Antioxidative caffeoylquinic acids and flavonoids from Hemerocallis fulva flowers

Article abstract of DOI:10.1021/jf201166b

Tumor necrosis factor-α (TNF-α)-induced reactive oxygen species (ROS) production in HepG2 was used to screen hepatocyte protective compounds from the flowers of Hemerocallis fulva. Three new polyphenols, n-butyl 4-trans-O-caffeoylquinate (1), kaempferol 3-O-{α-l- rhamnopyranosyl(1→6)[α-l-rhamnopyranosyl(1→2)]} -β-d-galactopyranoside (2), and chrysoeriol 7-O-[β-d- glucuronopyranosyl(1→2)(2-O-trans-feruloyl)-β-d-glucuronopyranoside (3), together with four caffeoylquinic acid derivatives (4-7), eight known flavones (8-15), one naphthalene glycoside, stelladerol (16), one tryptophan derivative (17), adenosine (18), and guanosine (19) were isolated from the bioactive fractions of the aqueous ethanol extract of H. fulva flowers. The structures of isolated compounds were characterized by means of spectroscopic data. Compounds 1-3 were described as first isolated natural products. Among the above-mentioned compounds, the caffeoylquinic acid derivatives are the major components with potent free radical scavenging activity in HepG2 cells and are for the first time isolated from H. fulva flowers. A convenient ultraperformance liquid chromatography (UPLC) method was also developed to simultaneously separate and identify caffeoylquinic acids and flavonoids promptly.

Full text of DOI:10.1021/jf201166b

ARTICLE
pubs.acs.org/JAFC
Antioxidative Caffeoylquinic Acids and Flavonoids from
Hemerocallis fulva Flowers
Yun-Lian Lin,* Chung-Kuang Lu, Yeh-Jeng Huang, and Hong-Jhang Chen
National Research Institute of Chinese Medicine, Taipei 112, Taiwan
ABSTRACT: Tumor necrosis factor-R (TNF-R)-induced reactive oxygen species (ROS) production in HepG2 was used to screen
hepatocyte protective compounds from the flowers of Hemerocallis fulva. Three new polyphenols, n-butyl 4-trans-O-caffeoylquinate
(
7
1), kaempferol 3-O-{R-L-rhamnopyranosyl(1f6)[R-L-rhamnopyranosyl(1f2)]}-β-D-galactopyranoside (2), and chrysoeriol
-O-[β-D-glucuronopyranosyl(1f2)(2-O-trans-feruloyl)-β-D-glucuronopyranoside (3), together with four caffeoylquinic acid
derivatives (4ꢀ7), eight known flavones (8ꢀ15), one naphthalene glycoside, stelladerol (16), one tryptophan derivative (17),
adenosine (18), and guanosine (19) were isolated from the bioactive fractions of the aqueous ethanol extract of H. fulva flowers. The
structures of isolated compounds were characterized by means of spectroscopic data. Compounds 1ꢀ3 were described as first
isolated natural products. Among the above-mentioned compounds, the caffeoylquinic acid derivatives are the major components
with potent free radical scavenging activity in HepG2 cells and are for the first time isolated from H. fulva flowers. A convenient
ultraperformance liquid chromatography (UPLC) method was also developed to simultaneously separate and identify caffeoyl-
quinic acids and flavonoids promptly.
KEYWORDS: Hemerocallis fulva, flowers, caffeoylquinic acids, flavonoids, TNF-R, free radical scavenger, UPLC
’
INTRODUCTION
polarimeter (Hachioji, Tokyo, Japan). Ultraviolet (UV) spectra were
measured on a Hitachi U-3310 spectrophotometer (Hitachi, Tokyo,
Japan). Nuclear magnetic resonance (NMR) spectra were run in
Excess production of reactive oxygen species (ROS) and/or
defective cellular antioxidant systems have been implicated in
many pathological conditions including chronic liver injury and
fibrogenesis.
through the production of ROS during chronic liver damage.
Tumor necrosis factor-R (TNF-R), a pro-inflammatory cyto-
kine, plays a central role in hepatocyte cell death during liver
injury. Multiple studies have shown that TNF-R signaling is
associated with enhanced generation of ROS, which significantly
contributes to TNF-R-induced cell death. Phenols such as
lignans and flavonoids derived from fruits and vegetables have
been regarded as hepatoprotective agents.
Hemerocallis fulva L. (daylily) belongs to the Liliaceae family.
Both fresh and dried flowers of the plant have been commonly
used as a vegetable in eastern Asia. Different kinds of compounds
CD
3 6
OD or DMSO-d on a Varian Unity INOVA-500 or VNMRS 600
(Varian, Palo Alto, CA). ESIMS and HRESIMS mass spectra were
1
ꢀ3
Many inflammatory cytokines were expressed3
recorded on a Finnigan MAT LCQ and Finnigan Mat 95S, respectively.
UPLC was performed on a Waters AcQuity Ultra Performance LC
(
Milford, MA). Silica gel (230ꢀ400 mesh) and a semipreparative Si
column (LiChrosorb Si-60) (Merck, Darmstadt, Germany) and Sepha-
dex LH-20 (Amersham Biosciences, Uppsala, Sweden) were used for
column chromatography. Solvents (analytical grade) were purchased
from Merck. Dulbecco's modified Eagle's medium (DMEM) and fetal
bovine serum (FBS) were from Gibco BRL (Grand Isand, NY). Di-
methyl sulfoxide (DMSO) and analytical grade solvents were from
Merck. Trypan blue, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazo-
lium bromide (MTT), and TNF-R were purchased from Sigma (St. Louis,
4
5
,6
7
ꢀ9
10
MO). CM-H DCF-DA was from Invitrogen (Carlsbad, CA). All other
2
1
1,12
13
14
including lactams,
aquinones,
carotenoids, steroidal saponins, anthr-
chemicals were of analytical grade and purchased from commercial suppliers.
Determination of Total Phenolic Content. The total phenolic
1
5,16
17
and flavones have been isolated from the aerial
1
8
parts or roots or flowers of Hemerocallis spp. As a part of our
interest in liver protective constituents in edible flowers, further
investigation on the chemical and biological studies of H. fulva
flowers was conducted. This paper reports on the isolation and
structural elucidation of three new compounds from H. fulva
flowers and the free radical scavenging activity of the isolates in
HepG2 cells. Meanwhile, a convenient ultraperformace liquid
chromatography (UPLC) method was developed to simulta-
neously separate and identify caffeoylquinic acids and flavonoids.
content was determined according to the FolinꢀCiocalteu method.
Briefly, 200 μL of each sample (2000 μg/mL) was separately mixed with
2 mL of distilled water, 1 mL of FolinꢀCiocalteu reagent, and 5 mL of
20% sodium carbonate. After incubation in the dark for 20 min at room
temperature, the absorbance was measured at 735 nm with a microplate
reader (Molecular Devices, Sunnyvale, CA). The amount of Folinꢀ
Ciocalteu reagent was substituted by the same amount of distilled water
in the blank. Gallic acid (0, 200, 400, 600, 800, and 1000 μg/mL) was
used to make a calibration curve. The total phenolic content was
expressed as milligram gallic acid equivalents (GAE) per gram of extract.
’
MATERIALS AND METHODS
Received: March 22, 2011
General Experimental Procedures. Infrared (IR) spectra were
Revised:
May 31, 2011
recorded on a Nicolet Avatar 320 FT-IR spectrophotometer (Thermo
Electron, Akron, OH). Optical rotation was measured on a Jasco P-2000
Accepted: July 15, 2011
Published: July 15, 2011
r 2011 American Chemical Society
8789
dx.doi.org/10.1021/jf201166b
|J. Agric. Food Chem. 2011, 59, 8789–8795

Journal of Agricultural and Food Chemistry
ARTICLE
Determination of Total Flavonoid Content. The AlCl
3
meth-
dd, J = 9.6, 2.4 Hz, H-4), 6.35 and 7.62 (1H each, d, J = 16.0 Hz, H-R,
18
00 00
H-β), 6.78 (1H, d, J = 7.8 Hz, H-5 ), 6.94 (1H, dd, J = 7.8, 2.0 Hz, H-6 ),
od was used to determine the total flavonoid content. Briefly, 0.5 mL of
each sample (2000 μg/mL) was separately mixed with 0.1 mL of 10%
aluminum chloride, 0.1 mL of 1 M potassium acetate, and 2.8 mL of
distilled water. After incubation at room temperature for 30 min, the
absorbance was measured at 415 nm with a microplate reader
Molecular Devices). The amount of 10% aluminum chloride was
substituted by the same amount of distilled water in the blank. Catechin
0, 100, 200, 400, 600, and 800 μg/mL) was used to make a calibration
0
0
13
3
7.06 (1H, d, J = 2.0 Hz, H-2 ); C NMR (125 MHz, CD OD), δ 14.0
0
0
0
(q, C-4 ), 20.0 (t, C-3 ), 31.6 (t, C-2 ), 38.3 (t, C-6), 42.3 (t, C-2), 65.6
(d, C-3), 66.4 (t, C-1 ), 69.2 (d, C-5), 76.5 (s, C-1), 78.8 (d, C-4), 115.2
(d, C-R), 115.2 (d, C-2 ), 116.5 (d, C-5 ), 123.0 (d, C-6 ), 127.8 (s,
C-1 ), 146.7 (s, C-3 ), 147.1 (d, C-β), 149.4 (d, C-4 ), 169.0 (s, C-9 ),
0
0
0
00
00
0
0
00
00
00
(
0
175.3 (s, C-7) Key HMBC correlations: H-1 , H-2, and H-6/C-7; H-4,
0
0
ꢀ
(
H-R, and H-β/C-9 . ESIMS, m/z 409 [M ꢀ H] . HRESIMS, 409.1536
+
curve. The total flavonoid content was expressed as milligram catechin
equivalents (CE) per gram of extract.
25 9
[M + Na] (calcd 409.1499 for C20H O ).
Kaempferol 3-O-{R-L-rhamnopyranosyl(1f6)-[R-L-rhamnopyranosyl-
(1f2)]}-β-D-galactopyranoside (2): yellow amorphous; IR (KBr) νmax
Evaluation of Reactive Oxygen Species in HepG2 Cells.
ROS in HepG2 cells were measured by the conversion of nonfluorescent
ꢀ
1
(cm ), 3410, 1667, 1608, 1505, 1069; UV (MeOH) λmax nm (log ε),
1
0
0
0 0
2
(
,7 -dichlorofluorescein diacetate (DCF-DA) into 2 ,7 dichlorofluorescein
265 (3.96), 348 (3.89); H NMR (500 MHz, DMSO-d
6
), δ 0.79 and
0
000
000
DCF). PBS-rinsed cells were treated with TNF-R (200 ng/mL) with or
without extract or isolated compound for 2 h and then incubated with 8
0.94 (3H each, d, J = 6.0 Hz, rhamnosyl H-6 and 6 ), 4.30 and 5.04
0000 000
(1H each, br s, anomeric H-1 and -1 ), 5.47 (1H, d, J = 7.5 Hz,
00
anomeric H-1 ), 6.18 and 6.38 (1H each, d, J = 2.0 Hz, H-6 and -8), 6.86
μM CM-H DCFDA for 30 min at 37 °C. After PBS washing, fluoro-
2
0 0 0 0
and 7.93 (2H each, d, J = 8.5 Hz, H-(3 (5 )) and -2 (6 )), 10.10, 10.80,
metric analysis was performed at the indicated time points with the
excitation and emission wavelengths at 485 and 530 nm, respectively, in
0
13
and 12.63 (1H each, br s, OH-4 , OH-7, OH-5); C NMR (125 MHz,
19
0000 000
a FLEX Station (Molecular Devices Corp.).
DMSO-d ), δ 17.3 (q, C-6 ), 17.7 (q, C-6 ), 66.9 (t, galactopyranosyl
6
00 000
C-6 ), 68.2 (d, rhamnopyranosyl, C-5 ), 68.3 (d, rhamnopyranosyl,
Cell Viability Assay. HepG2 cells were cultured in DMEM
medium (containing 10% FBS, pH 7.2ꢀ7.4). During 24 h of incubation,
cells in serum-free medium were exposed to TNF-R in the absence or
presence of extract or isolated pure compounds at the indicated
concentrations for 6 h by MTT assay. The cell viability of various test
groups was determined by the following equation: (absorbance of the
test group/absorbance of the control) ꢁ 100%. All test samples
mentioned above were dissolved in DMSO. The final concentration
of DMSO was <0.1% and the group of 0.1% DMSO was used as the
0
000
000 0000
C-5 ), 70.3 (d, rhamnopyranosyl C-2 (2 )), 70.5 (d, rhamnopyr-
anosyl C-3 ), 70.6 (d, rhamnopyranosyl, C-3 ), 70.6 (d, galactopyr-
anosyl C-4 ), 71.8 (d, rhamnopyranosy C-4 (4 )), 75.7 (d,
galactopyranosyl C-3 ), 77.1 (d, galactopyranosyl C-2 ), 77.2 (d,
galactopyranosyl C-5 ), 93.5 (d, C-8), 98.6 (d, C-6), 98.7 (d, galacto-
pyranosyl C-1 ), 100.6 (d, rhamnosyl, C-1 ), 100.8 (d, rhamnosyl,
C-1 ), 104 (s, C-10), 115.0 (d, C-3 (5 )), 120.9 (s, C-1 ), 130.7 (d,
C-2 (6 )), 132.6 (s, C-3), 156.4 (s, C-9), 156.9 (s, C-2), 159.8 (C-4 ),
161.2 (s, C-5), 164.0 (s, C-7), 177.2 (s, C-4). Key HMBC correlations:
0
000
000
0
0
000
0000
0
0
00
0
0
0
0
000
0
0
000
0
0
0
0
0
20
control group.
0
Isolation and Purification Process. The air-dried flowers of H.
fulva (10 kg) were purchased from Taitung, Taiwan, in 2009. The dried
flowers (10 kg) were soaked with aqueous ethanol (80% EtOH) at 60 °C
overnight (3 ꢁ 80 L). The combined extracts were evaporated under
reduced pressure in a rotary vacuum evaporator. The dry extract (HF,
H-6, H-8, 7-OH (δ 10.80)/C-7; 5-OH (δ 12.63)/C-5, C-6, C-10; 4 -OH
0
0
000
0
0
0
00
0000
00
00
000
00
(δ 10.10), H-2 (6 ), H-3 (5 )/C-4 ; H-1 /C-3, C-2 , C-3 ; H-1 /C-2 ,
0
00
000
0000
00
0000
0000
000
000
C-2 , C-3 , C-5 ; H-1 /C-6 , C-2 , C-3 , C-5 ; H-6 /C-4 ,
0
00
0000
0000
0000
ꢀ
C-5 ; H-6 /C-4 , C-5 . ESIMS: m/z 739 [M ꢀ H] .
Chrysoeriol 7-O-[(2-O-trans-feruloylglucuronopyranosyl) (1f2)]-
O-β-D-glucuronopyranoside (3): yellow amorphous powder; IR
9
52 g) was sucessively partitioned with ethyl acetate (EtOAc) and n-
butanol (BuOH) to give EtOAc- (HF-EtOAc, 353 g), BuOH- (HF-
BuOH, 136 g), and water-soluble fractions (HF-H O, 450 g). The
ꢀ
1
(KBr) νmax (cm ), 3358, 1716, 1657, 1605, 1593, 1514, 1262, 1187,
1159, 1088, 1056; UV (MeOH) λmax nm (log ε), 248 (3.95), 270 (3.88),
334 (4.00); (+NaOAc) λmax nm (log ε), 249 (3.95), 270 (3.88), 335
(4.00); (+AlCl ) λ nm (log ε), 277 (3.95), 296 (3.88), 338 (3.97),
2
EtOAc-soluble fraction was then subjected to silica gel (230ꢀ400 mesh)
column chromatography with successive elution by a Hex/EtOA/
MeOH gradient solvent system to obtain six fractions: A (57 g, Hex),
B (112 g, 25% EtOAc/Hex), C (79 g, 50% EtOAc/Hex), D (27 g, 75%
EtOAc/Hex), E (24 g, EtOAc), and F (48 g, MeOH). Fractions D and E
were separated on a Sephadex LH-20 column (MeOH) repeatedly to
3
max
387 (3.85); (+AlCl +HCl) λ nm (log ε), 278 (3.95), 296 (3.88), 337
3
max
1
(3.98), 387 (3.85); H NMR (500 MHz, DMSO-d ), δ 3.54 (1H, m,
6
0
0
0
00
000
H-2 ), 4.63 (1H, t, J = 8.5 Hz, H-2 ), 4.90 (1H, d, J = 8.5 Hz, anomeric
00
H-1 ), 5.37 (1H, d, J = 7.5 Hz, anomeric H-1 ), 6.39 and 7.49 (1H each,
d, J = 15.5 Hz, feruloyl H-R and H-β), 6.41 (1H, d, J = 2.0 Hz, H-6), 6.76
(1H, d, J = 8.0 Hz, feruloyl H-5), 6.78 (1H, d, J = 2.0 Hz, H-8), 6.93 (1H,
2
give 1 (38 mg), 4 (0.95 g), 8 (32 mg), and 9 (35 mg). The H O-soluble
fraction was chromatogramed over a Diaion HP-20 column and eluted
with a gradient of MeOH in H O. Fraction 75% MeOH/H O eluate was
0
2
2
d, J = 8.5 Hz, H-5 ), 6.94 (1H, s, H-3), 7.06 (1H, dd, J = 8.0, 2.0 Hz,
further purified on a Sephadex LH-20 column (80% MeOH/H O) to
feruloyl H-6), 7.24 (1H, d, J = 2.0 Hz, feruloyl H-2), 7.56 (1H, dd, J = 8.5,
2
0 0
2.0 Hz, H-6 ), 7.57 (1H, d, J = 2.0 Hz, H-2 ), 9.55 (1H, br s, feruloyl
yield 18 (252 mg) and 19 (123 mg). The MeOH eluates of the H
2
O-
0
13
soluble fraction (18 g) and HF-EtOAc-F (52 g) were combined with the
BuOH-soluble fraction and further separated on Sephadex LH-20
4-OH), 9.99 (1H, br s, 4 -OH), 12.96 (1H, br s, 5-OH); C NMR (125
MHz, DMSO-d ), δ 55.7 (q, OCH ), 56.0 (q, OCH ), 71.4 (d,
6
3
3
0
0
0
0
0
(
MeOH) and RP-18 (preparative and semipreparative) columns re-
peatedly to give 2 (127 mg), 3 (36 mg), 5 (1.79 g), 6 (64 mg), 7 (85 mg),
0 (78 mg), 11 (141 mg), 12 (108 mg), 13 (43 mg), 14 (34 mg), 15 (18
glucuronopyranosyl-C-4 ), 71.9 (d, glucuronopyranosyl-C-4 ), 73.4
0
0
00
(d, glucuronopyranosyl-C-2 ), 73.9 (d, glucuronopyranosyl-C-3 ), 74.9
0
0
0
0
0
1
(d, glucuronopyranosyl-C-5 ), 75.0 (d, glucuronopyranosyl-C-3 ),
000
75.6 (d, glucuronopyranosyl-C-5 ), 79.9 (d, glucuronopyranosyl-C-
mg), 16 (25 mg), and 17 (34 mg).
0
00
00
n-Butyl 4-trans-O-caffeoylquinate (1): colorless amorphous powder;
2 ), 94.9 (d, C-8), 97.2 (glucuronopyranosyl-C-1 ), 99.2 (d, C-6),
100.7 (d, glucuronopyranosyl-C-1 ), 103.5 (s, C-3), 105.5 (s, C-10),
110.4 (d, C-2 ). 111.2 (d, feruloyl-C-2), 115.0 (d, feruloyl-C-R), 115.5
(d, feruloyl-C-5), 115.8 (d, C-5 ), 120.5 (d, C-6 ), 121.5 (s, feruloyl
C-1), 122.9 (d, feruloyl C-6), 125.8 (s, C-1 ), 144.6 (d, feruloyl C-β),
147.9 (s, C-3 ), 148.1 (s, feruloyl C-3 ), 149.2 (s, feruloyl C-4), 150.9 (s,
2
5
ꢀ1
000
[
1
2
R]
D
= ꢀ8.08° (MeOH, c 0.15); IR (KBr) νmax (cm ), 3406, 1718,
0
686, 1627, 1508, 1441, 1069, 1049, 934; UV (MeOH) λ , nm (log ε),
20 (3.82), 233 (3.78), 243 (3.78), 295 (3.89), 329 (3.98); H NMR
500 MHz, CD
max
1
0
0
0
0
(
3
0
OD), δ 0.93 (3H, t, J = 7.2 Hz, H-4 ), 1.41 (2H, hepta,
0
0
0
J = 7.2 Hz, H-3 ), 1.64 (2H, hepta, J = 7.2 Hz, H-2 ), 2.01 (1H, dd, J =
0
1
6
3.2, 10.2 Hz, H-6a), 2.06 (1H, m, H-2a), 2.16ꢀ2.20 (2H, m, H-2b, -
C-4 ), 156.9 (s, C-9), 161.2 (s, C-5), 162.2 (s, C-7), 164.2 (s, C-2), 165.8
0
00
b), 4.13 (2H, hepta, J = 7.2 Hz, H-1 ), 4.28 (1H, ddd, J = 10.2, 9.6, 4.8
(s, feruloyl CdO), 169.9 (s, glucuronopyranosyl C-6 ), 170.1 (s,
000
Hz, w1/2 = 20.0 Hz, H-3), 4.31 (1H, m, Jw1/2 = 7.6 Hz, H-5), 4.83 (1H,
glucuronopyranosyl C-6 ), 182.1 (s, C-4). Key HMBC correlations:
8
790
dx.doi.org/10.1021/jf201166b |J. Agric. Food Chem. 2011, 59, 8789–8795

Journal of Agricultural and Food Chemistry
ARTICLE
Table 1. Phenol and Flavonoid Contents of the Aqueous
Ethanol Extract and the Subtractions of H. fulva
EtOAc), 0.56 (HF-BuOH) and 0.51 (HF-H O), respectively.
The results suggest that both flavonoids and phenols play major
roles in ROS scavenging in the active fractions.
Compound Isolation and Structural Identification. Nine-
teen compounds including three new compounds (1ꢀ3),
together with four known caffeoylquinic acid derivatives, chloro-
2
total phenols
(mg gallic acid equiv/g
extracts)
total flavonoids
(mg catechin equiv/g
extracts)
extract of H. fulva
flowers
2
6,27
26
genic acid methyl ester (4),
chlorogenic acid (5), 4-caf-
HF
30.9 ( 0.7
25.3 ( 2.7
34.6 ( 2.7
28.8 ( 0.7
15.1 ( 0.9
10.4 ( 2.3
19.5 ( 1.3
14.6 ( 0.5
28
28
feoylquinic acid (6), and 5-caffeoylquinic acid (7), eight
known flavones, including kaempferol (8), quercetin (9),
astragalin (10), isoquercitrin (11), kaempferol 3-O-rutino-
HF-EA
HF-BuOH
26
26
26
26
HF-H
2
O
29
29,30
side(12), rutin (13),
quercetin 3-O-{R-L-rhamnopyranosyl-
(
(
1f6)[R-L-rhamnopyranosyl(1f2)]}-β-D-galactopyranoside
31
14), chrysoeriol 7-O-β-D-glucuronopyranosyl(1f2)-O-β-D-
H-3/C-2, C-4; 5-OH/C-5, C-6 and C-10; H-8/C-6, C-7, C-9 and
28
0
00
00
000
00
00
000
00
C-10; H-2 /feruloyl CdO and C-1 ; H-2 /C-1 , C-1 , C-3 ,
glucuronopyranoside (15), a naphthalene glycoside, stelladerol
16 32
00
000
000
C-4 ; H-1 /C-7; H-1 /C-2 . Key NOE correlations: OCH
3
(δ
H
(16), a tryptophan derivative, lycoperodine-1 (17), adeno-
26
0
3
.73)/ feruloyl H-2; OCH
3
ꢀ
(δ
H
3.79)/H-2 . ESIMS: m/z 827 [M ꢀ
sine (18), and guanosine (19) were isolated from high phenolic
subfractions of the aqueous ethanol extract of H. fulva flowers
(Figure 1). The structures of known compounds were elucidated
by spectral methods and their spectroscopic data compared with
those of an authentic sample or with published data. Compound
4 is probably an artifact during the isolation.
ꢀ
ꢀ
H] , 651 [M ꢀ H ꢀ 175] , 527 [M ꢀ H ꢀ 300] , 299. HRESIMS:
+
8
27.1831[M ꢀ H] (calcd 827.1670 for C H O ).
3
8 35 21
Hydrolysisof2 and3: Compound 2(10mg) was dissolvedin5%HCl/
aqueous ethanol (2 mL) at 80 °C for 30 min. After neutralization with
NaHCO , the mixture was evaporated. The residue was resuspended in
3
H O and then filtered to yield kaempferol (3 mg). The sugar components
2
Compound 1 was obtained as a colorless amorphous powder,
25
in the filtrate were identified by TLC (on Si gel, developed with ethyl
acetate/butanone/formic acid/H O = 5:3:1:1) as D-galactose (R 0.20)
2 f
[
R] = ꢀ8.08° (MeOH, c 0.15), and gave a molecular ion peak
D
ꢀ
at m/z 409 [M ꢀ H] in agreement with a molecular formula of
and L- rhamnose (R 0.55) in comparison with authentic samples.
f
C H O by HRESIMS and DEPT NMR. The IR spectrum
2
0 25 9
Compound 3 (5 mg) was conducted by basic hydrolysis (5% NaOH in
ꢀ1
showed absorption bands for hydroxyl (3406 cm ), ester
1718 cm ), conjugated carbonyl (1686 cm ), and an aro-
matic group (1627 and 1508 cm ). H and C NMR indicated
5
0% aqueous ethanol at 80 °C) for 1 h to give compound 15 and
ꢀ1
ꢀ1
(
ferulic acid.
ꢀ1
1 13
UPLC Analysis of Isolated Caffeoylquinic Acids and Fla-
vonoids. UPLC analyses were carried out on a Waters AcQuity Ultra
Performance LC, which was equipped with a UPLC column
signals for a butyroyl [δ 0.93 (3H, t, J = 7.2 Hz), 1.41 and 1.64
H
and 4.13 (2H each, hepta, J = 7.2 Hz); δ 14.0 (q), 20.0 (t), 31.6
C
(
1
t), and 66.4 (t)], two methylenes [δ 2.01 (1H, dd, J = 13.2,
H
(
ACQUITY BEH C18, 1.7 μm, 2.1 ꢁ 100 mm) and a guard column
0.2 Hz), 2.06 (1H, m), and 2.16ꢀ2.20 (2H, m); δ 38.3 (t, C-6)
H
C
(VanGuard BEH C18, 1.7 μm, 2.1 ꢁ 5 mm) at a constant flow rate of
and 42.3 (t, C-2)], three oxymethines [δ 4.28 (1H, ddd, J =
0
.4 mL/min. Mobile phases A (acetonitrile) and B (0.1% formic acid)
1
0.2, 9.6, 4.8 Hz), 4.31 (1H, m), 4.83 (1H, dd, J = 9.6, 2.4 Hz); δC
were run on a programmed protocol: 0ꢀ3 min, 2ꢀ18% A; 3ꢀ5 min,
6
5.6 (d, C-3), 69.2 (d, C-5), and 78.8 (d, C-4)], one trans-
1
8ꢀ18% A; 5ꢀ7 min, 18ꢀ40% A; 7ꢀ9 min, 40ꢀ70% A. The DAD was
caffeoyl [δ 6.35 and 7.62 (1H each, d, J = 16.0 Hz,), 6.78 (1H, d,
H
set at a wavelength between 200 and 400 nm. The chromatographs were
plotted at an absorbance of 254 nm. An aliquot of sample (1 μL) was
directly introduced into the column through the sample manager.
Statistical Analysis. Data are expressed as the mean ( SEM. One-
way analysis of variance (ANOVA) was used. The statistical significance
between control and compound-treated group was evaluated by Stu-
dent’s t test. A p < 0.05 was considered to be statistically significant.
J = 7.8 Hz), 6.94 (1H, dd, J = 7.8, 2.0 Hz), and 7.06 (1H, d, J = 2.0
Hz); δ 115.2 (d), 115.2 (d), 116.5 (d), 123.0 (d), 127.8 (d),
C
1
46.7 (s), 147.1 (d), 149.4 (s), and 169.0 (s)], a quaternary
oxygenated carbon [δ 76.5 (s)], and a carbonyl [δ 175.3 (s)].
C
C
From the data mentioned above, compound 1 was suggested as
the n-butyl ester of 4-caffeoylquinic acid. The coupling constants
of H and H (J = 9.6 Hz) with a large half-width value (J =
3
4
w1/2
2
0.0 Hz), and a small one between H and H (J = 2.4 Hz) in the
4 5
’
RESULTS AND DISCUSSION
quinate moiety indicated they had trans-diaxial protons and
33
Phenol and Flavonoid Compositions. Phenolic compounds
axialꢀequatorial gauche configuration, respectively. The acyla-
0
are ubiquitously distributed in the plant kingdom. Phenolic
compounds such as caffeoylquinic acid derivatives and flavonoids
are the most abundant polyphenols in human diet and have been
shown to possess different kinds of bioactivities.
phenolic contents and flavonoids were shown to exhibit anti-
oxidant capacity and redox properties, such as reducing agents,
hydrogen donors, and reactive oxidative species scavengers. As
Table 1 shows, the total phenolic contents of HF, HF-EA, HF-
BuOH, and H O were 30.9 ( 0.7, 25.3 ( 2.7, 34.6 ( 2.7, and
tion of hydroxyl groups produced the downfield shifts of H-1
[δ 4.13 (2H, hepta, J = 7.2 Hz)] and H-4 [δ 4.83 (1H, dd, J =
9.6, 2.4 Hz)], which displayed the presence of acyloxy groups at
both positions. This assignment was further confirmed by
HMBC correlations from H-1 , H-2, and H-6 to C-7 and from
H
H
21ꢀ24
Levels of
0
0
0
H-4, H-R, and H-β to C-9 .
25
Compound 2 showed the quasi molecular ion peak at m/z 739
ꢀ
1
1
13
[M ꢀ H] . H and C NMR signals were assigned by 1D-
1
TOCSY, Hꢀ H-COSY, HMQC, and HMBC NMR experi-
ments and suggested kaempferol, a galactopyranosyl, and two
rhamnopranosyl units were included in the skeleton. The upfield
2
2
8.8 ( 0.7 mg gallic acid equiv (GAE)/g of extract, respectively.
The flavonoid contents of three subfractions were 10.4 ( 2.3,
1
9.5 ( 1.3, and 14.6 ( 0.5 mg catechin equiv (CE)/g of
shift (about 1.8 ppm) of C-3 (δ 132.6) and a lower field shift
(about 9.7 ppm) of C-2 (δ 156.9) in the kaempferol moiety
(compared to kaempferol (8)) indicated O-glycosidation at
C-3. The coupling constants of the anomeric proton signals
C
extracts), respectively. Only the HF-BuOH subfraction has a
flavnoid content higher than that in crude extract (15.1 ( 0.9 mg
CE/g of extracts). Total flavonoid/total phenolic contents in the
crude extract and three subfractions were 0.49 (HF), 0.41 (HF-
C
21
at δ 5.47 (1H, d, J = 7.5 Hz), 4.30, and 5.04 (1H each, br s), and
H
8
791
dx.doi.org/10.1021/jf201166b |J. Agric. Food Chem. 2011, 59, 8789–8795

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 1. Structures of the isolated compounds.
two methyl group at δ 0.79 and 0.94 (3H each, d, J = 6.0 Hz)
H
revealed 2 with one β-D-galactosyl and two R-L-rhamnosyl units.
Acidic hydrolysis of 2 yielded kaempferol, D-galactose, and L-
rhamnose. The downfield shift of C-2 (δ 77.1) and C-6 (δ
C
C
6
6.9) of the galactopyranosyl indicated the interglycosidic
linkage at both positions. The assignment was further supported
by J and J correlations between H-1 (δ 5.47) and C-3 (δ
C
0
0
2
3
H
000 00
32.6), between H-1 (δH 5.04) and C-2 (δ 77.1), and
C
1 13
1
0
000
00
between H-1 (δ 4.30) and C-6 (δ 66.9). The H and
C
H
C
NMR spectral data of 2 were similar to those of 14 except for
signals assignable to the aglycone moiety in place of kaempferol
in 2. The structure was therefore assigned as kaempferol 3-O-
{
R-L-rhamnopyranosyl(1f6)-[R-L-rhamnopyranosyl(1f2)]}-β-D-
galactopyranoside.
The molecular formula of compound 3 was determined to be
1
3
C H O on the basis of C and DEPT spectra and the
3
8 36 21
+
HRESIMS molecular ion at m/z 827.1831 [M ꢀ H] . The UV
characteristic absorption bands at 232, 270, and 333 nm indicated
34
compound 2 to be a flavone derivative. The presence of a signal
at δ 12.96 and a bathochromic shift of 30 nm of band II (248 nm
H
shifts to 278 nm) in the presence of AlCl /HCl indicated that C-5
3
3
4
Figure 2. Effects of the ethanol extract of H. fulva on TNF-R-induced
was hydroxylated. Meanwhile, the absence of a 7-hydroxyl group
was elucidated by no bathochromic shift on adding shift reagent
4
HepG2 cell ROS production. HepG2 cells were seeded at 8 ꢁ10 cells/
3
4
1
13
well in 96 wells overnight. PBS-rinsed cells were treated with TNF-R
NaOAc. The H, C, and DEPT spectra revealed the presence
of chrysoeriol, two glucuronopyranosyl, and trans-feruloyl moi-
eties. Alkali hydrolysis gave compound 15 and ferulic acid. Long-
(10 ng/mL) with or without HF extract for 2 h and then incubated with
8
μM of CM-H DCFDA for 30 min at 37 °C. The fluorometric analysis
2
was measured with the excitation and emission wavelengths at 485 and
0
00
range correlations between H-2 [δ 4.63 (t, J = 8.5 Hz)] and
H
5
30 nm. Each column represents the mean ( SEM of three experiments.
0
00
carbonyl of feruloyl (δ 165.8), between H-1 [δ 4.90 (d, J =
C
H
Resveratrol was used as a positive control. (#) P < 0.05, compared with
00 00
.5 Hz)] and C-2 (δ 79.9), and between H-1 [δ 5.37 (d, J =
C H
8
7
control; (/) P < 0.05, compared with TNF-R-induced group.
.5 Hz) and C-7 (δ 162.2) elucidated their connectivities. The
C
structure was further confirmed by the NOE correlations between
methoxyl at δ 3.73 and feruloyl H-2 at δ 7.24 and between
Effects of Subfractions and Isolated Polyphenols on ROS
scavenging in HepG2 Cells. Flavonoids, phenethyl glycosides,
H
H
0
methoxyl at δ 3.79 and H-2 at δ 7.57 (d, J = 2.0 Hz).
H
H
8
792
dx.doi.org/10.1021/jf201166b |J. Agric. Food Chem. 2011, 59, 8789–8795

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 3. Effects of the isolated polyphenols and flavonoids from H. fulva on (A) ROS production and (B) cell viability in TNF-R-induced Hep G2 cells.
A) PBS-rinsed cells were treated with TNF-R (10 ng/mL) with or without test compound for 2 h AND then incubated with 8 μM CM-H DCFDA for
(
2
3
0 min at 37 °C. The fluorometric analysis was measured with the excitation and emission wavelengths at 485 and 530 nm. (B) cell viability in TNF-R-
induced HepG2 cells was tested by MTT assay. Each column represents the mean ( SEM of three experiments. Resveratrol was used as a positive
control. (#) P < 0.05, compared with control group; (/) P < 0.05, compared with TNF-R-induced group.
naphthalene glycosides, and carotenoids have been isolated and
crude extract of the flowers of H. fulva and three subfractions
significantly inhibited TNF-R-induced ROS production in
HepG2 cells at the concentration of 50 μg/mL and had no
direct cytotoxicity (data not shown). The HF-BuOH subfrac-
tion preserved the most potent activity and diminished ROS to
a level similar to that of the control group. The isolated
caffeoylquinic acids and flavonoids (1ꢀ17) were also tested
for ROS scavenging activity (Figure 3A). Among those com-
pounds, caffeoylquinic acid derivatives 1 and 4ꢀ7, flavonoids 3
and 10ꢀ14, and anthraquinone 16 significantly inhibited
claimed as the antioxidative principles of daylily (Hemerocallis
1
3,16
17
spp.) flowers
or leaves. In this study, caffeoylquinate deri-
vatives including n-butyl 4-trans-O-caffeoylquinate (1, 38 mg),
methyl chloriginate (4, 0.95 g), chlorogenic acid (5, 1.79 g),
4
(
-caffeoylquinic acid (6, 64 mg), and 5-caffeoylquinic acid
7, 85 mg) were isolated as the major constituents in H. fulva
flowers. Chlorogenic acid and the related caffeoylquinic acids
from natural resources have been focused on various bioactiv-
2
2ꢀ24
ities including antioxidative effect.
As Figure 2 shows, the
8
793
dx.doi.org/10.1021/jf201166b |J. Agric. Food Chem. 2011, 59, 8789–8795

Journal of Agricultural and Food Chemistry
ARTICLE
’
AUTHOR INFORMATION
Corresponding Author
*
Phone: 886-2-2820-1999, ext. 6531. Fax: 886-2-2826-4266.
E-mail: yllin@nricm.edu.tw.
’
ABBREVIATIONS USED
HF, aqueous ethanolic extract of the flowers of H. fulva; EtOAc,
ethyl acetate; BuOH, n-butanol; HF-EtOAc, ethyl acetate-solu-
ble subfraction of HF; HF-BuOH, n-butanol-soluble subfrac-
tion of HF; HF-H O, H O-soluble subfraction of HF; Hex,
2
2
hexane; MeOH, methanol; GAE, gallic acid equivalents; CE,
catechin equivalents; IR, infrared spectra; UV, ultraviolet spectra;
NMR, nuclear magnetic resonance; ESIMS, electrospray ioniza-
tion mass spectra; UPLC, ultraperformance liquid chromatogra-
phy; DMSO, dimethyl sulfoxide; DMEM, Dulbecco's modified
Eagle's medium; FBS, fetal bovine serum; TNF-R, tumor necro-
sis factor-R; MTT, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-
tetrazolium bromide; ROS, reactive oxygen species; DCF-DA,
0
0
2
,7 -dichlorofluorescein diacetate; ANOVA, one-way analysis of
variance.
’
REFERENCES
Figure 4. UPLC chromatogram of (A) isolated caffeoylquinic acids and
flavonoids and (B) crude aqueous ethanol extract. (1) n-Butyl 4-caf-
feoylquinate; (2) kaempferol 3-O-{R-L-rhamnopyaranosyl(1f6)-[R-L-
rhamnopyranosyl(1f2)]}-β-D-galactopyranoside; (3) chrysoeriol 7-O-
(
1) Droge, W. Free radical in the physiological control of cell
function. Physiol. Rev. 2002, 82, 47–95.
2) Friedman, S. L. Molecular regulation of hepatic fibrosis, an
integrated cellular response to tissue injury. J. Biol. Chem. 2000,
75, 2247–2250.
3) Baroni, G. S.; D’Ambrosio, L.; Ferretti, G.; Casini, A.; Sario,
(
[(2-O-trans-feruloylglucuronopyranosyl) (1f2)]-O-β-D-glucuronopyr-
2
anoside; (4) methyl chlorogenate; (5) chlorogenic acid; (6) 4-caffeoyl-
quinic acid; (7) 5-caffeoylquinic acid; (8) kaempferol; (9) quercetin;
(
A. D.; Salzano, R.; Ridolfi, F.; Saccomanno, S.; Jezequel, A. M.;
Benedetti, A. Fibrogenic effect of oxidative stress on rat hepatic stellate
cells. Hepatology 1998, 27, 720–726.
(
10) astragalin; (11) isoquercitrin; (12) kaempferol 3-O-rutinoside;
(
13) rutin; (14) quercetin 3-O-{R-L-rhamnopyranosyl(1f6)[R-L-
rhamnopyranosyl(1f2)]}-β-D-galactopyranoside; (15) chrysoeriol 7-O-
β-D-glucuronopyranosyl(1f2)-O-β-D-glucuronopyranoside; (16) stel-
laderol; (17) lycoperodine-1.
(4) Wullaert, A.; Heyninck, K.; Beyaert, R. Mechanisms of crosstalk
between TNF-induced NF-kB and JNK activation in hepatocytes.
Biochem. Pharmacol. 2006, 72, 1090–1101.
(5) Sakon, S.; Xue, X.; Takekawa, M.; Sasazuki, T.; Okazaki, T.;
Kojima, Y.; Piao, J. H.; Yagita, H.; Okumura, K.; Doi, T.; Nakano, H. NF-
kB inhibits TNF-R-induced accumulation of ROS that mediate pro-
longed MAPK activation and necrotic cell death. EMBO J. 2003,
ROS production in HepG2 cells at 50 μM without direct
toxicity under the working concentration (Figure 3B). From
our results, both caffeoylquinic acid derivatives and flavo-
noids are responsible for the ROS scavenging activity in H.
fulva.
Identification of Caffeoylquinic Acids and Flavonoids by
UPLC. The isolated caffeoylquinic acids and flavonoids were
analyzed by a UPLC system. A complete UV chromatogram
22, 3898–3909.
(6) Kamata, H.; Honda, S.; Maeda, S.; Chang, L.; Hirata, H.; Karin,
M. Reactive oxygen species promote TNFR-induced death and sus-
tained JNK activation by inhibiting MAP kinase phosphatases. Cell 2005,
1
20, 649–661.
7) Gebhardt, R. Oxidative stress, plant-derived antioxidants and
liver fibrosis. Planta Med. 2002, 68, 289–296.
8) Prasad, S.; Phromnoi, K.; Yadav, V. R.; Chaturvedi, M. M.;
(
(
254 nm) obtained with the standard solution is shown in
(
Figure 4. Peaks were identified by comparison with the
retention time and DAD spectra of the authentic samples
Aggarwal, B. B. Targeting inflammatory pathways by flavonoids for
prevention and treatment of cancer. Planta Med. 2010, 76, 1044–1063.
(
Figure 4). The results indicated that caffeoylquinic acids and
(9) Lin, Y. L.; Wu, C. H.; Huang, Y. J.; Luo, M. H.; Wang, C. N.;
flavonoids gave good resolutions except rutin (13) and chry-
soeriol 7-O-β-D-glucuronopyranosyl(1f2)-O-β-D-glucurono-
pyranoside (15) overlapping. Our results suggest that UPLC
analysis can be considered as a convenient analytical method
for caffeoylquinic acids and flavonoids simultaneously and
promptly.
This is the first report that H. fulva flowers contain high levels
of caffeoylquinic acids and possess potent ROS scavenging
activity. Our results suggest that caffeoylquinic acids, especially
chlorogenic acid in H. fulva flowers, at least partially contributed
to their ROS scavenging effect. Thus, H. fulva flowers may be a
source of natural ROS scavenger and beneficial to the health of
consumers.
Shiao, M. D.; Huang, Y. T. In vitro protective effects of salvianolic acid B
on primary hepatocytes and hepatic stellate cells. J. Ethnopharmacol.
2006, 105, 215–222.
(10) Yang, T. I. A List of Plants in Taiwan; Natural Publishing: Taipei,
Taiwan, 1982; p 353.
(11) Inoue, T.; Iwagoe, K.; Konishi, T.; Kiyosawa, S.; Fujiwara, Y.
Novel 2,5-dihydrofuryl-7-lactam derivatives from Hemerocallis fulva L.
var. kwanzo Regel. Chem. Pharm. Bull. 1990, 38, 3187–3189.
(12) Inoue, T.; Konishi, T.; Kiyosawa, S.; Fujiwara, Y. 2,5-Dihydro-
furyl-γ-lactam derivatives from from Hemerocallis fulva L. var. kwanso
Regel. II. Chem. Pharm. Bull. 1994, 42, 154–155.
(13) Tai, C. Y.; Chen, B. H. Analysis and stability of carotenoids in
the flowers of daylily (Hemerocallis disticha) as affected by various
treatments. J. Agric. Food Chem. 2000, 48, 5962–5968.
8
794
dx.doi.org/10.1021/jf201166b |J. Agric. Food Chem. 2011, 59, 8789–8795

Journal of Agricultural and Food Chemistry
14) Konishi, T.; Fujiwara, Y.; Konoshima, T.; Kiyosawa, S.; Nishi,
ARTICLE
(
M.; Miyahara, K. Steroidal saponins from Hemerocallis fulva var. Kwanso.
Chem. Pharm. Bull. 2001, 49, 318–320.
(
15) Cichewicz, R. H.; Kim, K. C.; McKerrow, J. H.; Nair, M. G.
Kwanzoquinones AꢀG and other constituents of Hemerocallis fulva
Kwanso’ roots and their activity against the human pathogenic trema-
tode Schistosoma mansoni. Tetrahedron 2002, 58, 8597–8606.
16) Cichewicz, R. H.; Nair, M. G. Isolation and characterization of
‘
(
stelladerol, a new antioxidant naphthalene glycoside, and other antiox-
idant glycosides from edible daylily (Hemerocallis) flowers. J. Agric. Food
Chem. 2002, 50, 87–91.
(
17) Zhang, Y.; Cichewicz, R. H.; Nair, M. G. Lipid peroxidation
inhibitory compounds from daylily (Hemerocallis fulva) leaves. Life Sci.
004, 75, 753–763.
18) Moyo, M.; Ndhlala, A. R.; Finnie, J. F.; Van Staden, J. Phenolic
2
(
composition, antioxidant and acetylcholinesterase inhibitory activities of
Sclerocarya birrea and Harpephyllum caffrum (Anacardiaceae) extracts.
Food Chem. 2010, 123, 67–76.
(19) Tsai, M. K.; Lin, Y. L.; Huang, Y. T. Effects of salvianolic acids, A
and B on oxidative stress and hepatic fibrosis in rats. Toxicol. Appl.
Pharmacol. 2010, 242, 155–164.
(20) Lin, Y. L.; Wu, C. H.; Luo, M. H.; Huang, Y. J.; Wang, C. N.;
Shiao, M. S.; Huang, Y. T. Effect of salvianolic acid B on oxidative
damage in primary hepatocytes and hepatic stellate cells. J. Ethnophar-
macol. 2006, 105, 215–222.
(21) Matsuda, H.; Morikawa, T.; Ando, S.; Toguchida, I.; Yoshikawa,
M. Structural requirements of flavonoids for nitric oxide production
inhibitory activity and mechanism of action. Bioorg. Med. Chem. 2003,
11, 1995–2000.
(22) David dos Santos, M.; Almeida, M. C.; Lopes, N. P.; Petto de
Souza, G. E. Evaluation of the anti-inflammation, analgesic and anti-
pyretic activities of the natural polyphenol chlorogenic acid. Biol. Pharm.
Chem. 2006, 29, 2236–2240.
(23) Bouayed, J.; Rammal, H.; Dicko, A.; Younos, C.; Soulimani, R.
Chlorogenic acid, a polyphenol from Prunus domestica (Mirabelle), with
coupled anxiolytic and antioxidant effects. J. Neurol. Sci. 2007,
2
62, 77–84.
24) Lin, L. C.; Kuo, Y. C.; Chou, C. J. Immunomodulatory
principles of Dichrocephala bicolor. J. Nat. Prod. 1999, 62, 405–408.
25) Robards, K.; Prenzler, P.; Tucker, G.; Swatsitang, P.; Glover, W.
(
(
Phenolic compounds and their role in oxidative processes in fruits. Food
Chem. 1999, 66, 401–436.
(
26) Lin, Y. L.; Wang, W. Y.; Kuo, Y. H.; Chen, C. F. Nonsteroidal
constituents from Solanum incanum. J. Chin. Soc. 2000, 47, 247–251.
27) Lin, L. C.; Kuo, Y. C.; Chou, C. J. Immunomodulatory
principles of Dichrocephala bicolor. J. Nat. Prod. 1999, 62, 405–408.
28) Lin, Y. L.; Wang, W. Y. Chemical constituents of Biophytum
sensitivum. Chin. Pharm. J. 2003, 55, 71–75.
29) Lin, Y. L.; Lee, H. P.; Huang, R. L.; Ou, J. C.; Kuo, Y. H. Studies
on the constituents of Hyptis rhomboides. Chin. Pharm. J. 1993,
5, 61–68.
30) Lin, Y. L.; Kuo, Y. H.; Shiao, M. S.; Chen, C. C.; Ou, J. C.
Flavonoid glycosides from Terminalia catappa. J. Chin. Chem. Soc. 2000,
7, 253–256.
31) Yasukawa, K.; Sekine, H.; Takido, M. Two flavonol glycosides
from Lysimachia fortunei. Phytochemistry 1989, 28, 2215–2216.
32) Yahara, S.; Uda, N.; Yoshio, E.; Yae, E. Steroidal alkaloid
glycosides from tomato (Lycopersicon esculentum). J. Nat. Prod. 2004,
7, 500–502.
33) Corse, J.; Lundin, R. E.; Sondheimer, E.; Waiss, A. C. Con-
(
(
(
4
(
4
(
(
6
(
formation analyses of D-(ꢀ)-quinic acid and some of its derivatives by
nuclear magnetic resonance. Phytochemistry 1966, 5, 767–776.
(34) Mabry, T. J.; Markham, K. R.; Thomas, M. B. The Systematic
Identification of Flavonoids; Springer-Verlag: New York, 1970; pp 41ꢀ61.
8
795
dx.doi.org/10.1021/jf201166b |J. Agric. Food Chem. 2011, 59, 8789–8795