- Email: [email protected]
MS
Accepted Manuscript Antioxidant profiling of vine tea (Ampelopsis grossedentata): off-line coupling heart-cutting HSCCC with HPLC-DAD-QTOF-MS/MS Qingping Gao, Ruyi Ma, Lin Chen, Shuyun Shi, Ping Cai, Shuihan Zhang, Haiyan Xiang PII: DOI: Reference:
S0308-8146(16)31971-9 http://dx.doi.org/10.1016/j.foodchem.2016.11.122 FOCH 20255
To appear in:
Food Chemistry
Received Date: Revised Date: Accepted Date:
1 July 2016 31 October 2016 22 November 2016
Please cite this article as: Gao, Q., Ma, R., Chen, L., Shi, S., Cai, P., Zhang, S., Xiang, H., Antioxidant profiling of vine tea (Ampelopsis grossedentata): off-line coupling heart-cutting HSCCC with HPLC-DAD-QTOF-MS/MS, Food Chemistry (2016), doi: http://dx.doi.org/10.1016/j.foodchem.2016.11.122
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Graphical Abstract
50 7-10
5-6
2 Absorbance (mAU)
40
A. grossedentata
30 20 4
Exract + DPPH 0
5
10 15 Time (min)
2 50
8 3
1
6 5
Absorbance (mAU)
40
R3 OR1
O
25
60
OH
OH
20
HSCCC
R2
O
15
Exract
0
HO
12-13 14 11
3
1 10
14 13 11
30 20
4 9
10
10 12
7
e d
15
c b a
0 0
5
10 15 Time (min)
1
20
25
1
Antioxidant profiling of vine tea (Ampelopsis grossedentata): off-line coupling
2
heart-cutting HSCCC with HPLC–DAD–QTOF-MS/MS
3 4
Qingping Gaoa,b, Ruyi Mab, Lin Chenc, Shuyun Shib, *, Ping Caic, Shuihan Zhangc,
5
Haiyan Xiangd,**
6 7
a
8
410078, P. R. China
9 10 11 12 13 14
b
Department of Orthodontics, Xiangya Hospital, Central South University, Changsha
College of Chemistry and Chemical Engineering, Central South University,
Changsha 410083, P. R. China c
Research Institute of Chinese Medicine, Hunan Academy of Chinese Medicine,
Changsha 410013, P. R. China d
School of Pharmaceutical Sciences, Southern Medical University, Guangzhou
510515, P.R. China
15 16 17
* Corresponding Author. Tel.: +86 731 88879616.
18
** Corresponding Author. Tel.: +86 20 62789419
19
E-mail addresses: [email protected] (S. Shi), [email protected] (H. Xiang)
20 21
Running title: Antioxidant flavonoids in vine tea
2
22
ABSTRACT
23
Vine tea with strong antioxidant activity is commonly consumed as healthy
24
tea/beverage. However, detailed information about its antioxidants is incomplete.
25
Here, off-line hyphenation of heart-cutting high-speed countercurrent chromatography
26
(HSCCC)
27
detector‒quadrupole
28
(HPLC–DAD‒QTOF-MS/MS) were described for systematic profiling antioxidants in
29
vine
30
1,1-diphenyl-2-picryl-hydrazyl radical‒high performance liquid chromatography
31
(DPPH‒HPLC). Subsequently, stepwise HSCCC using petroleum ether–ethyl
32
acetate–methanol–water (4:9:4:9, v/v/v/v) and (4:9:5:8, v/v/v/v) as solvent systems
33
was optimized to fractionate and enrich antioxidants from ethyl acetate fraction of
34
vine tea. Finally, heart-cutting mode was used to collect five interesting HSCCC
35
fractions for HPLC–DAD‒QTOF-MS/MS analysis. Desirable orthogonality between
36
HSCCC and HPLC led to identification of fifteen antioxidant flavonoids, while four
37
minor flavonoids were first reported in vine tea. Results showed that the developed
38
system is efficient to comprehensively explore antioxidants from complex natural
39
herbs.
tea.
with
At
high
performance
liquid
time-of-flight
first,
chromatography‒diode
tandem
antioxidants
were
mass
rapidly
array
spectrometry
screened
by
40
Vine
tea;
41
Keywords:
42
HPLC–DAD–QTOF-MS/MS
Flavonoid;
3
Antioxidant
activity;
HSCCC;
43
1. Introduction
44
Ampelopsis grossedentata, a plant belonging to Vitaceae family, is distributed
45
widely in mountainous areas of southern China. The tender leaves and stems of A.
46
grossedentata, also called vine tea, have been commonly consumed as healthy tea,
47
beverage and herbal medicine for hundreds of years. Pharmaceutical investigations
48
show that vine tea exhibits significant bioactivities of antioxidant (Hou et al., 2014;
49
Ye, Wang, Duncan, Eigel, & O'Keefe, 2015), anti-inflammatory (Chen et al., 2015;
50
Hou et al., 2015), antiviral (Yan, & Zheng, 2009), antitumor (Zhou et al., 2014),
51
anti-diabetic and anti-hyperglycemic properties (Chen, Wu, Zou, & Gao, 2016).
52
Antioxidant activity has sparked interest due to its relevant to some metabolic
53
diseases (Hu et al., 2016; Zhou et al., 2015). Flavonoids, a kind of phenolic
54
antioxidants existed in many foods and herbs (Karabin, Hudcova, Jelinek, & Dostalek,
55
2015; Mattila, Hellström, Karhu, Pihlava, & Veteläinen, 2016), are found to be the
56
major metabolites in vine tea (Du, Cai, Xia, & Ito; 2002; Du, Chen, Jerz, &
57
Winterhalter, 2004; Gao et al., 2009; Wang, Zheng, Xu, & Zheng, 2002).
58
Dihydromyricetin, a dihydroflavonol with higher 1,1-diphenyl-2-picryl-hydrazyl
59
radical (DPPH) scavenging activity than butylated hydroxytoluene, accounts for
60
around 20% (w/w) on dry weight of vine tea (Du et al., 2004). Thus, most reports
61
focused on the bioactivity assays of dihydromyricetin or crude extract of vine tea.
62
Nevertheless, the therapeutic efficacies of natural products are achieved by
63
combinatorial compounds rather than single or two major compounds. Moreover,
64
some minor compounds (less than 0.1% (w/w) on dry weight of herbs) are found to
65
present significant biological activities and have been developed as clinical drugs or
66
lead compounds for drug discovery (e.g. paclitaxel, vincristine) (Oberlies, & Kroll,
67
2004). Then it is important to comprehensively investigate antioxidant flavonoids in 4
68
vine tea.
69
Compounds in natural products exist with different abundances and structures.
70
DPPH, a paramagnetic compound with an odd electron, would capture one or more
71
hydrogen atoms of antioxidants after spiking with them. Then higher performance
72
liquid chromatography (HPLC) peak areas of antioxidants would reduce or disappear.
73
Therefore, DPPH–HPLC technology has been developed to rapidly and effectively
74
screen antioxidants from complex natural products by comparing their HPLC peak
75
areas before and after spiking with DPPH (Hu et al., 2015; Qiu et al., 2012; Zhao et
76
al., 2015). Notably, no sample pretreatment was contained. HPLC‒diode array
77
detector‒quadrupole
78
(HPLC–DAD‒QTOF-MS/MS) provided ultraviolet (UV) spectra, high-resolution MS
79
and MS/MS spectra, which were useful for structural identification even when
80
standard
81
Segura-Carretero, & Fernández-Gutiérrez, 2016; Sarah et al., 2016; Pihlava, &
82
Kurtelius, 2016; Zhang et al., 2016). High-abundant compounds in vine tea (e.g.
83
dihydromyricetin, myricetin, and myricitrin) have been analyzed and characterized by
84
HPLC system in detail (Du et al., 2004; Gao et al., 2009). From Fig. 1a, four major
85
peaks (I‒IV) existed, in addition, some minor compounds were observed when HPLC
86
chromatogram was zoomed for 13 folds (Fig. 1b). Then, it was necessary to develop
87
some methods to identify minor compounds, especially for those overlapping with
88
major compounds. Recently, two-dimensional (2D) HPLC was developed.
89
Orthogonal columns used in 2D HPLC would provide high peak capacity and
90
resolution for comprehensive compounds analysis (Li et al., 2006; Yang et al., 2016).
91
However, for compounds at very low abundance, some preparative methods were still
92
irreplaceable to enrich them.
compounds
time-of-flight
were
not
tandem
available
5
mass
(Abu-Reidah,
spectrometry
Arráez-Román,
93
High-speed countercurrent chromatography (HSCCC), a unique liquid–liquid
94
partition chromatography method based on partitioning of compounds between two
95
immiscible liquid phases with a support-free matrix, no irreversible adsorption, low
96
risk of sample denaturation and total sample recovery, is an optimal choice for purify
97
major compounds and enrich minor compounds from complex matrix (Esatbeyoglu,
98
Wray, & Winterhalter, 2015; Zhang, et al., 2015; Zhao et al., 2015). Major compounds
99
were knocked out from M. doumeri by HSCCC, and then thirty minor antioxidants
100
were enriched and identified by HPLC–DAD–QTOF-MS/MS (Zhao et al., 2015). To
101
make full use of the orthogonality between HSCCC and HPLC, off-line
102
comprehensive HSCCC × HPLC‒MS/MS was developed, and eighty-five compounds
103
were detected in Citrus limetta (Rodríguez-Rivera, Lugo-Cervantes, Winterhalter, &
104
Jerz, 2014). However, heart-cutting mode is effective by only collection of interesting
105
HSCCC effluents for HPLC analysis.
106
Thus, the aim of this research is to investigate the antioxidant activity and
107
antioxidant compounds in vine tea. In order to achieve antioxidant information, total
108
flavonoid contents and DPPH scavenging activities of three fractions with different
109
polarities were firstly evaluated, and DPPH‒HPLC was then used to rapidly screen
110
antioxidants from the fraction with strongest activity. Co-eluted and minor
111
antioxidants could be successfully separated, enriched and then identified by making
112
full use of the orthogonality between HSCCC and HPLC. The resultant approach
113
found fifteen antioxidants in ethyl acetate fraction of vine tea, and four minor ones
114
were reported in vine tea for the first time.
115 116
2. Experimental
117
2.1. Chemicals and reagents 6
118
All organic solvents used for extraction and separation were of analytical grade
119
(Chemical Reagent Factory of Hunan Normal University, Hunan, China). Methanol
120
and formic acid (Sinopharm Chemical Reagent Co., Ltd, Shanghai, China) were used
121
for HPLC analysis. Deionized water (18.2 MΩ) was obtained from a Milli-Q water
122
purification system (Millipore, Bedford, MA, USA). Ascorbic acid (VC) and DPPH
123
(95%) were bought from Sigma-Aldrich (Shanghai Division). Nine flavonoid
124
standards, dihydromyricetin, myricetin, dihydroquercetin, quercetin, hesperitin,
125
kaempferol,
126
kaempferol-3-O-α-L-rhamnoside, were purchased from the National Institute for the
127
Control of Pharmaceutical and Biological Products (Beijing, China). Phloridzin and
128
phloretin with purities over 98 % were purified from Malus doumeri by two-step
129
HSCCC in our laboratory. Firstly, phloridzin was purified by HSCCC using ethyl
130
acetate–n-butanol–methanol–25 mM ammonium acetate solution (3.5:1.5:1:4, v/v/v/v)
131
as solvent system. Then, extruded sample from the first HSCCC separation was
132
injected into the second HSCCC to purify phloretin using petroleum ether–ethyl
133
acetate–methanol–water (1:2:1.5:1.5, v/v/v/v) as solvent system. Their structures were
134
identified by UV, MS and nuclear magnetic resonance (NMR) (Zhao et al., 2015).
135
2.2. Preparation of vine tea extract
myricitrin,
quercetin-3-O-α-L-rhamnoside,
and
136
Vine tea was collected in May 2015 from Jianghua (Hunan province of China,
137
Northern altitude 25°15', longitude 112°46', altitude: 610 m). The plant material was
138
identified using a species identification key (Editorial committee of flora of China,
139
1998) by Prof. Zhaoming Xie, Research Institute of Chinese Medicine, Hunan
140
Academy of Chinese Medicine, Changsha, China. After collection, vine tea was
141
immediately dried at 40 °C in an oven with air circulation. Dry vine tea was ground
142
and sieved, and materials between 180 and 250 µm was used for extraction. Vine tea 7
143
powder (25.0 g) was extracted with 70% ethanol (250 mL, three times) at 85°C (each
144
for 3 h). After filtration, organic solvent was removed from the combined extracts
145
under reduced pressure to yield a total crude extract (11.4 g). Crude extract,
146
suspended in water (100 mL), was then extracted successively with 3 × 100 mL of
147
petroleum ether, ethyl acetate and n-butanol. After extraction, solvents were removed
148
under reduced pressure, and then petroleum ether fraction (0.2 g), ethyl acetate
149
fraction (1.9 g) and n-butanol fraction (1.4 g) were stored at 4ºC for further
150
experiments (Liang et al., 2015).
151
2.3. Determination of total flavonoid contents
152
Total flavonoid contents were estimated by NaNO2‒Al(NO3)3 colorimetric method
153
(Zheng, Xia, & Lu, 2015). Vine tea extract (1.0 mg/mL, 0.4 mL) was mixed with
154
NaNO2 solution (4 %, 0.4 mL). After standing for 6 min, Al(NO3)3 solution (9 %, 0.4
155
mL) were added and incubated for another 6 min at room temperature. Then NaOH
156
solution (1.6 mol/L, 2.0 mL) was added and kept for 12 min. Finally, absorbance was
157
measured at 510 nm against the control. Total flavonoids contents were calculated
158
using a standard calibration of quercetin (4.0‒14.0 µg/mL) and expressed as mg of
159
quercetin equivalent (QE) per g of dry weight (DW) sample (mg QE/g DW), while
160
data are presented as means ± standard deviation for three replicates.
161
2.4. DPPH radical scavenging assay
162
DPPH scavenging activity, with VC as standard, was assayed as previously
163
reported procedures (Zhao et al., 2015). Briefly, samples or standards with different
164
concentrations (25 µL, 2, 5, 10, 30, 50, 100, and 200 µg/mL) were mixed with DPPH
165
solution (40 µL, 0.4 mg/mL) and then diluted with methanol to 250 µL. After
166
incubating at 37 °C for 30 min, the absorbance was measured at 517 nm. DPPH
167
solution (0.064 mg/mL) without any sample was a control. The antioxidant activity is 8
168
expressed as percentage of DPPH radical elimination: [(Ablank–Asample)/Ablank] × 100 %
169
(Ablank and Asample are the absorbance of the control and sample after addition of
170
DPPH). Sample concentration providing 50% inhibition (IC50) was calculated from
171
the graph plotting inhibition percentage.
172
2.5. DPPH‒HPLC experiment
173
HPLC analysis was carried out using an Agilent 1200 series HPLC system (Agilent
174
Technologies, Santa Clara, CA), equipped with an online vacuum degasser, a
175
QuatPump, a manual injection valve with a 20 µL sample loop, a thermostated
176
column compartment, a diode assay detector (DAD), and an Agilent ChemStation.
177
Chromatographic separation was performed on a Waters Symmetry C18 column (150
178
mm × 3.9 mm i.d., 5 µm, Waters, MA, USA) in tandem with a Phenomenex C18 guard
179
cartridge (4.0 mm × 3.0 mm, Phenomenex, Torrance, CA). Water containing 0.1%
180
formic acid (A) and methanol containing 0.1% formic acid (B) were selected as the
181
mobile phase in gradient elution: 0–5 min, 25% B; 5–20 min, 25–60% B; 20–25 min,
182
60–75% B. The flow rate and the column temperature were set at 0.8 mL/min and
183
25°C, respectively. The chromatogram was acquired at 254 nm.
184
DPPH‒HPLC experiment was carried out according to our previous reports (Hu et
185
al., 2015; Peng, Zhang, & Shi, 2016). Typically, ethyl acetate fraction of vine tea (2.0
186
mg/mL, 300 µL) and DPPH (30.0 mg/mL, 300 µL) were mixed and incubated at
187
37 °C for 30 min. After passing through a 0.45 µm polytetrafluoroethylene syringe
188
filter, reaction mixtures were analyzed by HPLC in comparison with the control,
189
adding methanol (300 µL) into ethyl acetate fraction of vine tea (2.0 mg/mL, 300 µL).
190
Subsequently, by comparison with the chromatogram of control, peaks in reaction
191
mixtures disappeared or decreased could be considered as potential antioxidants.
192
2.6. HSCCC separation 9
193
Preparative HSCCC was performed using a model TBE-300B HSCCC (Shanghai
194
Tauto Biotechnique Co. Ltd., Shanghai, China), which was equipped with three
195
multilayered coils connected in series (diameter of tube, 1.6 mm, total capacity 260
196
mL) and a 20 mL manual sample loop. The rotation speed was adjustable, ranging
197
from 0 to 1000 rpm. In addition, HSCCC system was equipped with a TBP-1002
198
pump, a TBD-2000 UV detector, a HX-1050 constant temperature regulator (Beijing
199
Boyikang Lab Implement Co. Ltd., Beijing, China), and a WH V4.0 workstation
200
(Shanghai Wuhao Information Technology Co. Ltd., Shanghai, China).
201
Suitable solvent systems for HSCCC were selected according to K values of target
202
compounds as our previously described (Zhang, et al., 2015; Zhao et al., 2015). A
203
series of solvent systems composed of petroleum ether–ethyl acetate–methanol–water
204
were prepared, and ethyl acetate fraction of vine tea (0.25 mg) was added into solvent
205
system (1 mL). After thorough equilibration for 20 min, 20 µL of upper and lower
206
phases of solvent system were analyzed by HPLC, respectively, to obtain K values of
207
each compound in ethyl acetate fraction of vine tea. The K is defined as K =
208
Aupper/Alower, where Aupper and Alower were the HPLC peak area of target compound in
209
the upper and lower phases, respectively.
210
Two solvent systems composed of petroleum ether–ethyl acetate–methanol–water
211
(4:9:4:9, v/v/v/v) and (4:9:5:8, v/v/v/v) were selected for HSCCC separation, which
212
were prepared in a separation funnel according to volume ratios, and thoroughly
213
equilibrated by shaking repeatedly. The upper phase (stationary phase) and lower
214
phase (mobile phase) were then separated and degassed by ultrasonication for 30 min
215
shortly prior to use. Ethyl acetate fraction of vine tea (6.0 mg/mL) in 20 mL of lower
216
phase of petroleum ether–ethyl acetate–methanol–water (4:9:4:9, v/v/v/v) was
217
prepared as sample solution. HSCCC column was first entirely filled with the upper 10
218
phase of petroleum ether–ethyl acetate–methanol–water (4:9:4:9, v/v/v/v) as the
219
stationary phase; subsequently, the lower mobile phase was then pumped into the inlet
220
of the column at the flow rate of 1.5 mL/min, while the apparatus was run at 900 rpm;
221
after a clear mobile phase eluted at the tail outlet and the hydrodynamic equilibrium
222
was reached, 20 mL of sample solution (6.0 mg/mL) was then injected into the
223
injection valve. HSCCC separation was performed by a stepwise mode with lower
224
phase of petroleum ether–ethyl acetate–methanol–water (4:9:4:9, v/v/v/v) as the
225
mobile phase in the period of 0–170 min and then lower phase of petroleum
226
ether–ethyl acetate–methanol–water (4:9:5:8, v/v/v/v) in the period of 170–350 min.
227
The HSCCC chromatogram was acquired at 254 nm. After HSCCC separation was
228
completed, the solvent in HSCCC column was pumped out by pressured nitrogen.
229
2.7. HPLC–DAD–QTOF-MS/MS analysis of collected HSCCC fractions
230
HSCCC fractionation was performed as a function of time, with a collection time of
231
6.0 min per tube. The collected fractions were dried, and then redissolved in methanol
232
for HPLC–DAD–QTOF-MS/MS analysis.
233
HPLC separation column, mobile phase, gradient program and flow rate were the
234
same
as
those
presented
in
section
235
Accurate-Mass QTOF LC/MS system with an Agilent Jet Stream electrospray (ESI)
236
interface was coupled in parallel by splitting the mobile phase 1:3 using an adjustable
237
high-pressure stream splitter (Valco Instrument Company, Houston, TX, USA). The
238
MS settings for flavonoid characterization were optimized according to previously
239
reported (Zhao, Zhang, Guo, & Shi, 2015): scan range 100–1000 m/z, capillary
240
voltage 3.5 kV, dry gas temperature 320 ºC, dry gas flow rate 10.0 L/min, sheath gas
241
temperature 400 ºC, sheath gas flow rate 12 L/min, nebulizing gas pressure 35 psi,
242
fragmentor voltage 160 V. For collision-induced dissociation (CID) experiments, 11
2.5.
In
addition,
Agilent
6530
243
keeping MS1 static, the precursor ion of interest was selected using the quadrupole
244
analyzer and the product ions were analyzed using a time-of-flight (TOF) analyzer.
245
The mass axis was calibrated using mixtures provided by manufacturer.
246
2.8. Statistical analysis
247
Total flavonoid content and DPPH radical scavenging assay were conducted in
248
triplicate (n =3), and the results were expressed as mean ± SD (standard deviation),
249
while p < 0.05 represented a statistically significant difference. Statistical evaluation
250
was performed using Statistical Analysis System (version 9.2, SAS Institute Inc., Cary,
251
NC).
252 253
3. Results and discussion
254
3.1. Determination of total flavonoid contents and antioxidant activity
255
Flavonoids were considered as the widely existed antioxidants in natural products,
256
and our previous report indicated that total flavonoid contents had high positive
257
correlation to the antioxidant activities (Zheng et al., 2015). Accelerated solvent
258
extraction was always used to partition complex matrix. Here, 70% ethanol extract of
259
vine tea was fractionated successively by petroleum ether, ethyl acetate and n-BuOH.
260
The results indicated that ethyl acetate fraction yielded the highest total flavonoid
261
content (12.19 ± 2.03 mg QE/g DW), followed by n-butanol fraction (7.74 ± 1.16 mg
262
QE/g DW). Petroleum ether fraction had the lowest value (0.26 ± 0.09 mg QE/g DW).
263
Therefore, extraction solvent significantly (p < 0.05) affected the total flavonoid
264
contents. IC50 values of ethyl acetate fraction and n-butanol fraction to DPPH
265
scavenging were 3.05 ± 0.50 and 4.90 ± 0.52 µg/mL, respectively. However,
266
petroleum ether fraction did not show any antioxidant activities (IC50 > 20 µg/mL).
267
The DPPH scavenging activities exhibited the same trends with total flavonoid 12
268
contents. It is noted that ethyl acetate fraction and n-butanol fraction exhibited strong
269
antioxidant capacities in comparison to positive control, VC (IC50, 19.17 ± 3.21
270
µg/mL) (p < 0.05). Therefore, vine tea was rich in natural antioxidants for scavenging
271
biologically relevant radicals.
272
3.2. DPPH–HPLC analysis
273
The mobile phase system was firstly optimized to obtain reliable chromatographic
274
results. The final results showed that best resolution, shortest analysis time and lowest
275
pressure variations were achieved when a gradient elution mode composed of solvent
276
A (0.1% formic acid in water) and B (0.1% formic acid in methanol) was programmed
277
as follows: 0–5 min, 25% B; 5–20 min, 25–60% B; 20–25 min, 60–75% B. The flow
278
rate was 0.8 mL/min while the column temperature was 25°C. UV spectra were
279
recorded from 190 to 400 nm, while the chromatogram was acquired at 254 nm.
280
Flavonoid profiling of ethyl acetate fraction and n-butanol fraction of vine tea by
281
HPLC were almost the same, and the difference between them was their quantities
282
(Fig. 1), then subsequently, ethyl acetate fraction of vine tea was investigated. It was
283
obvious that there existed four major peaks (I‒IV) (Fig. 1b). Vine tea was reported to
284
contain two types of flavonoids, dihydroflavonols and flavonols (Du et al., 2004; Gao
285
et al., 2009; Wang et al., 2002). Dihydroflavonols had maximum UV absorption band
286
near 285 nm, while flavonols showed two maximum UV absorptions at 250‒270 and
287
340‒380 nm. A comprehensive UV screening showed that peaks II and III contained
288
two or more flavonoids, and UV spectra of peak IV (λmax: 249, 298, 333 nm) was not
289
a kind of flavonoid. In addition, there existed some minor flavonoids (Fig. 1c).
290
However, because of the co-elution and matrix interference, UV and MS data of
291
minor
292
HPLC–QTOF-MS/MS analysis. Therefore, fractionation and enrichment of minor
flavonoids
were
difficult
to
13
be
accurately
achieved
only
by
293
flavonoids were inevitable.
294
DPPH–HPLC was always used as a rapid method to screen antioxidants from
295
complex mixtures without sample pretreatment. It was believed structures of
296
antioxidants will be changed after they react with DPPH. Thus, HPLC peak areas of
297
antioxidants will decrease or disappear. Fig. 2 showed the chromatogram of ethyl
298
acetate fraction of vine tea before and after reaction with DPPH, which presented that
299
peak areas of fifteen flavonoids (1–15) disappeared or decreased obviously after
300
spiking with DPPH. Therefore, flavonoids 1–15 possessed potent antioxidant activity.
301
3.3. HSCCC separation
302
Most solvent systems were limited to separating compounds with a narrow range of
303
polarities, and then it was needed to apply a multistep elution using more than one
304
solvent system. Petroleum ether–ethyl acetate–methanol–water solvent system,
305
providing a broad range of polarities by modifying volume ratios of four solvents, has
306
been testified as be useful in purification of components in ethyl acetate fraction
307
(Zhao et al., 2015). The 0.5≤K≤2.0 in "test tube" experiments is always considered as
308
suitable K values for HSCCC. According to systematic analysis of K values of target
309
antioxidant flavonoids, stepwise HSCCC separation using petroleum ether–ethyl
310
acetate–methanol–water (4:9:4:9, v/v/v/v) and (4:9:5:8, v/v/v/v) as solvent systems
311
were selected (Fig. 3A). Obviously, HSCCC fractionated ethyl acetate fraction of vine
312
tea to four peaks, but each HSCCC peak contained a major compound and some
313
minor compounds. Collected HSCCC fractions were analyzed by off-line HPLC, and
314
five fractions (a‒e) contained target flavonoids (Fig. 3B). It is noted that HSCCC and
315
HPLC have orthogonality, which make it possible to identify co-eluted flavonoids.
316
3.4. Characterization of antioxidant flavonoids by HPLC‒DAD‒QTOF-MS/MS
317
A MS detector with positive ion mode in combination with DAD was applied to 14
318
interpret antioxidants by comparison with standards or published data. Table 1
319
showed the UV, MS and MS/MS spectral data and characterization results for these
320
compounds. Compounds 1‒3, 5, 9, 11, and 14 had typical UV spectra of
321
dihydroflavonols/flavanones/dihydrochalcones, while UV spectra of compounds 4,
322
6‒7, 8, 10, 12‒13, 15 were for flavonols.
323
Dihydroflavonols cannot be differentiated from flavanones/dihydrochalcones only
324
by UV spectra. However, dihydroflavonols are always more polar and be firstly eluted
325
in HPLC analysis. The early retention times of compounds 1‒3 and 5 suggested that
326
they might be dihydroflavonols, while compounds 9, 11 and 14 with lower polarities
327
might be flavanones/dihydrochalcones (Fig. 2).
328
Compound 2, the major compound in vine tea, was identified as dihydromyricetin
329
by comparison with standard, which was then taken as an example to explain the
330
fragmentation details of dihydroflavonols (Fig. 4A). The cleavages of C-ring at
331
positions 1/3, 1/2 and 0/2 were their characteristic fragmentations. Therefore, the
332
structurally informative ions, m/z 153.0182 (1,3A0+), 149.0237 (0,2A0+‒H2O), and
333
139.0394 (1,2B0+), were diagnostic for A- and B-ring substitutions. In addition,
334
dihydromyricetin displayed a fragment ion at m/z 303.0504 (C15H11O7) for the loss of
335
H2O of C-ring.
336
Both compounds 1 and 3 afforded [M+H]+ ion at m/z 321.06 (C15H13O8), the same
337
with that of 2. It was hard to distinguish compounds 2 and 3 by their UV and MS/MS
338
spectra, but they had different HPLC retention behaviors. As a result, the structural
339
difference between compounds 2 and 3 should be the configurations of C‒2 and C‒3.
340
Then, by comparison with previously published data (Zheng et al., 2014), compound
341
3 was tentatively identified as iso-dihydromyricetin. Compound 1 presented MS/MS
342
base peak at m/z 321.0605 ([M‒H2O+H]+) and typical fragment ions at m/z 167.0341 15
343
(0,2A0+) and 153.0183 (1,3A0+), indicating the same A-ring structure with that of 2.
344
Thinks to shorter retention time, compound 1 was allowed to be elucidated as
345
3-dihydroxyquercetin, which was previously reported in fermented vine tea (Hu et al.,
346
2016). The mass spectra of compound 5 displayed [M+H]+ ion at m/z 305.0656
347
(C15H13O7, 16 Da less than that of 2), and the 16 Da mass difference was also found
348
for
349
between compounds 5 and 2 was the numbers of hydroxyl groups on B-ring. Based on
350
the observations, compound 5 was tentatively established as dihydroquercetin, which
351
was finally confirmed using standard.
1,2
B0+ fragment ion between compounds 5 and 2. Thus, the structural difference
352
Compounds 9 and 11 exhibited [M+H]+ ions at m/z 437.1443 (C21H25O10) and
353
275.0914 (C15H15O5), respectively. The fragment ion with a high intensity of 9 at m/z
354
275.0912 ([M‒162+H]+, the same with the parent ion of 11, indicated the existence of
355
an additional glucosyl unit in 9. In addition, the m/z 169.0550 (C8H9O4) might be the
356
characteristic fragment ion of dihydrochalcones by the cleavage between C‒α and
357
C‒β. Thus, compounds 9 and 11 were assigned as phloridzin and phloretin (Fig. 5),
358
respectively, by comparing with their standards, while the existence of them in vine
359
tea had not been previously reported. Compound 14 showed [M+H]+ ion at m/z
360
303.0862 (C16H15O6) along with three fragment ions at m/z 288.0631 ([M‒CH3+H]+),
361
275.0914 ([M‒CO+H]+), and 153.0182 (1,3A0+), which was consistent with the
362
previously isolated compound, hesperitin.
363
The RDA rearrangement 1,3A0+ ion for the cleavage of C-ring was the characteristic 1,2
B0+ fragment ion was typical for
364
fragmentation for flavonols and flavones, however,
365
flavonols (Zhao et al., 2015). As shown in Fig. 4B, myricetin (compound 8),
366
unequivocally identified by comparing their HPLC‒DAD‒MS/MS data with those of
367
standard, displayed parent ion [M+H]+ at m/z 319.0450 (C15H11O8) and typical 16
368
fragment ions at m/z 153.0185 (1,3A0+) and 139.0393 (1,2B0+). Compounds 13 and 15
369
exhibited [M+H]+ ions at m/z 303.0500 (C15H11O7, 16 Da less than that of myricetin)
370
and 287.0549 (C15H11O6, 32 Da less than that of myricetin), which gave the same
371
1,3
A0+ fragment ion but different
1,2
B0+ fragment ion. Then it was evident that
372
compound 13 have two hydroxyl groups on B-ring, and 15 contained only one
373
hydroxyl group on B-ring. By comparing with standards, compounds 13 and 15 were
374
finally elucidated as quercetin and kaempferol. Compounds 4 and 6 afforded [M+H]+
375
ions at m/z 481.0960 (C21H21O13) and 465.1005 (C21H21O12), which gave the same
376
base peak at m/z 319.0448 (C15H11O8), mostly probably by loss of a glucosyl residue
377
(162 Da) and a rhamnosyl group (146 Da). Accordingly, compound 4 was tentatively
378
established as myricetin-3-O-β-D-glucoside, while compound 6 was deduced as
379
myricitrin (myricetin-3-O-α-L-rhamnoside) by comparing with standard. Using
380
similar
381
quercetin-3-O-β-D-xyloside,
382
kaempferol-3-O-α-L-rhamnoside by comparing them with standards or previously
383
reported data (Riethmüller et al., 2015). It was noted that compound 7 and 12 has not
384
been previously reported in vine tea (Fig. 5).
principles,
compounds
7,
10
and
12
were
quercetin-3-O-α-L-rhamnoside,
assigned
as and
385 386
4. Conclusions
387
In this study, total flavonoid contents in vine tea extracts with different polarities
388
and their antioxidant activities were studied. Ethyl acetate and n-butanol fractions
389
contained more flovonoids contents and exhibited stronger antioxidant capacities than
390
petroleum ether fraction and VC. Same HPLC flavonoid profiles with different
391
quantities were observed between ethyl acetate and n-butanol fractions. DPPH‒HPLC
392
analysis was then used to screen fifteen antioxidants in ethyl acetate fraction. The 17
393
co-eluted or minor antioxidants were separated or enriched by stepwise HSCCC, and
394
five interesting HSCCC fractions were heart-cut and further analyzed offline by
395
HPLC–DAD–QTOF-MS/MS. Structure of each antioxidant was finally elucidated by
396
comparison of retention time, DAD and MS/MS data, easily achieved because of the
397
orthogonality between HSCCC and HPLC, with standards or published data. It is
398
worth mentioning that four minor antioxidants were reported in vine tea for the first
399
time. The comprehensive knowledge of antioxidants in vine tea could make benefit
400
for its further application in food supplementation, cosmetics and medicinal products.
401
Moreover, the developed method provided an efficient way to investigate bioactive
402
compounds, especially for co-eluted and minor ones, in complex system.
403 404
Acknowledgments
405
This work was supported by the National Natural Science Foundation of China
406
(21275163, 81402935), the Science and Technology Program of Hunan Province,
407
China (2015NK3037), Program of Survey and Monitoring of Chinese Medicines for
408
National Drugs ([2011] 76).
18
409
References
410
Abu-Reidah, I., Arráez-Román, D., Segura-Carretero, A., & Fernández-Gutiérrez, A.
411
(2016). Extensive characterisation of bioactive phenolic constituents from globe
412
artichoke (Cynara scolymus L.) by HPLC–DAD-ESI-QTOF-MS. Food Chemistry,
413
141, 2269‒2277.
414
Chen, J., Wu, Y. C., Zou, J. W., & Gao, K. (2016). α-Glucosidase inhibition and
415
antihyperglycemic activity of flavonoids from Ampelopsis grossedentata and the
416
flavonoid derivatives. Bioorganic & Medicinal Chemistry, 24, 1488‒1494.
417
Chen, S. H., Zhao, X. L., Wan, J., Ran, L., Qin, Y., Wang, X. F., Gao, Y. X., Shu, F. R.,
418
Zhang, Y., Liu, P., Zhang, Q. Y., Zhu, J. D., & Mi, M. T. (2015). Dihydromyricetin
419
improves glucose and lipid metabolism and exerts anti-inflammatory effects in
420
nonalcoholic fatty liver disease: A randomized controlled trial. Pharmacological
421
Research, 99, 74‒81.
422
Du, Q. Z., Cai, W. J., Xia, M., & Ito, Y. (2002). Purification of (+)-dihydromyricetin
423
from leaves extract of Ampelopsis grossedentata using high-speed countercurrent
424
chromatography with scale-up triple columns. Journal of Chromatography A, 973,
425
217‒220.
426
Du, Q. Z., Chen, P., Jerz, G., & Winterhalter, P. (2004). Preparative separation of
427
flavonoid glycosides in leaves extract of Ampelopsis grossedentata using
428
high-speed counter-current chromatography. Journal of Chromatography A, 1040,
429
147‒149.
430 431
Editorial committee of flora of China, (1998). Flora of China. Beijing: Science Press, vol. 48, p. 53.
432
Esatbeyoglu, T., Wray, V., & Winterhalter, P. (2015). Isolation of dimeric, trimeric,
433
tetrameric and pentameric procyanidins from unroasted cocoa beans (Theobroma 19
434
cacao L.) using countercurrent chromatography. Food Chemistry, 179, 278‒289.
435
Gao, J., Liu, B., Ning, Z., Zhao, R., Zhang, A., & Wu, Q. (2009). Characterization and
436
antioxidant activity of flavonoid-rich extracts from leaves of Ampelopsis
437
grossedentata. Journal of Food Biochemistry, 336, 808‒820.
438
Hou, X. L., Tong, Q., Wang, W. Q., Shi, C. Y., Xiong, W., Chen, J., Liu, X., & Fang, J.
439
G. (2015). Suppression of inflammatory responses by dihydromyricetin, a
440
flavonoid from Ampelopsis grossedentata, via inhibiting the activation of NF-kappa
441
B and MAPK signaling pathways. Journal of Natural Products, 78, 1689‒1696.
442
Hou, X., Zhang, J. F., Ahmad, H., Zhang, H., Xu, Z. W., & Wang, T. (2014).
443
Evaluation of antioxidant activities of ampelopsin and its protective effect in
444
lipopolysaccharide- induced oxidative stress piglets. PLoS One, 9, e108314.
445
Hu, X., Zhao, H., Shi, S., Li, H., Zhou, X., Jiao, F., Jiang, X., Peng, D., & Chen, X.
446
(2015). Sensitive characterization of polyphenolic antioxidants in Polygonatum
447
odoratum by selective solid phase extraction and high performance liquid
448
chromatography‒diode array detector‒quadrupole time-of-flight tandem mass
449
spectrometry. Journal of Pharmaceutical and Biomedical Analysis, 112, 15‒22.
450
Hu, X., Chen, L., Shi, S. Y., Cai, P., Liang, X. J., & Zhang, S. H. (2016). Antioxidant
451
capacity
and
phenolic
compounds
of
452
HPLC‒DAD‒QTOF‒MS/MS. Journal of Pharmaceutical and Biomedical Analysis,
453
124, 254‒260.
Lonicerae
macranthoides
by
454
Karabin, M., Hudcova, T., Jelinek, L., & Dostalek, P. (2015). Biotransformations and
455
biological activities of hop flavonoids. Biotechnology Advances, 33, 1063‒1090.
456
Li, J. F., Fang, H., Yan, X., Chang, F. R., Wu, Z., Wu, Y. L., & Qiu, Y. K. (2016).
457
On-line comprehensive two-dimensional normal-phase liquid chromatography×
458
reversed-phase liquid chromatography for preparative isolation of toad venom. 20
459
Journal of Chromatography A, 1456, 169‒175.
460
Liang, X. J., Zhang, Y. P., Chen, W., Cai, P., Zhang, S. H., Chen, X. Q., & Shi, S. Y.
461
(2015). High-speed counter-current chromatography coupled online to high
462
performance liquid chromatography-diode array detector-mass spectrometry for
463
purification, analysis and identification of target compounds from natural products.
464
Journal of Chromatography A, 1385, 69‒76.
465
Mattila, P. H., Hellström, J., Karhu, S., Pihlava, J., & Veteläinen, M. (2016). High
466
variability
467
North-European currant (Ribes spp.) varieties. Food Chemistry, 204, 14‒20.
468 469
in
flavonoid
contents
and
composition
between
different
Oberlies, N. H., & Kroll, D. J. (2004). Camptothecin and taxol: Historic achievements in natural products research. Journal of Natural Products, 67, 129‒135.
470
Peng, M. J., Zhang, Y. P., & Shi, S. Y. (2016). Separation of polar antioxidants from
471
Rhizoma Polygonatum odorati by high-speed counter-current chromatography with
472
a hydrophilic solvent system. Journal of Liquid Chromatography & Related
473
Technologies, 39, 171‒177.
474 475
Pihlava, J., & Kurtelius, T. (2016). Determination of benzoxazinoids in wheat and rye beers by HPLC-DAD and UPLC-QTOF MS. Food Chemistry, 204, 400‒408.
476
Qiu, J., Chen, L., Zhu, Q., Wang, D., Wang, W., Sun, X., Liu, X., & Du, F. (2012).
477
Screening natural antioxidants in peanut shell using DPPH‒HPLC‒DAD‒TOF/MS
478
methods. Food Chemistry, 135, 2366‒2371.
479
Riethmüller, E., Tóth, G., Alberti, Á., Végh, K., Burlini, I., Könczöl, Á., Balogh, G. T.
480
& Kéry, Á. (2015). First characterisation of flavonoid- and diarylheptanoid-type
481
antioxidant phenolics in Corylus maxima by HPLC-DAD-ESI-MS. Journal of
482
Pharmaceutical and Biomedical Analysis, 107, 159‒167.
483
Rodríguez-Rivera, M. P., Lugo-Cervantes, E., Winterhalter, P., & Jerz, G. (2014). 21
484
Metabolite profiling of polyphenols in peels of Citrus limetta Risso by combination
485
of preparative high-speed countercurrent chromatography and LC–ESI–MS/MS.
486
Food Chemistry, 158, 139–152.
487
Sarah, S. A., Faradalila, W. N., Salwani, M. S., Amin, I., Karsani, S. A., & Sazili, A. Q.
488
(2016). LC–QTOF-MS identification of porcine-specific peptide in heat treated
489
pork identifies candidate markers for meat species determination. Food Chemistry,
490
199, 157‒164.
491
Wang, D. Y., Zheng, Z. Z., Xu, S. Y., & Zheng, S. Z. (2002). Four new isoflavones
492
from Ampelopsis grossedentata. Journal of Asian Natural Products Research, 4,
493
303–308.
494
Yan, L., & Zheng, Z. W. (2009). Effect of ampelopsis of Ampelopsis grossedentata on
495
duck hepatitis B virus. China Journal of Chinese Material Medicine, 34, 908‒910.
496
Yang, W., Zhang, J., Yao, C., Qiu, S., Chen, M., Pan, H., Shi, X., Wu, W., & Guo, D.
497
(2016). Method development and application of offline two-dimensional liquid
498
chromatography/quadrupole time-of-flight mass spectrometry-fast data directed
499
analysis for comprehensive characterization of the saponins from Xueshuantong
500
Injection. Journal of Pharmaceutical and Biomedical Analysis, 128, 322‒332.
501
Ye, L. Y., Wang, H. J., Duncan, S. E.; Eigel, W. N., & O'Keefe, S. F. (2015).
502
Antioxidant activities of vine tea (Ampelopsis grossedentata) extract and its major
503
component dihydromyricetin in soybean oil and cooked ground beef. Food
504
Chemistry, 172, 416‒422.
505
Zhang, S., Cui, Y., Li, L., Li, Y., Zou, P., Luo, L., & Sun, B. (2015). Preparative
506
HSCCC isolation of phloroglucinolysis products from grape seed polymeric
507
proanthocyanidins as new powerful antioxidants. Food Chemistry, 188, 422‒429.
508
Zhang, S. H., Hu, X.; Shi, S. Y., Huang, L. Q., Chen, W., Chen, L., & Cai, P. (2016). 22
509
Typical ultraviolet spectra in combination with diagnostic mass fragmentation
510
analysis for the rapid and comprehensive profiling chlorogenic acids in the buds of
511
Lonicera macranthoides. Analytical and Bioanalytical Chemistry, 408, 3659‒3672.
512
Zhao, H., Hu, X., Chen, X., Shi, S., Jiang, X., Liang, X., Chen, W., & Zhang, S.
513
(2015). Analysis and improved characterization of minor antioxidants from leaves
514
of Malus doumeri using a combination of major constituents' knockout with
515
high-performance
516
time-of-flight tandem mass spectrometry. Journal of Chromatography A, 1398,
517
57‒65.
liquid
chromatography-diode
array
detector-quadrupole
518
Zhao, H. D., Zhang, Y. P., Guo, Y., & Shi, S. Y. (2015). Indentification of
519
α-glucosidase inhibitors in Radix Astragali and its human microsomal metabolites
520
using ultrafiltration HPLC–DAD–MSn. Journal of Pharmaceutical and Biomedical
521
Analysis, 104, 31–37.
522
Zheng, M., Xia, Q., & Lu, S. (2015). Study on drying methods and their influences on
523
effective components of loquat flower tea. LWT-Food Science and Technology, 63,
524
14‒20.
525
Zheng, X. J., Xiao, H., Zeng, Z., Sun, Z. W., Lei, C., Dong, J. Z., & Wang, Y. (2014).
526
Composition and serum antioxidation of the main flavonoids from fermented vine
527
tea (Ampelopsis grossedentata). Journal of Functional Foods, 9, 290‒294.
528
Zhou, X. L., Zhang, Y. P., Zhao, H. D., Liang, J. S., Zhang, Y., & Shi, S. Y. (2015).
529
Antioxidant homoisoflavonoids from Polygonatum odoratum. Food Chemistry, 186,
530
63‒68.
531
Zhou, Y., Shu, F. R., Liang, X. Y., Chang, H., Shi, L. Y., Peng, X. L., Zhu, J. D., & Mi,
532
M. T. (2014). Ampelopsin induces cell growth inhibition and apoptosis in breast
533
cancer cells through ROS generation and endoplasmic reticulum stress pathway. 23
534
PLoS One, 9, e89021.
24
535
Figure captions
536
Fig. 1. HPLC full chromatogram of n-butanol fraction (a) and ethyl acetate fraction (b)
537
of vine tea, and y-axis zoomed-in chromatogram (c) of ethyl acetate fraction of vine
538
tea.
539
Fig. 2. HPLC chromatogram of ethyl acetate fraction of vine tea before (a) and after
540
(b) addition of DPPH. The co-eluted compounds were clearly marked after HSCCC
541
fractionation.
542
Fig. 3. HSCCC chromatogram of ethyl acetate fraction of vine tea (Fr. a, 86‒92 min;
543
Fr. b, 104‒110 min; Fr. c, 128‒134 min; Fr. d, 300‒306 min; Fr. e, 324‒330 min) (A)
544
and HPLC chromatogram of Frs. a‒e (B).
545
Fig. 4. MS/MS spectra of dihydromyricetin (A) and myricetin (B) with its collision
546
energy at 50 eV.
547
Fig. 5. Structures of four compounds (7, 9, 11 and 12) that were firstly reported in
548
vine tea.
25
Table 1 Identified antioxidant flavonoids in vine tea. No.
tR (min)
λmax (nm)
[M+H]+ (m/z) (∆ ppm)
Molecular formula
Fragment ions (m/z) (intensity)
Identification
303.0502 (100), 167.0341 (25),
3-Dihydroxyquercetin
(neutral form) 1
2.716
292
321.0605 (‒2.8)
C15H12O8
153.0183 (12) 2
7.207
291
321.0601 (‒2.8)
C15H12O8
303.0504 (6), 153.0182 (77),
Dihydromyricetin
149.0237 (41), 139.0394 (100) 3
10.222
293
321.0602 (‒2.5)
C15H12O8
303.0504 (7), 153.0184 (71),
Iso-dihydromyricetin
149.0230 (43), 139.0393 (100)
4
14.127
265, 357
481.0960 (‒0.6)
C21H20O13
319.0450 (100)
Myricetin-3-O-β-D-glucoside
5
14.681
290
305.0656 (‒2.0)
C15H12O7
287.0553 (3), 153.0184 (82),
Dihydroquercetin
149.0236 (55), 123.0444 (100)
6
15.326
259, 354
465.1005 (‒0.2)
C21H20O12
319.0448 (100)
Myricitrin
7
17.604
267, 354
435.0909 (‒0.1)
C20H18O11
303.0500 (100)
Quercetin-3-O-β-D-xyloside
8
17.677
253, 374
319.0450 (‒1.3)
C15H10O8
291.0502 (5), 153.0185 (28),
Myricetin
26
139.0393 (100) 9
17.719
284
437.1443 (‒1.8)
C21H24O10
275.0912 (100), 169.0549 (22)
Phloridzin
10
17.811
257, 352
449.1059 (‒0.6)
C21H20O11
303.0501 (100)
Quercetin-3-O-α-L-rhamnoside
11
19.231
285
275.0914 (‒1.8)
C15H14O5
169.0550 (100)
Phloretin
12
19.586
263, 342
433.1125 (‒0.3)
C21H20O10
287.0547 (100)
Kaempferol-3-O-α-L-rhamnoside
13
19.688
254, 367
303.0500 (‒1.6)
C15H10O7
275.0552 (10), 153.0183 (20),
Quercetin
123.0444 (100) 14
21.150
284
303.0862 (‒2.3)
C16H14O6
288.0631 (100), 275.0914 (13),
Hesperitin
153.0182 (37)
15
23.953
265, 364
287.0549 (‒2.4)
C15H10O6
259.0602 (4), 153.0185 (34), 107.0495 (100)
27
Kaempferol
Fig. 1 250 20 150 10
100 c
0
I
b
II III IV
-10
0
5
10 15 Time (min)
20
28
50 0 -50
a
-100 25
Absorbance (mAU)
Absorbance (mAU)
200
Fig. 2 2
Absorbance (mAU)
150
7-10
100 5-6 50 a
3
1
12-13 11 14
4
15
0 -50 b 0
5
10 15 Time (min)
20
29
25
Fig. 3 ×105 A
Absorbance (254 nm)
2.0
1.5
1.0
0.5 a
0
b
c
100
400
170 min mobile phase change
150
B
d
200 250 Time (min)
e
300
350
2
Absorbance (mAU)
300 14
e 15
8
200
5
1
11 13
3
9 12 10
100 7 4 5
c b
6
a
0 0
d
10 15 Time (min)
20
30
25
Fig. 4 100
A
153.0182
80 Relative intensity/%
139.0394 1,2 + B0
1,3
OH
+
A0
OH
60
HO
O OH
40
149.0237 0,2 + A0 -H2O
OH
O +
[M+H] 321.0601 + [M+H-H2O] 303.0504
20
0 100
100
OH
B
150
200
m/z
250
300
139.0393 1,2 + B0 OH
80 Relative intensity/%
350
OH HO
O
OH
60 OH OH
O
40 20 0 100
153.0185 1,3 + A
150
+
[M+H] + [M+H-CO] 319.0450 291.0502
200
250
300
m/z
31
350
Fig. 5 R1
OH
OH HO
HO
O
OH
OR2
OH
OR
O
7 R1 =OH, R2 = β-D-xyl
O
9 R = β-D-Glc
12 R1 =H, R2 = α-L-rha 11 R = H
32
Highlights • Off-line coupling heart-cutting HSCCC with HPLC was developed. • Developed system with high orthogonality resolved similar structures. • Fifteen antioxidative flavonoids were identified by DAD and MS/MS. • Four minor flavonoids were reported in vine tea for the first time.
33
S0308-8146(16)31971-9 http://dx.doi.org/10.1016/j.foodchem.2016.11.122 FOCH 20255
To appear in:
Food Chemistry
Received Date: Revised Date: Accepted Date:
1 July 2016 31 October 2016 22 November 2016
Please cite this article as: Gao, Q., Ma, R., Chen, L., Shi, S., Cai, P., Zhang, S., Xiang, H., Antioxidant profiling of vine tea (Ampelopsis grossedentata): off-line coupling heart-cutting HSCCC with HPLC-DAD-QTOF-MS/MS, Food Chemistry (2016), doi: http://dx.doi.org/10.1016/j.foodchem.2016.11.122
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Graphical Abstract
50 7-10
5-6
2 Absorbance (mAU)
40
A. grossedentata
30 20 4
Exract + DPPH 0
5
10 15 Time (min)
2 50
8 3
1
6 5
Absorbance (mAU)
40
R3 OR1
O
25
60
OH
OH
20
HSCCC
R2
O
15
Exract
0
HO
12-13 14 11
3
1 10
14 13 11
30 20
4 9
10
10 12
7
e d
15
c b a
0 0
5
10 15 Time (min)
1
20
25
1
Antioxidant profiling of vine tea (Ampelopsis grossedentata): off-line coupling
2
heart-cutting HSCCC with HPLC–DAD–QTOF-MS/MS
3 4
Qingping Gaoa,b, Ruyi Mab, Lin Chenc, Shuyun Shib, *, Ping Caic, Shuihan Zhangc,
5
Haiyan Xiangd,**
6 7
a
8
410078, P. R. China
9 10 11 12 13 14
b
Department of Orthodontics, Xiangya Hospital, Central South University, Changsha
College of Chemistry and Chemical Engineering, Central South University,
Changsha 410083, P. R. China c
Research Institute of Chinese Medicine, Hunan Academy of Chinese Medicine,
Changsha 410013, P. R. China d
School of Pharmaceutical Sciences, Southern Medical University, Guangzhou
510515, P.R. China
15 16 17
* Corresponding Author. Tel.: +86 731 88879616.
18
** Corresponding Author. Tel.: +86 20 62789419
19
E-mail addresses: [email protected] (S. Shi), [email protected] (H. Xiang)
20 21
Running title: Antioxidant flavonoids in vine tea
2
22
ABSTRACT
23
Vine tea with strong antioxidant activity is commonly consumed as healthy
24
tea/beverage. However, detailed information about its antioxidants is incomplete.
25
Here, off-line hyphenation of heart-cutting high-speed countercurrent chromatography
26
(HSCCC)
27
detector‒quadrupole
28
(HPLC–DAD‒QTOF-MS/MS) were described for systematic profiling antioxidants in
29
vine
30
1,1-diphenyl-2-picryl-hydrazyl radical‒high performance liquid chromatography
31
(DPPH‒HPLC). Subsequently, stepwise HSCCC using petroleum ether–ethyl
32
acetate–methanol–water (4:9:4:9, v/v/v/v) and (4:9:5:8, v/v/v/v) as solvent systems
33
was optimized to fractionate and enrich antioxidants from ethyl acetate fraction of
34
vine tea. Finally, heart-cutting mode was used to collect five interesting HSCCC
35
fractions for HPLC–DAD‒QTOF-MS/MS analysis. Desirable orthogonality between
36
HSCCC and HPLC led to identification of fifteen antioxidant flavonoids, while four
37
minor flavonoids were first reported in vine tea. Results showed that the developed
38
system is efficient to comprehensively explore antioxidants from complex natural
39
herbs.
tea.
with
At
high
performance
liquid
time-of-flight
first,
chromatography‒diode
tandem
antioxidants
were
mass
rapidly
array
spectrometry
screened
by
40
Vine
tea;
41
Keywords:
42
HPLC–DAD–QTOF-MS/MS
Flavonoid;
3
Antioxidant
activity;
HSCCC;
43
1. Introduction
44
Ampelopsis grossedentata, a plant belonging to Vitaceae family, is distributed
45
widely in mountainous areas of southern China. The tender leaves and stems of A.
46
grossedentata, also called vine tea, have been commonly consumed as healthy tea,
47
beverage and herbal medicine for hundreds of years. Pharmaceutical investigations
48
show that vine tea exhibits significant bioactivities of antioxidant (Hou et al., 2014;
49
Ye, Wang, Duncan, Eigel, & O'Keefe, 2015), anti-inflammatory (Chen et al., 2015;
50
Hou et al., 2015), antiviral (Yan, & Zheng, 2009), antitumor (Zhou et al., 2014),
51
anti-diabetic and anti-hyperglycemic properties (Chen, Wu, Zou, & Gao, 2016).
52
Antioxidant activity has sparked interest due to its relevant to some metabolic
53
diseases (Hu et al., 2016; Zhou et al., 2015). Flavonoids, a kind of phenolic
54
antioxidants existed in many foods and herbs (Karabin, Hudcova, Jelinek, & Dostalek,
55
2015; Mattila, Hellström, Karhu, Pihlava, & Veteläinen, 2016), are found to be the
56
major metabolites in vine tea (Du, Cai, Xia, & Ito; 2002; Du, Chen, Jerz, &
57
Winterhalter, 2004; Gao et al., 2009; Wang, Zheng, Xu, & Zheng, 2002).
58
Dihydromyricetin, a dihydroflavonol with higher 1,1-diphenyl-2-picryl-hydrazyl
59
radical (DPPH) scavenging activity than butylated hydroxytoluene, accounts for
60
around 20% (w/w) on dry weight of vine tea (Du et al., 2004). Thus, most reports
61
focused on the bioactivity assays of dihydromyricetin or crude extract of vine tea.
62
Nevertheless, the therapeutic efficacies of natural products are achieved by
63
combinatorial compounds rather than single or two major compounds. Moreover,
64
some minor compounds (less than 0.1% (w/w) on dry weight of herbs) are found to
65
present significant biological activities and have been developed as clinical drugs or
66
lead compounds for drug discovery (e.g. paclitaxel, vincristine) (Oberlies, & Kroll,
67
2004). Then it is important to comprehensively investigate antioxidant flavonoids in 4
68
vine tea.
69
Compounds in natural products exist with different abundances and structures.
70
DPPH, a paramagnetic compound with an odd electron, would capture one or more
71
hydrogen atoms of antioxidants after spiking with them. Then higher performance
72
liquid chromatography (HPLC) peak areas of antioxidants would reduce or disappear.
73
Therefore, DPPH–HPLC technology has been developed to rapidly and effectively
74
screen antioxidants from complex natural products by comparing their HPLC peak
75
areas before and after spiking with DPPH (Hu et al., 2015; Qiu et al., 2012; Zhao et
76
al., 2015). Notably, no sample pretreatment was contained. HPLC‒diode array
77
detector‒quadrupole
78
(HPLC–DAD‒QTOF-MS/MS) provided ultraviolet (UV) spectra, high-resolution MS
79
and MS/MS spectra, which were useful for structural identification even when
80
standard
81
Segura-Carretero, & Fernández-Gutiérrez, 2016; Sarah et al., 2016; Pihlava, &
82
Kurtelius, 2016; Zhang et al., 2016). High-abundant compounds in vine tea (e.g.
83
dihydromyricetin, myricetin, and myricitrin) have been analyzed and characterized by
84
HPLC system in detail (Du et al., 2004; Gao et al., 2009). From Fig. 1a, four major
85
peaks (I‒IV) existed, in addition, some minor compounds were observed when HPLC
86
chromatogram was zoomed for 13 folds (Fig. 1b). Then, it was necessary to develop
87
some methods to identify minor compounds, especially for those overlapping with
88
major compounds. Recently, two-dimensional (2D) HPLC was developed.
89
Orthogonal columns used in 2D HPLC would provide high peak capacity and
90
resolution for comprehensive compounds analysis (Li et al., 2006; Yang et al., 2016).
91
However, for compounds at very low abundance, some preparative methods were still
92
irreplaceable to enrich them.
compounds
time-of-flight
were
not
tandem
available
5
mass
(Abu-Reidah,
spectrometry
Arráez-Román,
93
High-speed countercurrent chromatography (HSCCC), a unique liquid–liquid
94
partition chromatography method based on partitioning of compounds between two
95
immiscible liquid phases with a support-free matrix, no irreversible adsorption, low
96
risk of sample denaturation and total sample recovery, is an optimal choice for purify
97
major compounds and enrich minor compounds from complex matrix (Esatbeyoglu,
98
Wray, & Winterhalter, 2015; Zhang, et al., 2015; Zhao et al., 2015). Major compounds
99
were knocked out from M. doumeri by HSCCC, and then thirty minor antioxidants
100
were enriched and identified by HPLC–DAD–QTOF-MS/MS (Zhao et al., 2015). To
101
make full use of the orthogonality between HSCCC and HPLC, off-line
102
comprehensive HSCCC × HPLC‒MS/MS was developed, and eighty-five compounds
103
were detected in Citrus limetta (Rodríguez-Rivera, Lugo-Cervantes, Winterhalter, &
104
Jerz, 2014). However, heart-cutting mode is effective by only collection of interesting
105
HSCCC effluents for HPLC analysis.
106
Thus, the aim of this research is to investigate the antioxidant activity and
107
antioxidant compounds in vine tea. In order to achieve antioxidant information, total
108
flavonoid contents and DPPH scavenging activities of three fractions with different
109
polarities were firstly evaluated, and DPPH‒HPLC was then used to rapidly screen
110
antioxidants from the fraction with strongest activity. Co-eluted and minor
111
antioxidants could be successfully separated, enriched and then identified by making
112
full use of the orthogonality between HSCCC and HPLC. The resultant approach
113
found fifteen antioxidants in ethyl acetate fraction of vine tea, and four minor ones
114
were reported in vine tea for the first time.
115 116
2. Experimental
117
2.1. Chemicals and reagents 6
118
All organic solvents used for extraction and separation were of analytical grade
119
(Chemical Reagent Factory of Hunan Normal University, Hunan, China). Methanol
120
and formic acid (Sinopharm Chemical Reagent Co., Ltd, Shanghai, China) were used
121
for HPLC analysis. Deionized water (18.2 MΩ) was obtained from a Milli-Q water
122
purification system (Millipore, Bedford, MA, USA). Ascorbic acid (VC) and DPPH
123
(95%) were bought from Sigma-Aldrich (Shanghai Division). Nine flavonoid
124
standards, dihydromyricetin, myricetin, dihydroquercetin, quercetin, hesperitin,
125
kaempferol,
126
kaempferol-3-O-α-L-rhamnoside, were purchased from the National Institute for the
127
Control of Pharmaceutical and Biological Products (Beijing, China). Phloridzin and
128
phloretin with purities over 98 % were purified from Malus doumeri by two-step
129
HSCCC in our laboratory. Firstly, phloridzin was purified by HSCCC using ethyl
130
acetate–n-butanol–methanol–25 mM ammonium acetate solution (3.5:1.5:1:4, v/v/v/v)
131
as solvent system. Then, extruded sample from the first HSCCC separation was
132
injected into the second HSCCC to purify phloretin using petroleum ether–ethyl
133
acetate–methanol–water (1:2:1.5:1.5, v/v/v/v) as solvent system. Their structures were
134
identified by UV, MS and nuclear magnetic resonance (NMR) (Zhao et al., 2015).
135
2.2. Preparation of vine tea extract
myricitrin,
quercetin-3-O-α-L-rhamnoside,
and
136
Vine tea was collected in May 2015 from Jianghua (Hunan province of China,
137
Northern altitude 25°15', longitude 112°46', altitude: 610 m). The plant material was
138
identified using a species identification key (Editorial committee of flora of China,
139
1998) by Prof. Zhaoming Xie, Research Institute of Chinese Medicine, Hunan
140
Academy of Chinese Medicine, Changsha, China. After collection, vine tea was
141
immediately dried at 40 °C in an oven with air circulation. Dry vine tea was ground
142
and sieved, and materials between 180 and 250 µm was used for extraction. Vine tea 7
143
powder (25.0 g) was extracted with 70% ethanol (250 mL, three times) at 85°C (each
144
for 3 h). After filtration, organic solvent was removed from the combined extracts
145
under reduced pressure to yield a total crude extract (11.4 g). Crude extract,
146
suspended in water (100 mL), was then extracted successively with 3 × 100 mL of
147
petroleum ether, ethyl acetate and n-butanol. After extraction, solvents were removed
148
under reduced pressure, and then petroleum ether fraction (0.2 g), ethyl acetate
149
fraction (1.9 g) and n-butanol fraction (1.4 g) were stored at 4ºC for further
150
experiments (Liang et al., 2015).
151
2.3. Determination of total flavonoid contents
152
Total flavonoid contents were estimated by NaNO2‒Al(NO3)3 colorimetric method
153
(Zheng, Xia, & Lu, 2015). Vine tea extract (1.0 mg/mL, 0.4 mL) was mixed with
154
NaNO2 solution (4 %, 0.4 mL). After standing for 6 min, Al(NO3)3 solution (9 %, 0.4
155
mL) were added and incubated for another 6 min at room temperature. Then NaOH
156
solution (1.6 mol/L, 2.0 mL) was added and kept for 12 min. Finally, absorbance was
157
measured at 510 nm against the control. Total flavonoids contents were calculated
158
using a standard calibration of quercetin (4.0‒14.0 µg/mL) and expressed as mg of
159
quercetin equivalent (QE) per g of dry weight (DW) sample (mg QE/g DW), while
160
data are presented as means ± standard deviation for three replicates.
161
2.4. DPPH radical scavenging assay
162
DPPH scavenging activity, with VC as standard, was assayed as previously
163
reported procedures (Zhao et al., 2015). Briefly, samples or standards with different
164
concentrations (25 µL, 2, 5, 10, 30, 50, 100, and 200 µg/mL) were mixed with DPPH
165
solution (40 µL, 0.4 mg/mL) and then diluted with methanol to 250 µL. After
166
incubating at 37 °C for 30 min, the absorbance was measured at 517 nm. DPPH
167
solution (0.064 mg/mL) without any sample was a control. The antioxidant activity is 8
168
expressed as percentage of DPPH radical elimination: [(Ablank–Asample)/Ablank] × 100 %
169
(Ablank and Asample are the absorbance of the control and sample after addition of
170
DPPH). Sample concentration providing 50% inhibition (IC50) was calculated from
171
the graph plotting inhibition percentage.
172
2.5. DPPH‒HPLC experiment
173
HPLC analysis was carried out using an Agilent 1200 series HPLC system (Agilent
174
Technologies, Santa Clara, CA), equipped with an online vacuum degasser, a
175
QuatPump, a manual injection valve with a 20 µL sample loop, a thermostated
176
column compartment, a diode assay detector (DAD), and an Agilent ChemStation.
177
Chromatographic separation was performed on a Waters Symmetry C18 column (150
178
mm × 3.9 mm i.d., 5 µm, Waters, MA, USA) in tandem with a Phenomenex C18 guard
179
cartridge (4.0 mm × 3.0 mm, Phenomenex, Torrance, CA). Water containing 0.1%
180
formic acid (A) and methanol containing 0.1% formic acid (B) were selected as the
181
mobile phase in gradient elution: 0–5 min, 25% B; 5–20 min, 25–60% B; 20–25 min,
182
60–75% B. The flow rate and the column temperature were set at 0.8 mL/min and
183
25°C, respectively. The chromatogram was acquired at 254 nm.
184
DPPH‒HPLC experiment was carried out according to our previous reports (Hu et
185
al., 2015; Peng, Zhang, & Shi, 2016). Typically, ethyl acetate fraction of vine tea (2.0
186
mg/mL, 300 µL) and DPPH (30.0 mg/mL, 300 µL) were mixed and incubated at
187
37 °C for 30 min. After passing through a 0.45 µm polytetrafluoroethylene syringe
188
filter, reaction mixtures were analyzed by HPLC in comparison with the control,
189
adding methanol (300 µL) into ethyl acetate fraction of vine tea (2.0 mg/mL, 300 µL).
190
Subsequently, by comparison with the chromatogram of control, peaks in reaction
191
mixtures disappeared or decreased could be considered as potential antioxidants.
192
2.6. HSCCC separation 9
193
Preparative HSCCC was performed using a model TBE-300B HSCCC (Shanghai
194
Tauto Biotechnique Co. Ltd., Shanghai, China), which was equipped with three
195
multilayered coils connected in series (diameter of tube, 1.6 mm, total capacity 260
196
mL) and a 20 mL manual sample loop. The rotation speed was adjustable, ranging
197
from 0 to 1000 rpm. In addition, HSCCC system was equipped with a TBP-1002
198
pump, a TBD-2000 UV detector, a HX-1050 constant temperature regulator (Beijing
199
Boyikang Lab Implement Co. Ltd., Beijing, China), and a WH V4.0 workstation
200
(Shanghai Wuhao Information Technology Co. Ltd., Shanghai, China).
201
Suitable solvent systems for HSCCC were selected according to K values of target
202
compounds as our previously described (Zhang, et al., 2015; Zhao et al., 2015). A
203
series of solvent systems composed of petroleum ether–ethyl acetate–methanol–water
204
were prepared, and ethyl acetate fraction of vine tea (0.25 mg) was added into solvent
205
system (1 mL). After thorough equilibration for 20 min, 20 µL of upper and lower
206
phases of solvent system were analyzed by HPLC, respectively, to obtain K values of
207
each compound in ethyl acetate fraction of vine tea. The K is defined as K =
208
Aupper/Alower, where Aupper and Alower were the HPLC peak area of target compound in
209
the upper and lower phases, respectively.
210
Two solvent systems composed of petroleum ether–ethyl acetate–methanol–water
211
(4:9:4:9, v/v/v/v) and (4:9:5:8, v/v/v/v) were selected for HSCCC separation, which
212
were prepared in a separation funnel according to volume ratios, and thoroughly
213
equilibrated by shaking repeatedly. The upper phase (stationary phase) and lower
214
phase (mobile phase) were then separated and degassed by ultrasonication for 30 min
215
shortly prior to use. Ethyl acetate fraction of vine tea (6.0 mg/mL) in 20 mL of lower
216
phase of petroleum ether–ethyl acetate–methanol–water (4:9:4:9, v/v/v/v) was
217
prepared as sample solution. HSCCC column was first entirely filled with the upper 10
218
phase of petroleum ether–ethyl acetate–methanol–water (4:9:4:9, v/v/v/v) as the
219
stationary phase; subsequently, the lower mobile phase was then pumped into the inlet
220
of the column at the flow rate of 1.5 mL/min, while the apparatus was run at 900 rpm;
221
after a clear mobile phase eluted at the tail outlet and the hydrodynamic equilibrium
222
was reached, 20 mL of sample solution (6.0 mg/mL) was then injected into the
223
injection valve. HSCCC separation was performed by a stepwise mode with lower
224
phase of petroleum ether–ethyl acetate–methanol–water (4:9:4:9, v/v/v/v) as the
225
mobile phase in the period of 0–170 min and then lower phase of petroleum
226
ether–ethyl acetate–methanol–water (4:9:5:8, v/v/v/v) in the period of 170–350 min.
227
The HSCCC chromatogram was acquired at 254 nm. After HSCCC separation was
228
completed, the solvent in HSCCC column was pumped out by pressured nitrogen.
229
2.7. HPLC–DAD–QTOF-MS/MS analysis of collected HSCCC fractions
230
HSCCC fractionation was performed as a function of time, with a collection time of
231
6.0 min per tube. The collected fractions were dried, and then redissolved in methanol
232
for HPLC–DAD–QTOF-MS/MS analysis.
233
HPLC separation column, mobile phase, gradient program and flow rate were the
234
same
as
those
presented
in
section
235
Accurate-Mass QTOF LC/MS system with an Agilent Jet Stream electrospray (ESI)
236
interface was coupled in parallel by splitting the mobile phase 1:3 using an adjustable
237
high-pressure stream splitter (Valco Instrument Company, Houston, TX, USA). The
238
MS settings for flavonoid characterization were optimized according to previously
239
reported (Zhao, Zhang, Guo, & Shi, 2015): scan range 100–1000 m/z, capillary
240
voltage 3.5 kV, dry gas temperature 320 ºC, dry gas flow rate 10.0 L/min, sheath gas
241
temperature 400 ºC, sheath gas flow rate 12 L/min, nebulizing gas pressure 35 psi,
242
fragmentor voltage 160 V. For collision-induced dissociation (CID) experiments, 11
2.5.
In
addition,
Agilent
6530
243
keeping MS1 static, the precursor ion of interest was selected using the quadrupole
244
analyzer and the product ions were analyzed using a time-of-flight (TOF) analyzer.
245
The mass axis was calibrated using mixtures provided by manufacturer.
246
2.8. Statistical analysis
247
Total flavonoid content and DPPH radical scavenging assay were conducted in
248
triplicate (n =3), and the results were expressed as mean ± SD (standard deviation),
249
while p < 0.05 represented a statistically significant difference. Statistical evaluation
250
was performed using Statistical Analysis System (version 9.2, SAS Institute Inc., Cary,
251
NC).
252 253
3. Results and discussion
254
3.1. Determination of total flavonoid contents and antioxidant activity
255
Flavonoids were considered as the widely existed antioxidants in natural products,
256
and our previous report indicated that total flavonoid contents had high positive
257
correlation to the antioxidant activities (Zheng et al., 2015). Accelerated solvent
258
extraction was always used to partition complex matrix. Here, 70% ethanol extract of
259
vine tea was fractionated successively by petroleum ether, ethyl acetate and n-BuOH.
260
The results indicated that ethyl acetate fraction yielded the highest total flavonoid
261
content (12.19 ± 2.03 mg QE/g DW), followed by n-butanol fraction (7.74 ± 1.16 mg
262
QE/g DW). Petroleum ether fraction had the lowest value (0.26 ± 0.09 mg QE/g DW).
263
Therefore, extraction solvent significantly (p < 0.05) affected the total flavonoid
264
contents. IC50 values of ethyl acetate fraction and n-butanol fraction to DPPH
265
scavenging were 3.05 ± 0.50 and 4.90 ± 0.52 µg/mL, respectively. However,
266
petroleum ether fraction did not show any antioxidant activities (IC50 > 20 µg/mL).
267
The DPPH scavenging activities exhibited the same trends with total flavonoid 12
268
contents. It is noted that ethyl acetate fraction and n-butanol fraction exhibited strong
269
antioxidant capacities in comparison to positive control, VC (IC50, 19.17 ± 3.21
270
µg/mL) (p < 0.05). Therefore, vine tea was rich in natural antioxidants for scavenging
271
biologically relevant radicals.
272
3.2. DPPH–HPLC analysis
273
The mobile phase system was firstly optimized to obtain reliable chromatographic
274
results. The final results showed that best resolution, shortest analysis time and lowest
275
pressure variations were achieved when a gradient elution mode composed of solvent
276
A (0.1% formic acid in water) and B (0.1% formic acid in methanol) was programmed
277
as follows: 0–5 min, 25% B; 5–20 min, 25–60% B; 20–25 min, 60–75% B. The flow
278
rate was 0.8 mL/min while the column temperature was 25°C. UV spectra were
279
recorded from 190 to 400 nm, while the chromatogram was acquired at 254 nm.
280
Flavonoid profiling of ethyl acetate fraction and n-butanol fraction of vine tea by
281
HPLC were almost the same, and the difference between them was their quantities
282
(Fig. 1), then subsequently, ethyl acetate fraction of vine tea was investigated. It was
283
obvious that there existed four major peaks (I‒IV) (Fig. 1b). Vine tea was reported to
284
contain two types of flavonoids, dihydroflavonols and flavonols (Du et al., 2004; Gao
285
et al., 2009; Wang et al., 2002). Dihydroflavonols had maximum UV absorption band
286
near 285 nm, while flavonols showed two maximum UV absorptions at 250‒270 and
287
340‒380 nm. A comprehensive UV screening showed that peaks II and III contained
288
two or more flavonoids, and UV spectra of peak IV (λmax: 249, 298, 333 nm) was not
289
a kind of flavonoid. In addition, there existed some minor flavonoids (Fig. 1c).
290
However, because of the co-elution and matrix interference, UV and MS data of
291
minor
292
HPLC–QTOF-MS/MS analysis. Therefore, fractionation and enrichment of minor
flavonoids
were
difficult
to
13
be
accurately
achieved
only
by
293
flavonoids were inevitable.
294
DPPH–HPLC was always used as a rapid method to screen antioxidants from
295
complex mixtures without sample pretreatment. It was believed structures of
296
antioxidants will be changed after they react with DPPH. Thus, HPLC peak areas of
297
antioxidants will decrease or disappear. Fig. 2 showed the chromatogram of ethyl
298
acetate fraction of vine tea before and after reaction with DPPH, which presented that
299
peak areas of fifteen flavonoids (1–15) disappeared or decreased obviously after
300
spiking with DPPH. Therefore, flavonoids 1–15 possessed potent antioxidant activity.
301
3.3. HSCCC separation
302
Most solvent systems were limited to separating compounds with a narrow range of
303
polarities, and then it was needed to apply a multistep elution using more than one
304
solvent system. Petroleum ether–ethyl acetate–methanol–water solvent system,
305
providing a broad range of polarities by modifying volume ratios of four solvents, has
306
been testified as be useful in purification of components in ethyl acetate fraction
307
(Zhao et al., 2015). The 0.5≤K≤2.0 in "test tube" experiments is always considered as
308
suitable K values for HSCCC. According to systematic analysis of K values of target
309
antioxidant flavonoids, stepwise HSCCC separation using petroleum ether–ethyl
310
acetate–methanol–water (4:9:4:9, v/v/v/v) and (4:9:5:8, v/v/v/v) as solvent systems
311
were selected (Fig. 3A). Obviously, HSCCC fractionated ethyl acetate fraction of vine
312
tea to four peaks, but each HSCCC peak contained a major compound and some
313
minor compounds. Collected HSCCC fractions were analyzed by off-line HPLC, and
314
five fractions (a‒e) contained target flavonoids (Fig. 3B). It is noted that HSCCC and
315
HPLC have orthogonality, which make it possible to identify co-eluted flavonoids.
316
3.4. Characterization of antioxidant flavonoids by HPLC‒DAD‒QTOF-MS/MS
317
A MS detector with positive ion mode in combination with DAD was applied to 14
318
interpret antioxidants by comparison with standards or published data. Table 1
319
showed the UV, MS and MS/MS spectral data and characterization results for these
320
compounds. Compounds 1‒3, 5, 9, 11, and 14 had typical UV spectra of
321
dihydroflavonols/flavanones/dihydrochalcones, while UV spectra of compounds 4,
322
6‒7, 8, 10, 12‒13, 15 were for flavonols.
323
Dihydroflavonols cannot be differentiated from flavanones/dihydrochalcones only
324
by UV spectra. However, dihydroflavonols are always more polar and be firstly eluted
325
in HPLC analysis. The early retention times of compounds 1‒3 and 5 suggested that
326
they might be dihydroflavonols, while compounds 9, 11 and 14 with lower polarities
327
might be flavanones/dihydrochalcones (Fig. 2).
328
Compound 2, the major compound in vine tea, was identified as dihydromyricetin
329
by comparison with standard, which was then taken as an example to explain the
330
fragmentation details of dihydroflavonols (Fig. 4A). The cleavages of C-ring at
331
positions 1/3, 1/2 and 0/2 were their characteristic fragmentations. Therefore, the
332
structurally informative ions, m/z 153.0182 (1,3A0+), 149.0237 (0,2A0+‒H2O), and
333
139.0394 (1,2B0+), were diagnostic for A- and B-ring substitutions. In addition,
334
dihydromyricetin displayed a fragment ion at m/z 303.0504 (C15H11O7) for the loss of
335
H2O of C-ring.
336
Both compounds 1 and 3 afforded [M+H]+ ion at m/z 321.06 (C15H13O8), the same
337
with that of 2. It was hard to distinguish compounds 2 and 3 by their UV and MS/MS
338
spectra, but they had different HPLC retention behaviors. As a result, the structural
339
difference between compounds 2 and 3 should be the configurations of C‒2 and C‒3.
340
Then, by comparison with previously published data (Zheng et al., 2014), compound
341
3 was tentatively identified as iso-dihydromyricetin. Compound 1 presented MS/MS
342
base peak at m/z 321.0605 ([M‒H2O+H]+) and typical fragment ions at m/z 167.0341 15
343
(0,2A0+) and 153.0183 (1,3A0+), indicating the same A-ring structure with that of 2.
344
Thinks to shorter retention time, compound 1 was allowed to be elucidated as
345
3-dihydroxyquercetin, which was previously reported in fermented vine tea (Hu et al.,
346
2016). The mass spectra of compound 5 displayed [M+H]+ ion at m/z 305.0656
347
(C15H13O7, 16 Da less than that of 2), and the 16 Da mass difference was also found
348
for
349
between compounds 5 and 2 was the numbers of hydroxyl groups on B-ring. Based on
350
the observations, compound 5 was tentatively established as dihydroquercetin, which
351
was finally confirmed using standard.
1,2
B0+ fragment ion between compounds 5 and 2. Thus, the structural difference
352
Compounds 9 and 11 exhibited [M+H]+ ions at m/z 437.1443 (C21H25O10) and
353
275.0914 (C15H15O5), respectively. The fragment ion with a high intensity of 9 at m/z
354
275.0912 ([M‒162+H]+, the same with the parent ion of 11, indicated the existence of
355
an additional glucosyl unit in 9. In addition, the m/z 169.0550 (C8H9O4) might be the
356
characteristic fragment ion of dihydrochalcones by the cleavage between C‒α and
357
C‒β. Thus, compounds 9 and 11 were assigned as phloridzin and phloretin (Fig. 5),
358
respectively, by comparing with their standards, while the existence of them in vine
359
tea had not been previously reported. Compound 14 showed [M+H]+ ion at m/z
360
303.0862 (C16H15O6) along with three fragment ions at m/z 288.0631 ([M‒CH3+H]+),
361
275.0914 ([M‒CO+H]+), and 153.0182 (1,3A0+), which was consistent with the
362
previously isolated compound, hesperitin.
363
The RDA rearrangement 1,3A0+ ion for the cleavage of C-ring was the characteristic 1,2
B0+ fragment ion was typical for
364
fragmentation for flavonols and flavones, however,
365
flavonols (Zhao et al., 2015). As shown in Fig. 4B, myricetin (compound 8),
366
unequivocally identified by comparing their HPLC‒DAD‒MS/MS data with those of
367
standard, displayed parent ion [M+H]+ at m/z 319.0450 (C15H11O8) and typical 16
368
fragment ions at m/z 153.0185 (1,3A0+) and 139.0393 (1,2B0+). Compounds 13 and 15
369
exhibited [M+H]+ ions at m/z 303.0500 (C15H11O7, 16 Da less than that of myricetin)
370
and 287.0549 (C15H11O6, 32 Da less than that of myricetin), which gave the same
371
1,3
A0+ fragment ion but different
1,2
B0+ fragment ion. Then it was evident that
372
compound 13 have two hydroxyl groups on B-ring, and 15 contained only one
373
hydroxyl group on B-ring. By comparing with standards, compounds 13 and 15 were
374
finally elucidated as quercetin and kaempferol. Compounds 4 and 6 afforded [M+H]+
375
ions at m/z 481.0960 (C21H21O13) and 465.1005 (C21H21O12), which gave the same
376
base peak at m/z 319.0448 (C15H11O8), mostly probably by loss of a glucosyl residue
377
(162 Da) and a rhamnosyl group (146 Da). Accordingly, compound 4 was tentatively
378
established as myricetin-3-O-β-D-glucoside, while compound 6 was deduced as
379
myricitrin (myricetin-3-O-α-L-rhamnoside) by comparing with standard. Using
380
similar
381
quercetin-3-O-β-D-xyloside,
382
kaempferol-3-O-α-L-rhamnoside by comparing them with standards or previously
383
reported data (Riethmüller et al., 2015). It was noted that compound 7 and 12 has not
384
been previously reported in vine tea (Fig. 5).
principles,
compounds
7,
10
and
12
were
quercetin-3-O-α-L-rhamnoside,
assigned
as and
385 386
4. Conclusions
387
In this study, total flavonoid contents in vine tea extracts with different polarities
388
and their antioxidant activities were studied. Ethyl acetate and n-butanol fractions
389
contained more flovonoids contents and exhibited stronger antioxidant capacities than
390
petroleum ether fraction and VC. Same HPLC flavonoid profiles with different
391
quantities were observed between ethyl acetate and n-butanol fractions. DPPH‒HPLC
392
analysis was then used to screen fifteen antioxidants in ethyl acetate fraction. The 17
393
co-eluted or minor antioxidants were separated or enriched by stepwise HSCCC, and
394
five interesting HSCCC fractions were heart-cut and further analyzed offline by
395
HPLC–DAD–QTOF-MS/MS. Structure of each antioxidant was finally elucidated by
396
comparison of retention time, DAD and MS/MS data, easily achieved because of the
397
orthogonality between HSCCC and HPLC, with standards or published data. It is
398
worth mentioning that four minor antioxidants were reported in vine tea for the first
399
time. The comprehensive knowledge of antioxidants in vine tea could make benefit
400
for its further application in food supplementation, cosmetics and medicinal products.
401
Moreover, the developed method provided an efficient way to investigate bioactive
402
compounds, especially for co-eluted and minor ones, in complex system.
403 404
Acknowledgments
405
This work was supported by the National Natural Science Foundation of China
406
(21275163, 81402935), the Science and Technology Program of Hunan Province,
407
China (2015NK3037), Program of Survey and Monitoring of Chinese Medicines for
408
National Drugs ([2011] 76).
18
409
References
410
Abu-Reidah, I., Arráez-Román, D., Segura-Carretero, A., & Fernández-Gutiérrez, A.
411
(2016). Extensive characterisation of bioactive phenolic constituents from globe
412
artichoke (Cynara scolymus L.) by HPLC–DAD-ESI-QTOF-MS. Food Chemistry,
413
141, 2269‒2277.
414
Chen, J., Wu, Y. C., Zou, J. W., & Gao, K. (2016). α-Glucosidase inhibition and
415
antihyperglycemic activity of flavonoids from Ampelopsis grossedentata and the
416
flavonoid derivatives. Bioorganic & Medicinal Chemistry, 24, 1488‒1494.
417
Chen, S. H., Zhao, X. L., Wan, J., Ran, L., Qin, Y., Wang, X. F., Gao, Y. X., Shu, F. R.,
418
Zhang, Y., Liu, P., Zhang, Q. Y., Zhu, J. D., & Mi, M. T. (2015). Dihydromyricetin
419
improves glucose and lipid metabolism and exerts anti-inflammatory effects in
420
nonalcoholic fatty liver disease: A randomized controlled trial. Pharmacological
421
Research, 99, 74‒81.
422
Du, Q. Z., Cai, W. J., Xia, M., & Ito, Y. (2002). Purification of (+)-dihydromyricetin
423
from leaves extract of Ampelopsis grossedentata using high-speed countercurrent
424
chromatography with scale-up triple columns. Journal of Chromatography A, 973,
425
217‒220.
426
Du, Q. Z., Chen, P., Jerz, G., & Winterhalter, P. (2004). Preparative separation of
427
flavonoid glycosides in leaves extract of Ampelopsis grossedentata using
428
high-speed counter-current chromatography. Journal of Chromatography A, 1040,
429
147‒149.
430 431
Editorial committee of flora of China, (1998). Flora of China. Beijing: Science Press, vol. 48, p. 53.
432
Esatbeyoglu, T., Wray, V., & Winterhalter, P. (2015). Isolation of dimeric, trimeric,
433
tetrameric and pentameric procyanidins from unroasted cocoa beans (Theobroma 19
434
cacao L.) using countercurrent chromatography. Food Chemistry, 179, 278‒289.
435
Gao, J., Liu, B., Ning, Z., Zhao, R., Zhang, A., & Wu, Q. (2009). Characterization and
436
antioxidant activity of flavonoid-rich extracts from leaves of Ampelopsis
437
grossedentata. Journal of Food Biochemistry, 336, 808‒820.
438
Hou, X. L., Tong, Q., Wang, W. Q., Shi, C. Y., Xiong, W., Chen, J., Liu, X., & Fang, J.
439
G. (2015). Suppression of inflammatory responses by dihydromyricetin, a
440
flavonoid from Ampelopsis grossedentata, via inhibiting the activation of NF-kappa
441
B and MAPK signaling pathways. Journal of Natural Products, 78, 1689‒1696.
442
Hou, X., Zhang, J. F., Ahmad, H., Zhang, H., Xu, Z. W., & Wang, T. (2014).
443
Evaluation of antioxidant activities of ampelopsin and its protective effect in
444
lipopolysaccharide- induced oxidative stress piglets. PLoS One, 9, e108314.
445
Hu, X., Zhao, H., Shi, S., Li, H., Zhou, X., Jiao, F., Jiang, X., Peng, D., & Chen, X.
446
(2015). Sensitive characterization of polyphenolic antioxidants in Polygonatum
447
odoratum by selective solid phase extraction and high performance liquid
448
chromatography‒diode array detector‒quadrupole time-of-flight tandem mass
449
spectrometry. Journal of Pharmaceutical and Biomedical Analysis, 112, 15‒22.
450
Hu, X., Chen, L., Shi, S. Y., Cai, P., Liang, X. J., & Zhang, S. H. (2016). Antioxidant
451
capacity
and
phenolic
compounds
of
452
HPLC‒DAD‒QTOF‒MS/MS. Journal of Pharmaceutical and Biomedical Analysis,
453
124, 254‒260.
Lonicerae
macranthoides
by
454
Karabin, M., Hudcova, T., Jelinek, L., & Dostalek, P. (2015). Biotransformations and
455
biological activities of hop flavonoids. Biotechnology Advances, 33, 1063‒1090.
456
Li, J. F., Fang, H., Yan, X., Chang, F. R., Wu, Z., Wu, Y. L., & Qiu, Y. K. (2016).
457
On-line comprehensive two-dimensional normal-phase liquid chromatography×
458
reversed-phase liquid chromatography for preparative isolation of toad venom. 20
459
Journal of Chromatography A, 1456, 169‒175.
460
Liang, X. J., Zhang, Y. P., Chen, W., Cai, P., Zhang, S. H., Chen, X. Q., & Shi, S. Y.
461
(2015). High-speed counter-current chromatography coupled online to high
462
performance liquid chromatography-diode array detector-mass spectrometry for
463
purification, analysis and identification of target compounds from natural products.
464
Journal of Chromatography A, 1385, 69‒76.
465
Mattila, P. H., Hellström, J., Karhu, S., Pihlava, J., & Veteläinen, M. (2016). High
466
variability
467
North-European currant (Ribes spp.) varieties. Food Chemistry, 204, 14‒20.
468 469
in
flavonoid
contents
and
composition
between
different
Oberlies, N. H., & Kroll, D. J. (2004). Camptothecin and taxol: Historic achievements in natural products research. Journal of Natural Products, 67, 129‒135.
470
Peng, M. J., Zhang, Y. P., & Shi, S. Y. (2016). Separation of polar antioxidants from
471
Rhizoma Polygonatum odorati by high-speed counter-current chromatography with
472
a hydrophilic solvent system. Journal of Liquid Chromatography & Related
473
Technologies, 39, 171‒177.
474 475
Pihlava, J., & Kurtelius, T. (2016). Determination of benzoxazinoids in wheat and rye beers by HPLC-DAD and UPLC-QTOF MS. Food Chemistry, 204, 400‒408.
476
Qiu, J., Chen, L., Zhu, Q., Wang, D., Wang, W., Sun, X., Liu, X., & Du, F. (2012).
477
Screening natural antioxidants in peanut shell using DPPH‒HPLC‒DAD‒TOF/MS
478
methods. Food Chemistry, 135, 2366‒2371.
479
Riethmüller, E., Tóth, G., Alberti, Á., Végh, K., Burlini, I., Könczöl, Á., Balogh, G. T.
480
& Kéry, Á. (2015). First characterisation of flavonoid- and diarylheptanoid-type
481
antioxidant phenolics in Corylus maxima by HPLC-DAD-ESI-MS. Journal of
482
Pharmaceutical and Biomedical Analysis, 107, 159‒167.
483
Rodríguez-Rivera, M. P., Lugo-Cervantes, E., Winterhalter, P., & Jerz, G. (2014). 21
484
Metabolite profiling of polyphenols in peels of Citrus limetta Risso by combination
485
of preparative high-speed countercurrent chromatography and LC–ESI–MS/MS.
486
Food Chemistry, 158, 139–152.
487
Sarah, S. A., Faradalila, W. N., Salwani, M. S., Amin, I., Karsani, S. A., & Sazili, A. Q.
488
(2016). LC–QTOF-MS identification of porcine-specific peptide in heat treated
489
pork identifies candidate markers for meat species determination. Food Chemistry,
490
199, 157‒164.
491
Wang, D. Y., Zheng, Z. Z., Xu, S. Y., & Zheng, S. Z. (2002). Four new isoflavones
492
from Ampelopsis grossedentata. Journal of Asian Natural Products Research, 4,
493
303–308.
494
Yan, L., & Zheng, Z. W. (2009). Effect of ampelopsis of Ampelopsis grossedentata on
495
duck hepatitis B virus. China Journal of Chinese Material Medicine, 34, 908‒910.
496
Yang, W., Zhang, J., Yao, C., Qiu, S., Chen, M., Pan, H., Shi, X., Wu, W., & Guo, D.
497
(2016). Method development and application of offline two-dimensional liquid
498
chromatography/quadrupole time-of-flight mass spectrometry-fast data directed
499
analysis for comprehensive characterization of the saponins from Xueshuantong
500
Injection. Journal of Pharmaceutical and Biomedical Analysis, 128, 322‒332.
501
Ye, L. Y., Wang, H. J., Duncan, S. E.; Eigel, W. N., & O'Keefe, S. F. (2015).
502
Antioxidant activities of vine tea (Ampelopsis grossedentata) extract and its major
503
component dihydromyricetin in soybean oil and cooked ground beef. Food
504
Chemistry, 172, 416‒422.
505
Zhang, S., Cui, Y., Li, L., Li, Y., Zou, P., Luo, L., & Sun, B. (2015). Preparative
506
HSCCC isolation of phloroglucinolysis products from grape seed polymeric
507
proanthocyanidins as new powerful antioxidants. Food Chemistry, 188, 422‒429.
508
Zhang, S. H., Hu, X.; Shi, S. Y., Huang, L. Q., Chen, W., Chen, L., & Cai, P. (2016). 22
509
Typical ultraviolet spectra in combination with diagnostic mass fragmentation
510
analysis for the rapid and comprehensive profiling chlorogenic acids in the buds of
511
Lonicera macranthoides. Analytical and Bioanalytical Chemistry, 408, 3659‒3672.
512
Zhao, H., Hu, X., Chen, X., Shi, S., Jiang, X., Liang, X., Chen, W., & Zhang, S.
513
(2015). Analysis and improved characterization of minor antioxidants from leaves
514
of Malus doumeri using a combination of major constituents' knockout with
515
high-performance
516
time-of-flight tandem mass spectrometry. Journal of Chromatography A, 1398,
517
57‒65.
liquid
chromatography-diode
array
detector-quadrupole
518
Zhao, H. D., Zhang, Y. P., Guo, Y., & Shi, S. Y. (2015). Indentification of
519
α-glucosidase inhibitors in Radix Astragali and its human microsomal metabolites
520
using ultrafiltration HPLC–DAD–MSn. Journal of Pharmaceutical and Biomedical
521
Analysis, 104, 31–37.
522
Zheng, M., Xia, Q., & Lu, S. (2015). Study on drying methods and their influences on
523
effective components of loquat flower tea. LWT-Food Science and Technology, 63,
524
14‒20.
525
Zheng, X. J., Xiao, H., Zeng, Z., Sun, Z. W., Lei, C., Dong, J. Z., & Wang, Y. (2014).
526
Composition and serum antioxidation of the main flavonoids from fermented vine
527
tea (Ampelopsis grossedentata). Journal of Functional Foods, 9, 290‒294.
528
Zhou, X. L., Zhang, Y. P., Zhao, H. D., Liang, J. S., Zhang, Y., & Shi, S. Y. (2015).
529
Antioxidant homoisoflavonoids from Polygonatum odoratum. Food Chemistry, 186,
530
63‒68.
531
Zhou, Y., Shu, F. R., Liang, X. Y., Chang, H., Shi, L. Y., Peng, X. L., Zhu, J. D., & Mi,
532
M. T. (2014). Ampelopsin induces cell growth inhibition and apoptosis in breast
533
cancer cells through ROS generation and endoplasmic reticulum stress pathway. 23
534
PLoS One, 9, e89021.
24
535
Figure captions
536
Fig. 1. HPLC full chromatogram of n-butanol fraction (a) and ethyl acetate fraction (b)
537
of vine tea, and y-axis zoomed-in chromatogram (c) of ethyl acetate fraction of vine
538
tea.
539
Fig. 2. HPLC chromatogram of ethyl acetate fraction of vine tea before (a) and after
540
(b) addition of DPPH. The co-eluted compounds were clearly marked after HSCCC
541
fractionation.
542
Fig. 3. HSCCC chromatogram of ethyl acetate fraction of vine tea (Fr. a, 86‒92 min;
543
Fr. b, 104‒110 min; Fr. c, 128‒134 min; Fr. d, 300‒306 min; Fr. e, 324‒330 min) (A)
544
and HPLC chromatogram of Frs. a‒e (B).
545
Fig. 4. MS/MS spectra of dihydromyricetin (A) and myricetin (B) with its collision
546
energy at 50 eV.
547
Fig. 5. Structures of four compounds (7, 9, 11 and 12) that were firstly reported in
548
vine tea.
25
Table 1 Identified antioxidant flavonoids in vine tea. No.
tR (min)
λmax (nm)
[M+H]+ (m/z) (∆ ppm)
Molecular formula
Fragment ions (m/z) (intensity)
Identification
303.0502 (100), 167.0341 (25),
3-Dihydroxyquercetin
(neutral form) 1
2.716
292
321.0605 (‒2.8)
C15H12O8
153.0183 (12) 2
7.207
291
321.0601 (‒2.8)
C15H12O8
303.0504 (6), 153.0182 (77),
Dihydromyricetin
149.0237 (41), 139.0394 (100) 3
10.222
293
321.0602 (‒2.5)
C15H12O8
303.0504 (7), 153.0184 (71),
Iso-dihydromyricetin
149.0230 (43), 139.0393 (100)
4
14.127
265, 357
481.0960 (‒0.6)
C21H20O13
319.0450 (100)
Myricetin-3-O-β-D-glucoside
5
14.681
290
305.0656 (‒2.0)
C15H12O7
287.0553 (3), 153.0184 (82),
Dihydroquercetin
149.0236 (55), 123.0444 (100)
6
15.326
259, 354
465.1005 (‒0.2)
C21H20O12
319.0448 (100)
Myricitrin
7
17.604
267, 354
435.0909 (‒0.1)
C20H18O11
303.0500 (100)
Quercetin-3-O-β-D-xyloside
8
17.677
253, 374
319.0450 (‒1.3)
C15H10O8
291.0502 (5), 153.0185 (28),
Myricetin
26
139.0393 (100) 9
17.719
284
437.1443 (‒1.8)
C21H24O10
275.0912 (100), 169.0549 (22)
Phloridzin
10
17.811
257, 352
449.1059 (‒0.6)
C21H20O11
303.0501 (100)
Quercetin-3-O-α-L-rhamnoside
11
19.231
285
275.0914 (‒1.8)
C15H14O5
169.0550 (100)
Phloretin
12
19.586
263, 342
433.1125 (‒0.3)
C21H20O10
287.0547 (100)
Kaempferol-3-O-α-L-rhamnoside
13
19.688
254, 367
303.0500 (‒1.6)
C15H10O7
275.0552 (10), 153.0183 (20),
Quercetin
123.0444 (100) 14
21.150
284
303.0862 (‒2.3)
C16H14O6
288.0631 (100), 275.0914 (13),
Hesperitin
153.0182 (37)
15
23.953
265, 364
287.0549 (‒2.4)
C15H10O6
259.0602 (4), 153.0185 (34), 107.0495 (100)
27
Kaempferol
Fig. 1 250 20 150 10
100 c
0
I
b
II III IV
-10
0
5
10 15 Time (min)
20
28
50 0 -50
a
-100 25
Absorbance (mAU)
Absorbance (mAU)
200
Fig. 2 2
Absorbance (mAU)
150
7-10
100 5-6 50 a
3
1
12-13 11 14
4
15
0 -50 b 0
5
10 15 Time (min)
20
29
25
Fig. 3 ×105 A
Absorbance (254 nm)
2.0
1.5
1.0
0.5 a
0
b
c
100
400
170 min mobile phase change
150
B
d
200 250 Time (min)
e
300
350
2
Absorbance (mAU)
300 14
e 15
8
200
5
1
11 13
3
9 12 10
100 7 4 5
c b
6
a
0 0
d
10 15 Time (min)
20
30
25
Fig. 4 100
A
153.0182
80 Relative intensity/%
139.0394 1,2 + B0
1,3
OH
+
A0
OH
60
HO
O OH
40
149.0237 0,2 + A0 -H2O
OH
O +
[M+H] 321.0601 + [M+H-H2O] 303.0504
20
0 100
100
OH
B
150
200
m/z
250
300
139.0393 1,2 + B0 OH
80 Relative intensity/%
350
OH HO
O
OH
60 OH OH
O
40 20 0 100
153.0185 1,3 + A
150
+
[M+H] + [M+H-CO] 319.0450 291.0502
200
250
300
m/z
31
350
Fig. 5 R1
OH
OH HO
HO
O
OH
OR2
OH
OR
O
7 R1 =OH, R2 = β-D-xyl
O
9 R = β-D-Glc
12 R1 =H, R2 = α-L-rha 11 R = H
32
Highlights • Off-line coupling heart-cutting HSCCC with HPLC was developed. • Developed system with high orthogonality resolved similar structures. • Fifteen antioxidative flavonoids were identified by DAD and MS/MS. • Four minor flavonoids were reported in vine tea for the first time.
33