tea volatile data using 24 compounds in the model. Ellipses are the 95% confidence interval for each group.
Table 7 Discriminant function analysis p-values. Peak #
Name
Partial lambda
F-remove
p-level
17 16 12 5 1 25 24 14 2 22 23 15 13 3 20 7 6 10 4 9 21 11 19 8
Heptanal p-Xylene 2-Penten-1-ol, (Z)Butanal, 3-methylDimethyl sulphide Unknown Unknown Hexanal Propanal, 2-methylUnknown Linalool Ethylbenzene Toluene 2-Butanone Furan, 2-pentyl1-Penten-3-ol Butanal, 2-methylFuran, 2-ethylFuran, 2-methylPentanal Cyclohexanone, 2,2,6-trimethyl1-Pentanol 5-Hepten-2-one, 6-methyl1-Penten-3-one
0.0017 0.0022 0.0023 0.0027 0.0030 0.0031 0.0034 0.0037 0.0043 0.0064 0.0066 0.0073 0.0089 0.0094 0.0131 0.0131 0.0238 0.0300 0.0402 0.0419 0.0961 0.1204 0.2898 0.3045
145.48 114.64 108.31 90.99 81.87 80.28 72.95 66.87 57.54 38.81 37.59 34.05 27.81 26.46 18.86 18.84 10.25 8.07 5.97 5.72 2.35 1.83 0.61 0.57
0.062 0.070 0.072 0.078 0.083 0.083 0.088 0.091 0.099 0.120 0.122 0.128 0.141 0.145 0.171 0.171 0.230 0.257 0.297 0.303 0.450 0.500 0.730 0.744
Reducing Antioxidant Power), and TRAP (Total Radical Trapping Antioxidant Parameter) (Bravo, Goya, & Lecumberri, 2007; Lee, Lee, & Lee, 2002; Pellegrini et al., 2003; Rice-Evans 1999). Even though theaflavin, one of the most prevalent phenolic compounds in black tea shows similar antioxidant capacity with EGCG which is a major tea catechin in green tea (Lin, Chen, Ho, & Lin-Shiau, 2000; Yoshida et al., 1999), the concentration of tea catechins including EGCG was significantly higher than theaflavin and thearubigin (Bravo et al., 2007; Lee et al., 2002). In this study, reduction in antioxidant capacity was also observed during tea fermentation and it could be not only because of the difference in concentration of green and black tea polyphenolics in each tea but also because of the decline of other antioxidants including flavonol glycosides (r = 0.77), caffeine (r = 0.77), saponin (Cabrera et al., 2003; Önning, Juillerat, Fay, & Asp, 1994) and ascorbic acid (Aucamp, Hara, &
Apostolides, 2000; Lee & Kader, 2000) by oxidation or heat exposure during the fermentation process.
3.4. CIELAB colour value changes by tea fermentation L⁄ (darkness to lightness) value significantly decreased during oxidation and this indicates that tea infusion colour became darker due to the increase of theaflavin and thearubigin (r = 0.97 and 0.80) (Table 5). Other colour parameters such as a⁄ (red to green) and b⁄ (yellow to blue) were also changed during tea fermentation. Changes in colour values (a⁄ and b⁄) were more affected by theaflavin concentration than thearubigin as observed in L⁄ changes. The correlation was 0.96 and 0.92 between a⁄ and b⁄ and theaflavin while it was 0.89 and 0.68 between a⁄ and b⁄ and thearubigin.
1340
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
Fig. 7. DFA of tea volatile data using 16 compounds with complete separation. Ellipses are the 95% confidence interval for each group.
3.5. Volatile compounds changes observed by GC–MS analysis Twenty-four compounds (and the internal standard) were used in the volatile analysis of the fermented teas. The compounds used are presented in Table 5 with their retention indices. All the compounds reported in Table 6 have been previously reported in tea (Nijssen, Visscher, Maarse, Willemsens, & Boelens, 1999). The changes in the volatile profile at various fermentation levels results in some compounds increasing in concentration while others decreased. Of the 24 compounds analysed, only four were most abundant in the 0% fermented tea which decreased with increasing fermentation level. These four compounds were dimethyl sulphide, 1-penten-3-ol, toluene and ethylbenzene (Fig. 4a). Conversely, twelve compounds increased in amount as the fermentation level increased (Fig. 4b, 4c). A principle component analysis (PCA) was performed on the data with the results presented in Fig. 5. There is a general clustering of the data based on fermentation level along the x-axis (factor 1). This is an important first step in analysing data with multivariate techniques as it helps guard against over-fitting the data (Goodner, Dreher, & Rouseff, 2001). PCA is described as an unsupervised learning technique which means that the algorithms are performed on the data and show the inherent structure of the data without knowledge of data groupings. A useful method is to proceed to a supervised learning situation where the algorithms utilised the data classifications in analysing the data which yields better clustering than PCA as the variable weights are adjusted to maximise the distance between the groups (Goodner et al., 2001). One such method is a Discriminant Function Analysis (DFA) which using the same dataset, produced very tight clustering (Fig. 6). As one would suspect, the clustering of the data is more tightly clustered and well defined. This is a pretty clear indication that the model has room to be simplified by removing variables. A simple method to reduce the number of variables is to consider the summary of statistics for the DFA. The partial lambdas, F-remove statistic, and the associated p-level are listed in Table 7. The p-level is the probability level for each level, and using an arbitrary cut-off of 0.2, reduces the number of variables to 16 from 24. Re-analysing these variables using the DFA produces the
clustering seen in Fig. 7. Naturally, reducing the number of variables decreases the separation of the clusters, but the clusters are completely separate as seen by the 95% confidence ellipses drawn around each group. When evaluating a model such as this, the coefficients of the variables indicates the relative importance of the variable to the model. The larger the value (regardless of positive or negative) indicates a more heavily weighted variable, and therefore more important to the model. The coefficients for the variables of Fig. 7 are listed in Table 8. Pentanal, heptanal, 3,5,5-trimethyl-2cyclohexen-1-one, and 3-methylbutanal are the most heavily weighted peaks in the x-axis. It is important to note, that these are merely the best variables for differentiating the groups, not necessarily the most important compounds regarding flavour. It would be possible to utilise this type of modelling for determining fermentation level. A more basic method might be to utilise a single compound such as 3-methylbutanal which shows a clear increase with each level of fermentation.
Table 8 Standardised coefficients for variables used in Fig. 7. Peak #
Name
Root 1
Root 2
1 2 3 5 7 9 13 14 15 16 17 20 21 23 24 25
Dimethyl sulphide Propanal, 2-methyl2-Butanone Butanal, 3-methyl1-Penten-3-ol Pentanal Toluene Hexanal Ethylbenzene p-Xylene Heptanal Furan, 2-pentyl2-Cyclohexen-1-one, 3,5,5-trimethylLinalool Unknown Unknown Eigenvalue Cum.Prop
2.4449 4.0888 2.4586 5.5612 0.7273 12.9230 2.2642 1.5853 0.5122 0.3824 8.2043 1.2341 6.4138 2.5752 4.3809 0.9925 142.5440 0.5767
7.2837 13.9718 2.4913 17.1174 9.2811 30.0223 6.0709 2.1391 1.5392 1.1096 21.5079 2.7636 14.7983 3.0897 4.4972 2.8460 74.8504 0.8795
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
4. Conclusion As a result of tea fermentation, tea catechins were significantly reduced due to the transformation to theaflavins and thearubigins and it resulted in loss of total soluble phenolic content and antioxidant capacity of teas. Loss in antioxidant capacity was likely due to oxidative or thermal degradation of antioxidants such as caffeine, saponin, ascorbic acid, and non-catechin polyphenolics (flavonol glycosides) rather than the conversion of tea catechins to theaflavins during fermentation process. Unlike as reported in previous studies, it was found that caffeine naturally present in tea decreased during oxidative degradation. The reason why previous studies reported a higher concentration of caffeine in black tea was likely due to geographical difference, leaf age, leaf quality, and extractability rather than naturally higher caffeine content in black tea. In general, volatile compounds increased in concentration as a result of tea fermentation. It was shown that it is possible to develop either univariate or multivariate models to predict tea fermentation. The tea market has grown extensively due to an increased awareness of the potential health benefits from bioactive polyphenolics present in teas along with consumers migrating away from carbonated soft drinks. Thus, results obtained from the present study provide fundamental and practical information to the tea industry and consumers concerning tea polyphenolics, antioxidant capacity, caffeine, and volatile composition.
References Ahmad, N., & Mukhtar, H. (1999). Green tea polyphenolics and cancer: Biologic mechanisms and practical implications. Nutrition Reviews, 57, 78–83. Appel, H. M., Govenor, H. L., D’Ascenzo, M., Siska, E., & Schultz, J. C. (2001). Limitation of folin assays of folier phenolics in ecological studies. Journal of Chemical Ecology, 27, 761–778. Astill, C., Birch, M. R., Dacombe, C., Humphrey, P. G., & Martin, P. T. (2001). Factors affecting the caffeine and polyphenol contents of black and green tea infusions. Journal of Agricultural and Food Chemistry, 49, 5340–5347. Atoui, A. K., Mansouri, A., Boskou, G., & Kefalas, P. (2005). Tea and herbal infusions: Their antioxidant activity and phenolic profile. Food Chemistry, 89, 27–36. Aucamp, J. P., Hara, Y., & Apostolides, Z. (2000). Simultaneous analysis of tea catechins, caffeine, gallic acid, theanine and ascorbic acid by micellar electrokinetic capillary chromatography. Journal of Chromatography A, 876, 235–242. Bravo, L., Goya, L., & Lecumberri, E. (2007). LC/MS characterization of phenolic constituents of mate (Ilex paraguariensis, St. Hil.) and its antioxidant activity compared to commonly consumed beverages. Food Research International, 40, 393–405. Buchner, N., Krumbein, A., Rohn, S., & Kroh, L. W. (2006). Effect of thermal processing on the flavonols rutin and quercetin. Rapid Communications in Mass Spectrometry, 20, 3229–3235. Cabrera, C., Giménez, R., & López, M. C. (2003). Determination of tea components with antioxidant activity. Journal of Agricultural and Food Chemistry, 51, 4427–4435. Chan, P. T., Fong, W. P., Cheung, Y. L., Huang, Y., Ho, W. K. K., & Chen, Z. Y. (1999). Jasmine green tea epicatechins are hypolipidemic in hamsters fed a high-fat diet. Journal of Nutrition, 129, 1094–1101. Chen, Z., Zhu, Q. Y., Wong, Y. F., Zhang, Z., & Chung, H. Y. (1998). Stabilizing effect of ascorbic acid on green tea catechins. Journal of Agricultural and Food Chemistry, 46, 2512–2516. Chen, Z., Zhu, Q. Y., Tsang, D., & Huang, Y. (2001). Degradation of green tea catechins in tea drinks. Journal of Agricultural and Food Chemistry, 49, 477–482. Cheng, T. O. (2006). Review. All teas are not created equal: The Chinese green tea and cardiovascular health. International Journal of Cardiology, 108, 301–308. Del Rio, D., Stewart, A. J., Mullen, W., Burns, J., Lean, M. E. J., Brighenti, F., et al. (2004). HPLC-MSn analysis of phenolic compounds and purine alkaloids in green and black tea. Journal of Agricultural and Food Chemistry, 52, 2807–2815. Dou, J., Lee, V. S. Y., Tzen, A. T. C., & Lee, M. (2008). Rapid identification of aclyated flavonol tetraglycosides in oolong teas using HPLC-MSn. Phytochemical Analysis, 19, 251–257. Dubick, M. A., & Omaye, S. T. (2007). Handbook of nutraceuticals and functional foods (3rd ed.). Boca Raton, FL: CRC press. Taylor & Francis group. Engelhardt, U. H., Finger, A., & Kuhr, S. (1993). Determination of flavone C-glycoside in tea. Zeitschrift für Lebensmitteluntersuchung und -Forschung A, 197, 239–244. Fernández, P. L., Martín, M. J., González, A. G., & Pablos, F. (2000). HPLC determination of catechins and caffeine in tea: Differentiation of green, black and instant teas. Analyst, 421, 425.
1341
Ferreres, F., Silva, B. M., Andrade, P. B., Seabra, R. M., & Ferreira, M. A. (2003). Approach to the study of C-glycosyl flavones by ion trap HPLC-PAD-ESI/MS/MS: Application to seeds of quince (cydonia oblonga). Phytochemical Analysis, 14, 352–359. Goodner, K. L., Dreher, J. G., & Rouseff, R. L. (2001). The dangers of creating false classifications due to noise in electronic nose and similar multivariate analyses. Sensors and Actuators B, 80, 261–266. Gupta, S., Saha, B., & Giri, A. K. (2002). Comparative antimutagenic and anticlastogenic effects of green tea and black tea: a review. Mutational research-Reviews in Mutation Research, 512, 37–65. Gupta, S., Ahmad, N., Nieminen, A., & Mukhtar, H. (2000). Growth inhibition, cellcycle dysregulation, and induction of apoptosis by green tea constituent ()-epigallocatechin-3-gallate in androgen-sensitive and androgen-insensitive humanprostate carcinoma cells. Toxicology and Applied Pharmacology, 164, 82–90. Han, L. K., Takaku, T., Limura, Y., & Okuda, H. (1999). Anti-obesity action of oolong tea. International Journal of Obesity, 23, 98–105. Haslam, E. (2003). Thoughts on thearubigins. Phytochemistry, 64, 61–73. Janssen, K., Mensink, R. P., Cox, F. J. J., Harryvan, J. L., Hovenier, R., Hollman, P. C. H., et al. (1998). Effects of the flavonoids quercetin and apigenin on hemostasis in healthy volunteers: Results from an in vitro and a dietary supplement study. The American Journal of Clinical Nutrition, 67, 255–262. Khokhar, S., & Magnusdottir, S. G. M. (2002). Total phenol, catechin, and caffeine contents of teas commonly consumed in the United Kingdom. Journal of Agricultural and Food Chemistry, 50, 565–570. Kiehne, A., & Engelhardt, U. H. (1996). Thermospary-LC–MS analysis of various groups of polyphenolic in tea. 1. Catechins, flavonols O-glycosides and flavone C-glycosides. Zeitschrift Lebensmittel-Untersuchung und Forschung, 202, 48–54. Kim, Y. Factors influencing antioxidant phytochemical stability of teas. (2008). PhD, Dissertation, Texas A&M University. Kuhr, S., & Engelhardt, U. H. (1991). Determination of flavanols, theogallin, gallic acid and caffeine in teas using HPLC. Zeitschrift Lebensmittel-Untersuchung und Forschung, 192, 526–529. Lee, J. H., & Talcott, S. T. (2002). Ellagic acid and ellagitannins affect on sedimentation in muscadine juice and wine. Journal of Agricultural and Food Chemistry, 50, 3971–3976. Lee, K. W., Lee, H. J., & Lee, C. Y. (2002). Antioxidant activity of black tea vs. green tea. American Society for Nutritional Science. Journal of Nutrition, 132, 785. Lee, S. K., & Kader, A. A. (2000). Preharvest and postharvest factors influencing vitamin C content of horticultural crops. Postharvest Biology and Technology, 20, 207–220. Lin, J. K., Chen, P. C., Ho, C. T., & Lin-Shiau, S. Y. (2000). Inhibition of xanthine oxidase and suppression of intracellular reactive oxygen species in HL-60 cells by theaflavin-3,3-digallate, (–)-epigallocatechin-3-gallate, and propylgallate. Journal of Agricultural and Food Chemistry, 48, 2736–2743. Lin, Y., Tsai, Y., Tsay, J., & Lin, J. (2003). Factors affecting the levels of tea polyphenols and caffeine in tea leaves. Journal of Agricultural and Food Chemistry, 51, 1864–1873. Makris, D. P., & Rossiter, J. T. (2000). Heat-induced, metal-catalyzed oxidative degradation of quercetin and rutin (quercetin 3-Orhamnosylglucoside) in aqueous model systems. Journal of Agricultural and Food Chemistry, 48, 3830–3838. Markowicz Bastos, D. H., Saldanha, L. A., Catharino, R. R., Sawaya, A. C. H. F., Cunha, I. B. S., Carvalho, P. O., et al. (2007). Phenolic antioxidants identified by ESI-MS from yerba maté (Ilex paraguariensis) and green tea (Camelia sinensis) extracts. Molecules, 12, 423–432. Nijssen, L. M., Visscher, C. A., Maarse, H., Willemsens, L. C., & Boelens, M. H. (1999). Volatile compounds in foods and beverages. Zeist, The Netherlands: TNO Nutrition and Food Research Institute. Obanda, M., Owuor, P. O., & Mang’oka, R. (2001). Changes in the chemical and sensory quality parameters of black tea due to variations of fermentation time and temperature. Food Chemistry, 75, 395–404. Obanda, M., Owuor, P. O., Mang’oka, R., & Kavoi, M. M. (2004). Changes in thearubigin fractions and theaflavin levels due to variations in processing conditions and their influence on black tea liquor brightness and total colour. Food Chemistry, 85, 163–173. Önning, G., Juillerat, M. A., Fay, F., & Asp, N. (1994). Degradation of oat saponins during heat processing – Effects of pH, stainless steel, and iron at different temperatures. Journal of Agricultural and Food Chemistry, 42, 2578–2582. Pellegrini, N., Serafini, M., Colombi, B., Del Rio, D., Salvatore, S., Bianchi, M., et al. (2003). Total antioxidant capacity of plant foods, beverages and oils consumed in Italy assessed by three different in vitro assays. Journal of Nutrition, 133, 2812–2819. Rice-Evans, C. (1999). Implications of the mechanisms of action of tea polyphenols as antioxidants in vitro for chemoprevention in humans. Proceedings of the Society for Experimental Biology and Medicine, 220, 262–266. Rietveld, A., & Wiseman, S. (2003). Antioxidant effects of tea: Evidence from human clinical trials. Journal of Nutrition, 133, 3258S–3292S. Stewart, A. J., Mullen, W., & Crozier, A. (2005). On-line high-performance liquid chromatography analysis of the antioxidant activity of phenolic compounds in green and black tea. Molecular Nutrition and Food Research, 49, 52–60. Swain, T., & Hillis, W. E. (1959). The phenolic constituents of Purmus domestica. I. The quantitative analysis of phenolic constituents. Journal of the Science of Food and Agriculture, 10, 63–68.
1342
Y. Kim et al. / Food Chemistry 129 (2011) 1331–1342
Talcott, S. T., & Lee, J. (2002). Ellagic acid and flavonoid antioxidant content muscadine wine and juice. Journal of Agricultural and Food Chemistry, 50, 3186–3192. Wang, H., Provan, G. J., & Helliwell, K. (2000). Tea flavonoids: Their functions, utilization and analysis. Trends in Food Science and Technology, 11, 152–160. Wang, H., & Helliwell, K. (2001). Determination of flavonols in green and black tea leaves and green tea infusions by high performance liquid chromatography. Food Research International, 34, 223–227. Yen, G., & Chen, H. (1995). Antioxidant activity of various tea extracts in relation to their antimutagenicity. Journal of Agricultural and Food Chemistry, 43, 27–32.
Yoshida, H., Ishikawa, T., Hosoai, H., Suzukawa, M., Ayaori, M., Hisada, T., et al. (1999). Inhibitory effect of tea flavonoids on the ability of cells to oxidize low density lipoprotein. Biochemical Pharmacology, 58, 1695–1703. Yoshioka, H., Sugiura, K., Kawahara, R., Fujita, T., Makino, M., Kamiya, M., et al. (1991). Formation of radicals and chemiluminescence during the autoxidation of tea catechins. Agricultural and Biochemical Chemistry, 55, 2717–2723. Zuo, Y., Chen, H., & Deng, Y. (2002). Simultaneous determination of catechins, caffeine and gallic acids in green, oolong, black and pu-erh teas using HPLC with a photodiode array detector. Talanta, 57, 307–316.