Catechin Glucosides: Occurrence, synthesis, and stability

Article abstract of DOI:10.1021/jf9034095

Catechins are flavonoids with suggested health benefits, but are unstable during storage, processing and, after ingestion, during gut transit. We hypothesized that catechin glucosidos, which occur in various plants, could be more stable than unsubstituted catechin, and additionally be deglucosylated in the gut and so act to deliver catechin in a form able to be absorbed. (+)-Catechin O-glucosides from various sources have been used in the course of this investigation. (+)-Catechin 3^'-O-β-D-glucopyranoside (C3^'G), (+)-catechin 5-O-β-D-glucopyranoside (C5G), and (+)-catechin 3-O-β-D-glucopyranoside (C3G) were chemically synthesized. (+)-Catechin 4^'-O-β-D-glucopyranoside (C4^'G) and (+)-catechin 7-O-β-D-glucopyranoside (C7G) were prepared enzymically using preparations from lentil and barley. In general, but with some exceptions, the (+)-catechin glucosidos were more stable between pH 4 and 8 than (+)-catechin, with C3^'G exhibiting greatest stability. The intestinal metabolism of (+)-catechin and all (+)-catechin glucosides in the gut was determined by perfusion of rat intestine in vivo. C3^'G and C5G were extensively deglycosylated in the gut, and C3^'G showed greatest apparent "absorption" as calculated by the difference between effluent and influent. The results show the potential of catechin glucosides, especially C3^'G, as more stable prescursors of catechin.

Full text of DOI:10.1021/jf9034095

2138 J. Agric. Food Chem. 2010, 58, 2138–2149
DOI:10.1021/jf9034095
Catechin Glucosides: Occurrence, Synthesis, and Stability
†
T
HOMAS
R
AAB,† DENIS
ANUEL
B
ARRON,† FRANCIA
†
A
RCE
V
ERA,† VANESSA
,†,‡
C
RESPY
,
M
O
LIVEIRA
,
AND
G
ARY
W
ILLIAMSON
*
†Nestle Research Center, Vers chez les Blanc, 1000 Lausanne 26, Switzerland, and ‡School of Food Science
and Nutrition, University of Leeds, Leeds, LS2 9JT, U.K.
Catechins are flavonoids with suggested health benefits, but are unstable during storage, proces-
sing and, after ingestion, during gut transit. We hypothesized that catechin glucosides, which occur
in various plants, could be more stable than unsubstituted catechin, and additionally be degluco-
sylated in the gut and so act to deliver catechin in a form able to be absorbed. (þ)-Catechin
O-glucosides from various sources have been used in the course of this investigation. (þ)-Catechin
30-O-β-
3-O-β-
D
-glucopyranoside (C30G), (þ)-catechin 5-O-β-
-glucopyranoside (C5G), and (þ)-catechin
D
D
D
-glucopyranoside (C3G) were chemically synthesized. (þ)-Catechin 40-O-β-
-glucopyrano-
side (C40G) and (þ)-catechin 7-O-β-
D-glucopyranoside (C7G) were prepared enzymically using
preparations from lentil and barley. In general, but with some exceptions, the (þ)-catechin gluco-
sides were more stable between pH 4 and 8 than (þ)-catechin, with C30G exhibiting greatest
stability. The intestinal metabolism of (þ)-catechin and all (þ)-catechin glucosides in the gut was
determined by perfusion of rat intestine in vivo. C30G and C5G were extensively deglycosylated in
the gut, and C30G showed greatest apparent “absorption” as calculated by the difference between
effluent and influent. The results show the potential of catechin glucosides, especially C30G, as more
stable prescursors of catechin.
KEYWORDS: Catechin; flavonoid; glucosidase; chemical synthesis; stability; absorption
INTRODUCTION
degraded (10). We therefore hypothesized that catechin glucosylation
could increase stability during processing and storage and also
possibly in the gut lumen. For quercetin, glucosides are not only
more stable but also more bioavailable (after gut deglycosylation), at
least in part due to the improved stability of the glucoside in the
intestine (11). The direct transport of intact glycosides into the
epithelial cells by sugar transporters such as SGLT1 (12) followed
by hydrolysis by cytosolic β-glucosidase (13) may contribute to
absorption, but lactase phlorizin hydrolase, a mammalian β-gluco-
sidase located outside the epithelial cells, is the main determinant of
uptake (14). Therefore, for any potential increase in absorption of
catechin glucosides, they would need to be deglucosylated in the gut.
Only a few studies have been reported on increased stability of
catechins so far. These were mainly chemical derivatization either
using methyl(15) oracetyl groups (16), or the addition ofyeasts to
Catechins belong to the flavan-3-ol class of flavonoids and are
ubiquitous in plants and widely found in many foods. The most
prominent sources of catechins are green and black tea (Camellia
sinensis), grapes/wine, certain fruits and cocoa. The stability of
catechins is pH-dependent, and in neutral or alkaline solution,
they are very unstable and decompose rapidly, whereas in acidic
solution, they are relatively stable (1). In recent years, there has
been considerable interest in the potential health benefits of
catechin-rich food and drink, although some processing methods
are likely to degrade catechins. Several studies have demonstrated
significant risk reduction for cardiovascular diseases in consu-
mers of black and green tea, and there is some evidence that green
tea at high levels of intake provides a benefit in preventing cancers
of the digestive tract, especially gastric cancer (2). The oral
bioavailability of catechins is relatively low (3). The major site
of absorption of tea catechins is the small intestine, where they
can be directly absorbed into epithelial cells by passive or
facilitated diffusion as demonstrated in the rat small intestine
in situ (4). The epithelial cells are important sites of catechin
conjugation (5, 6) comprising glucuronidation, sulfation and
methylation. Catechin levels in human plasma reach their peak 2
to 4 h after ingestion (7-9). One of the factors limiting bioavailability
may be instability in the gut lumen, since in simulated intestinal
juice at pH 8.5, (þ)-catechin and (-)-epicatechin were rapidly
hydroalcoholic solutions of catechins (17). (þ)-Catechin O-β-
D-
glucosides (Figure 1) are relatively rare natural products which
havebeenisolatedfromseveral plant species. Amongthe different
possible isomers, (þ)-catechin 7-O-β-
D-glucoside (compound 1
(C7G)) is probably the most common, having been isolated from
plants belonging to Ulmus (18), Hordeum (Barley) (19), Paeo-
niae (20), Fagopyrum (Buckwheat) (21), Betula (22), Pseudotsuga
(Douglas) (23), Rheum (Rhubarb) (24-26), Rhaphiolepis (25),
Vigna (27), and Schizandra (28). Usually, C7G coexists in plant
extracts with other isomeric glucosides such as (þ)-catechin 5-O-
β-
D
-glucoside (compound 4 (C5G)), (þ)-catechin 40-O-β-
D-gluco-
side (compound 3 (C40G)), or (þ)-catechin 30-O-β-
D-glucoside
*Correspondence to this author. E-mail: g.williamson@leeds.ac.uk.
(compound 2 (C30G)) (see references above). (þ)-Catechin
pubs.acs.org/JAFC
Published on Web 01/29/2010
© 2010 American Chemical Society

Article
J. Agric. Food Chem., Vol. 58, No. 4, 2010 2139
250 ꢀ 3 mm, RP-C18, 5 μm (Macherey-Nagel) column with a guard
column at a flow rate of 0.4 mL/min. Fluorescence detection was at 310 nm
with excitation at 280 nm. Solvent A was water with 0.05% formic acid,
and solvent B was acetonitrile. Gradient 1: time 0 min, 5% B; 20 min, 25%
B; 21 min, 100% B for 6 min, then return to baseline. Gradient 2: time
0 min, 5% B; 5 min, 10% B; 10 min, 10% B; 15 min, 15% B; 22 min, 15%
B; then return to baseline. Gradient 3: 0 min, 10% B; 10 min, 20% B; then
return to baseline.
Analytical HPLC with Triple Quadrupole Mass Spectrometric
Detection. Analytical HPLC with triple quadropole mass spectrometric
detection was performed on a Waters 2690 “Alliance” HPLC (Rupperswil,
Switzerland) with Micromass “Quattro-LC” mass spectrometer
(Manchester, U.K.) and Waters 996 diode array on a Nucleosil 100,
250 ꢀ 3 mm, RP-C18, 5 μm column (Macherey-Nagel) with a guard
column at 0.4 mL/min. Solvent A was water with 0.05% formic acid, and
solvent B was acetonitrile. The gradient (gradient 4) used was as follows:
0 min, 5% B; 20 min, 25% B; 21 min, 100% B for 6 min then return to
baseline. For mass spectrometric detection, electrospray voltage was
3.5 kV, cone voltage was 25, 35 and 50 V, source block temperature was
120 °C, desolvation temperature was 200 °C, with cycle time of 2.1 s, scan
range of m/z 50 to 1500, cone gas of approximately 60 L/h and desolvation
gas of approximately 600 L/h. For MS/MS, cone voltage was 30 V,
collision energy was 40, 50, 60 and 80 eV, with a cycle time of 0.6 s and a
scan range of m/z 50 to 500 and m/z 50 to 700.
Semipreparative HPLC. Semipreparative HPLC was performed
on an Agilent Hewlett-Packard series 1050 HPLC with diode array
detector on a Vydac C18 reverse phase column (22 ꢀ 250 mm) column
at 5 mL/min. Solvent A was water, and B was acetonitrile, with detection
at 280 nm. Gradient 5: 0 min, 5% B; 20 min, 20% B; 21 min, 50% B;
35 min, 50% B; return to baseline. Gradient 6: isocratic 88% water and
12% acetonitrile.
Figure 1. The structures of (þ)-catechin 7-O-β-glucopyranoside (C7G)
1, (þ)-catechin 30-O-β-glucopyranoside (C30G) 2, (þ)-catechin 40-O-β-
glucopyranoside (C40G) 3, (þ)-catechin 5-O-β-glucopyranoside (C5G) 4,
and (þ)-catechin 3-O-β-glucopyranoside (C3G) 5.
NMR Analysis. 1H NMR (300.13 MHz) and 13C NMR (90.56 MHz)
spectra were recorded on a Bruker DPX-360 apparatus, equipped with a
5 mm BBO gradient probehead, and using CD3OD as solvent. The
chemical shifts (in ppm) were expressed with respect to tetramethylsilane
(TMS) as an internal reference, and using CDCl3 or CD3OD as solvents.
The spectrometer was controlled using XWINNMR version 3.1 software
(Bruker Biospin Ltd., Germany), and the spectra were processed using
TopSpin (version 1.3 Bruker Biospin Ltd., Germany). To confirm the
molecular structure, one- and two-dimensional 1H-, 13C NMR, proton
NOE-difference spectra, two-dimensional homonuclear proton correla-
tion COSY and heteronuclear one- and multiple-bond 1H/13C inverse shift
correlation experiment using pulsed field gradients (HSQC,HMBC) were
performed.
3-O-β-D-glucoside (compound 5 (C3G)) is also a natural pro-
duct (29, 30), but it is usually more associated with procyanidin
glycosides, rather than with other (þ)-catechin O-glucoside
isomers. C3G also constitutes a major portion of the phenolic
compounds of the seed coat of lentils (31). Catechin O-R-D-
glucosides have been prepared by biotransformation reac-
tions (32). In contrast, very few publications (33) describe the
preparation of β-glucosides by biotransformation, and only the
chemical synthesis of C3G has been previously reported (34).
The purpose of this work was to prepare all the possible (þ)-
Vacuum Liquid Chromatography and Medium Pressure Liquid
Chromatography. Vacuum liquid chromatography was performed on
silica gel 60 (63-200 μM) from Merck, or on a bonded phase octadecyl
(C18) for flash chromatography (40 μM average particle diameter) from
Baker. Medium pressure liquid chromatography (MPLC) on silica or
catechin mono-O-β- -glucosides 1-5 (Figure 1) by chemical and/
D
or enzymic synthetic methods, and to investigate their stability as
compared to the (þ)-catechin aglycones in vitro and measure
metabolism in vivo in the rat gut lumen.
€
RP-18 was carried out on a Buchi apparatus comprising a 688 chromato-
graphic pump and a 684 fraction collector and equipped with a C 690
Sepacore glass column (dimensions: 36 ꢀ 460 mm, volume 470 mL) plus a
pretreatment column (volume 11 mL). The column plus the precolumn
were filled with 222 g of silica gel 60 (40-63 μm), Merck, or with 250 g of
LiChroprep RP-18 (40-63 μm).
MATERIALS AND METHODS
Materials. (þ)-Catechin hydrate, acetic anhydride, pyridine, 1-meth-
yl-2-pyrrolidone, imidazole, thiophenol, anhydrous K2CO3, tetrabuty-
lammonium bromide, sodium methoxide, acetyl chloride, triethylamine,
dry tetrahydrofuran, dry methanol, 1,2,3,4,6-penta-O-acetyl-β-
D
-(þ)-glu-
Enzymic Synthetic Methods. Preparation of Barley and Lentil
Extracts. Barley seeds (20 g) or lentils (15 or 30 g) were ground in a mill in
the presence of dry ice. Ground sample was extracted 3 times with 20 mL of
cold acetone (4 °C) in polypropylene centrifuge tubes. Each extraction was
carried out using a polytron (15 s at low speed). The acetone extracts were
discarded, and the extracted samples were air-dried and homogenized with
a mortar and a pestle to a fine powder. The powder was then extracted
for 1 h at 4 °C with 100 mL of 0.2 M sodium phosphate buffer pH 6.5
containing 0.1 M NaCl. After centrifugation at 10000g, 4 °C for 20 min,
the supernatant was filtered through a Whatman 541 filter. Desalting
of the filtrate (15 mL) was done using a Biogel PD-6DG column (volume
60 mL, flow rate 2.0 mL/min), elution with sodium phosphate buffer
(0.2 M, pH 6.5) containing 0.1 M NaCl, and collection of fractions
between 8 and 23 mL.
cose, uridine-50-diphosphoglucose disodium salt (UDP-glucose), 1-butyl-
3-methylimidazolium tetrafluoroborate, 1-ethyl-3-methylimidazolium tet-
rafluoroborate, 1,3-dimethylimidazolium methylsulfate, 1-butyl-3-methy-
limidazolium 2-(2-methoxyethoxy)ethylsulfate, 1-butyl-3-methylimida-
zolium octylsulfate, and 1-butyl-4-methylpyridinium tetrafluoroborate
were obtained from Fluka. Tributyltin methoxide, triphenylphosphine
and BF3 etherate were from Aldrich. Almond β-glucosidase and quercetin
were from Sigma. All aqueous solutions were prepared using water filtered
through a Milli-Q water system (Millipore, Bedford, MA), and all
chemicals were of reagant grade. Lentils and barley seeds were purchased
from local supermarkets (Switzerland). 14C-Polyethylene glycol (PEG) was
from GE Healthcare, formerly Amersham Biosciences.
Analytical HPLC with Diode Array and Fluorescence Detection.
Analytical HPLC with diode array and fluorescence detection was
performed on an Agilent series 1100 HPLC using a Nucleosil 100,
Enzymatic Synthesis of Compound 1 (C7G), Compound 2 (C30G), and
Compound 3 (C40G) Using Barley or Lentil Extracts. For analytical work

2140 J. Agric. Food Chem., Vol. 58, No. 4, 2010
Raab et al.
to determine product formation, the following solutions were incubated in
a shaker bath at 37 °C for 22 h: 50 μL of concentrated extract, and 110 μL
of 50 mM citrate phosphate buffer (pH 5.5) containing 5 mM MgCl2,
0.1% Tween, 5 mM UDPG, and 1.6 mM (þ)-catechin. After incubation,
10 μL of 6 N HCl and 80 μL of MeOH were added. Before analytical
HPLC-DAD analysis (gradient 2), 400 μL of MeOH was added to 200 μL
of the incubation solution. For preparative work, 12 ꢀ 15 mL of desalted
barley extract and 4 ꢀ 15 mL of desalted lentil extract were incubated each
Table 1. 1H and 13C NMR Spectral Data of Compound 1 (C7G) and
Comparison with Data from the Literaturea
position
δ 1H; J (Hz)
δ 1H; J (Hz) (19)
δ 13
C
δ 13C (19)
1
2
3
4
4.63 (7.4)
4.03 (m)
4.63 (7.4)
4.04 (m)
82.9
68.6
28.5
83.0
68.6
28.6
A 2.89 (dd 16.4, 5.4) A 2.9 (dd 16.3, 5.4)
B 2.56 (dd 16.3, 8.1) B 2.58 (dd 16.3, 8.0)
with 15 mg of MgCl2 6H2O, 15 mg of Tween 100, 50 mg of UDP-glucose
3
and 20 mg of (þ)-catechin for 20 h at 35 °C, respectively. After incubation,
the solutions were extracted 3 times with 20 mL of n-butanol (saturated
with water). The combined organic extracts were evaporated under
reduced pressure, and the residues were redissolved in 5 mL of a mixture
of water and methanol (2/1). Further purification of the major conversion
products was carried out by semipreparative HPLC with gradient 5.
Resulting fractions were concentrated by evaporation under reduced
pressure to ∼5 mL and then lyophilized. The residues were redissolved
in 200 μL of water and analyzed by HPLC-MS. This resulted in the
isolation of 40 mg of compound 1 (C7G) from the incubation experiment
with barley extract, and of 6.2 mg of compound 2 (C30G) and 15.5 mg of
compound 3 (C40G) from the incubation medium with lentil extracts. For
5
6
157.5 157.6
97.3 97.5
158.7 158.7
96.8 97.0
6.24 (2.2)
6.20 (2.1)
6.24 (2.2)
6.20 (2.4)
7
8
9
156.8 156.9
103.6 103.7
132.0 132.1
115.2 115.3
146.3 146.3
146.3 146.3
116.1 116.1
120.0 120.0
102.2 102.3
10
10
20
30
40
50
60
100
200
300
400
500
600
6.87 (1.5)
6.87 (1.9)
6.80 (8.1)
6.75 (dd 8.1, 1.6)
4.86 (7.1)
6.80 (8.1)
6.75 (dd 8.1, 1.9)
4.86 (7.6)
3.5-3.4
3.5-3.4
3.5-3.4
1H and 13C NMR data of compound 1 (C7G), see Table 1. For 1H and 13
C
NMR data of compound 2 (C30G) and compound 3 (C40G), see Table 2.
Hydrolysis of Products by Almond β-Glucosidase. Samples (100 μL)
from above were evaporated to dryness by a stream of nitrogen.
Citrate buffer (100 μL, 100 mM, pH 5.0) was added containing 3 U of a
β-glucosidase from almonds. After incubating for 3 h at 37 °C in a shaker,
300 μL of MeOH was added and the solutions were analyzed by HPLC-
DAD (gradient 2).
3.5-3.4
3.5-3.4
3.5-3.4
74.8
78.1
71.3
78.0
74.9
78.1
71.4
78.1
62.6
3.5-3.4
A 3.92 (dd 11.9, nd)
B 3.73 (dd 11.9, 4.4) B 3.73 (dd 12.0, 4.9)
3.5-3.4
A 3.92 (dd 11.9, 1.6) 62.5
1
a Chemical shift δ/ppm. Multiplicity abbreviations for H NMR: dd = doublet of
doublets, m = multiplet.
Chemical Synthetic Methods. N-Phenyl-2,2,2-trifluoroacetimi-
doyl Chloride 6. Compound 6 was prepared in one step, according to
the procedure of Tamura et al (35), by heating a mixture of trifluoroacetic
acid and aniline in carbon tetrachloride in the presence of triphenylpho-
sphine and triethylamine.
Table 2. 1H and C NMR Spectral Data of Compound 2 (C30G) and
13
Compound 3 (C40G)a
δ 1H; J (Hz)
[C40G 3]
δ 1H; J (Hz)
[C30G 2]
δ 13
C
δ 13
C
(2,3,4,6)-Tetra-O-acetyl-D-glucose 8. This compound was prepared by
an adaptation of the method of Nudelman et al (36). In a 250 mL 3-neck
round-bottom flask, 6.8 g (17.42 mmol) of 1,2,3,4,6-penta-O-acetyl-
position
[C40G 3] [C30G 2]
1
2
3
4
β-
D
-(þ) glucose 7 was equilibrated under argon. Twenty-five milliliters
4.62 (7.3)
3.98 (m)
4.61 (7.5)
4.01 (m)
82.5
68.8
82.8
68.7
28.7
of dry THF was added under stirring. After complete dissolution, 5 mL
(17.36 mmol) of tributyltin methoxide was added dropwise under stirring.
The medium was refluxed (oil bath at 80 °C) under stirring and argon
atmosphere for 1.5 h. It was poured onto 200 mL of 10% aqueous HCl and
extracted with 2 ꢀ 200 mL of CH2Cl2. The CH2Cl2 extract was washed
with 3 ꢀ 200 mL of H2O, concentrated under reduced pressure, and
purified by VLC on silica using a gradient of EtOAc in hexane as solvent.
This yielded 5.7 g (16.36 mmol; 94%) of a mixture of R- and β-2,3,4,
A 2.82 (dd 16.2, 5.3) A 2.86 (dd 16.2, 5.4) 28.4
B 2.51 (dd 16.2, 7.9) B 2.52 (dd 16.1, 8.1)
157.7
5
6
157.7
96.5
5.93 (1.9)
5.94 (2.2)
96.4
7
157.9
95.6
157.9
95.5
8
5.87 (1.9)
5.85 (2.2)
6-tetra-O-acetyl-D
-glucose 8 as a colorless oil. 1H NMR (360.13 MHz,
9
156.8
104.4
136.4
115.9
148.3
146.5
118.6
119.9
100.8
74.9
156.9
103.8
132.5
116.9
146.5
148.1
117.3
123.8
100.9
74.9
10
10
20
30
40
50
60
100
200
300
400
500
600
CDCl3): δ 2.03 (s, OAc), 2.04 (s, OAc), 2.05 (s, OAc), 2.09 (s, OAc), 2.10
(s, OAc), 3.67 (dd, J = 3.9 and 1.2 Hz, 1-OHR), 3.76 (ddd, J = 10.0, 4.8,
and 2.3 Hz, H5β), 3.97 (d, J = 8.7 Hz, 1-OHβ), 4.09-4.17 (m, H6R þ
H6β), 4.22-4.30 (m, H5R þ H6R þ H6β), 4.75 (brt, J = 8.4 Hz, H1β),
4.90 (m, H2R þ H2β), 5.09(m, H4R þ H4β), 5.27 (dd, J = 18.4 and 8.8 Hz,
H3β), 5.47 (brt, J = 3.7 Hz, H1R), 5.54 (brt, J = 9.8 Hz, H3R). 13C NMR
(90.56 MHz, CDCl3): δ 20.79 (OAc), 20.82 (OAc), 20.88 (OAc), 20.91
(OAc), 20.94 (OAc), 20.95 (OAc), 60.66 (C6β), 62.15 (C6R), 67.37 (C5R),
68.60 (C4β), 68.66 (C4R), 70.04 (C3R), 71.27 (C2R), 72.54 (C5β), 72.41
(C3β), 73.38 (C2β), 90.32 (C1R), 95.73 (C1β), 169.75 (OAc), 169.88 (OAc),
170.39 (OAc), 170.42 (OAc), 171.00 (OAc), 171.05 (OAc), 171.08 (OAc).
6.90 (1.2)
7.2 (1.6)
7.17 (8.3)
6.82 (dd 8.3, 1.4)
4.77 (6.9)
6.85 (8.2)
6.96 (dd 8.2, 1.6)
4.80 (7.4)
3.5-3.4
3.5-3.4
3.5-3.4
3.5-3.4
3.5-3.4
3.5-3.4
3.5-3.4
A 3.89 (dd 11.9, nd)
B 3.71 (dd 11.2, nd)
78.4
71.4
78.2
71.3
3.5-3.4 77.7
A 3.82 (dd 12.1, 1.4) 62.5
77.6
62.4
(2,3,4,6)-Tetra-O-acetyl-
acetimidate 9. 2.0 g (5.74 mmol) of 2,3,4,6-tetra-O-acetyl-
dissolved in 15 mL of CH2Cl2. 1.2 g (8.68 mmol) of anhydrous K2CO3,
0.28 (0.87 mmol) of tetrabutylammonium bromide, and 1.8
(8.67 mmol) of N-phenyl-2,2,2-trifluoroacetimidoyl chloride 6 in 5 mL
of CH2Cl2 were added successively under stirring at rt. The reaction was
started by addition of 0.2 mL of H2O. Stirring at rt was maintained for 2 h.
The medium was diluted with 50 mL of CH2Cl2 and 50 mL of H2O. The
organic layer was washed with 3 ꢀ 100 mL of H2O, dried over anhydrous
Na2SO4, concentrated under reduced pressure, and purified by VLC on
silica using a gradient of EtOAc in hexane as solvent. This yielded 2.63 g of
D
-glucopyranosyl-(N-phenyl)-2,2,2-trifluoro-
D-glucose 8 was
a Chemical shift δ/ppm. Multiplicity abbreviations for H NMR: dd = doublet of
doublets, m = multiplet. The values in bold correspond to the signals shifted
downfield compared to those of (þ)-catechin.
1
g
g
(360.13 MHz, CDCl3): δ 2.03 (s, OAc), 2.04 (s, OAc), 2.06 (s, OAc), 2.08 (s,
OAc), 2.10 (s, OAc), 5.10-5.26 (m), 5.30 (s), 5.54 (t, J = 9.8 Hz), 6.79 (d,
J = 7.7 Hz, H-200/600), 6.84 (d, J = 7.8 Hz, H-200/600), 7.13 (m, H400), 7.32
(m, H300/500). 13C NMR (90.56 MHz, CDCl3): δ 20.45 (OAc), 20.50 (OAc),
20.56 (OAc), 20.58 (OAc), 20.66 (OAc), 20.68 (OAc), 61.38 (C6), 67.72
(C5), 69.44, 69.79, 70.01, 70.19, 72.52, 72.74, 119.20 (C200/600), 119.31
2,3,4,6-tetra-O-acetyl-D-glucopyranosyl-(N-phenyl)-2,2,2-trifluoroaceti-
midate 0.1CH2Cl2 9 as a pale yellow liquid (4.98 mmol; 87%). 1H NMR
3

Article
J. Agric. Food Chem., Vol. 58, No. 4, 2010 2141
(C200/600), 124.71 (C400), 128.89 (C300/C500), 142.91 (C100), 169.00 (OAc),
169.32 (OAc), 169.48 (OAc), 169.75 (OAc), 170.00 (OAc), 170.17 (OAc),
(s, 3H, 3-OAc), 2.64 (dd, 1H, J = 16.5 and 6.4 Hz, H4), 2.75 (dd, 1H, J =
16.6 and 5.2 Hz, H4), 3.31 (m, 1H, H500), 3.44-3.51 (m, 3H, H200 þ H300),
3.71 (dd, 1H, J = 12.3 and 4.4 Hz, H600), 3.77 (dd, 1H, J = 12.3 and 2.5
Hz, H600), 4.75 (d, 1H, J = 7.3 Hz, H100), 4.96 (d, 1H, J = 6.2 Hz, H2), 5.24
(dd, 1H, J = 11.7 and 6.3 Hz, H3), 5.90 (1H, d, J = 2.3 Hz, H8), 5.96 (1H,
d, J = 2.3 Hz, H6), 6.83 (d, 1H, J = 8.3 Hz, H50), 6.91 (dd, 1H, J = 8.3
and 1.9 Hz, H60), 7.14 (d, 1H, J = 1.9 Hz, H20). 13C NMR (90.56 MHz,
CDCl3): δ 21.01 (3-OAc), 24.25 (C4), 62.14 (C600), 70.91 (C3 or C400), 71.00
(C3 or C400), 74.77 (C200), 77.59 (C300), 78.14 (C500), 79.29 (C2), 95.54 (C8),
96.64 (C6), 99.62 (C10), 103.63 (C100), 116.43 (C20), 117.01 (C50), 123.04
(C60), 131.44 (C100), 146.52 (C30), 148.25 (C40), 156.41 (C9), 157.65 (C5),
158.23 (C7), 172.06 (3-OAc).
Compound 2 (C30G). 35.5 mg (0.072 mmol) of 3-O-acetylcatechin 30-O-
β-D-glucoside 18 was added to 75 mL of sodium acetate buffer (0.1 M, pH
5.0). 1.5 g of tannase (KT-50, Kikkoman Corporation, Chiba, Japan) was
added and the solution incubated at 37 °C. Samples (100 μL) were
withdrawn at 2.5, 5, 6, 21, and 23 h and analyzed by HPLC-DAD. After
23 h, 84% of 3-O-acetylcatechin 30-O-β-
D-glucoside had been hydro-
170.55 (OAc), 170.58 (OAc). Anal. (C22H24F3NO10 0.1CH2Cl2) C: calcd,
3
50.28; found 50.09. H: calcd, 4.62; found 4.73. N: calcd, 2.65; found 2.62.
(þ)-3,5,7,30,4-Penta-O-acetylcatechin 11. (þ)-Catechin hydrate 10
(5 g, 17.22 mmol) was dissolved in 40 mL of pyridine, and the solution
was cooled to 0 °C in an ice bath. Acetic anhydride (40 mL) was carefully
added in portions of 10 mL. The medium was removed from the ice bath
and allowed to react at rt for 6 days. The medium was poured onto 400 mL
of ice-cold H2O and, after 2 h, was extracted with 3 ꢀ 100 mL of CH2Cl2.
The CH2Cl2 extract was washed with 2 ꢀ 200 mL of 1 N aqueous HCl and
2 ꢀ 200 mL of H2O. The organic solvent was evaporated under reduced
pressure. The CH2Cl2 extract was purified by VLC using a gradient of
EtOAc in hexane as solvent. This yielded 8.5 g of (þ)-3,5,7,30,40-penta-O-
acetylcatechin 11 as a white solid (16.98 mmol; 99%). 1H NMR (360.13
MHz, CDCl3): δ 1.93 (s, 3H, 3-OAc), 2.24 (s, 3H, OAc), 2.26 (s, 3H, OAc),
2.27 (s, 3H, OAc), 2.73 (dd, 1H, J = 17.2 and 6.6 Hz, H4), 2.86 (dd, 1H,
J = 16.9 and 5.0 Hz, H4), 5.27 (d, 1H, J = 6.5 Hz, H2), 5.32 (1H, m, H3),
6.61 (1H, d, J = 2.2 Hz, H6), 6.68 (1H, d, J = 2.2 Hz, H8), 7.28 (d, 1H,
J = 8.4 Hz, H50), 7.30 (brd, 1H, J = 2.1 Hz, H20), 7.36 (ddd, 1H, J = 8.4,
2.1, and 0.5 Hz, H60). 13C NMR (90.56 MHz, CDCl3): δ 20.48 (OAc),
20.50 (OAc), 20.62 (OAc), 20.79 (OAc), 20.93 (OAc), 24.76 (C4), 68.79
(C3), 78.51 (C2), 108.35 (C8), 109.91 (C6), 111.49 (C10), 122.87 (C20),
124.60 (C50), 125.25 (C60), 137.25 (C10), 143.39 (C30 þ C40), 150.76 (C5),
151.07 (C7), 155.33 (C9), 168.59 (OAc), 168.66 (OAc), 168.92 (OAc),
169.33 (OAc), 170.08 (3-OAc). The 13C NMR data agrees with published
data in the same solvent (37).
lyzed. The incubation solution was extracted three times with 50 mL of
n-butanol (saturated with water); the combined organic extracts were
evaporated under reduced pressure to a final volume of about 1 mL. Two
milliliters of acetonitrile/water (50/50) was added. Purification of C30G
was achieved by semipreparative HPLC (gradient 6), and frac-
tions were collected, combined and freeze-dried. A total of 22 mg
1
(0.048 mmol; 67%) of C30G 2 was obtained. H and 13C NMR data of
2: see Table 1.
Compound 4 (C5G). In a 250 mL 3-neck round-bottom flask, 670 mg
(12.4 mmol) of NaOMe was suspended in 60 mL of dry MeOH. The
solution was equilibrated under argon and cooled to 0 °C (ice bath).
360 mg (0.46 mmol based on a Mr of 788.7 for acetylated mono-
glucosylcatechins) of the mixture of compounds 15-17, dissolved in
2.5 mL of dry CH2Cl2, were added under inert atmosphere. The medium
was allowed to react for 4.5 h at 0 °C under stirring and argon atmosphere.
The reaction was stopped by the addition of 5.5 mL of 10% aqueous HCl
at 0 °C. Finally the medium was diluted with 60 mL of H2O (pH was
around 4). The experiment was repeated using 700 mg (12.96 mmol) of
NaOMe and 376.4 mg (0.48 mmol) of the mixture of compounds 15-17.
The media resulting from the two experiments were mixed, concentrated
under reduced pressure (removal of MeOH) to a volume of 10 mL, and
purified by MPLC on RP-18 using a gradient of MeOH in H2O as solvent.
This yielded 43 mg of pure compound 4 (C5G) and 14 mg of pure
compound 2 (C30G), beside fractions containing mixtures of catechin
glucosides 3-5. NMR data of compound 4 (C5G): 1H NMR (360.13
MHz, CD3OD) and 13C NMR (90.56 MHz, CDCl3) chemical shifts are
shown in Table 3.
(þ)-5,7,30,40-Tetra-O-acetylcatechin 19. (þ)-Catechin hydrate 10 (1 g,
3.35 mmol) was dissolved in 10 mL of DMF, and the solution was cooled
to 0 °C (ice bath). 2.3 mL (16.59 mmol) of triethylamine and 1.2 mL (16.89
mmol) of acetyl chloride were added dropwise, successively. The medium
was stirred at 0 °C for 10 min and at rt for 1 h. The medium was poured
onto 100 mL of H2O and left at 4 °C for 48 h. The precipitate was dissolved
in 50 mL of EtOAc and washed with 50 mL of 1 N aqueous HCl and 2 ꢀ
50 mL of H2O. It was purified by MPLC on silica using a gradient of
EtOAc in hexane as solvent. This yielded 533.3 mg (1.16 mmol; 35%) of
5,7,30,40-tetra-O-acetylcatechin 19 as a colorless oil. 1H NMR (360.13
MHz, CDCl3): δ 2.25 (s, 3H, OAc), 2.78 (s, 3H, OAc), 2.29 (s, 6H, OAc),
2.59 (dd, 1H, J = 16.4 and 9.4 Hz, H4), 2.93 (dd, 1H, J = 16.4 and 5.6 Hz,
H4), 3.88 (1H, m, H3), 4.69 (d, 1H, J = 8.4 Hz, H2), 6.54 (1H, d, J = 2.2 Hz,
H6), 6.58 (1H, d, J = 2.2 Hz, H8), 7.20 (d, 1H, J = 8.3 Hz, H50), 7.24
(d, 1H, J = 2.0 Hz, H20), 7.30 (dd, 1H, J = 8.4 and 2.0 Hz, H60). 13C
NMR (90.56 MHz, CDCl3): δ 20.57 (OAc), 20.63 (OAc), 20.72 (OAc),
21.04 (OAc), 28.12 (C4), 67.44 (C3), 81.14 (C2), 107.76 (C8), 108.72 (C6),
111.75 (C10), 122.35 (C20), 123.64 (C50), 125.59 (C60), 136.66 (C10), 142.17
(C30 or C40), 142.26 (C30 or C40), 149.48 (C5 or C7), 149.67 (C5 or C7),
155.03 (C9), 168.30 (OAc), 168.42 (OAc), 168.54 (OAc), 169.08 (OAc).
Anal. (C23H22O10 0.1 CH2Cl2) C: calcd, 59.42; found 59.61. H: calcd, 4.79;
found 5.03.
(þ)-3,7,30,40-Tetra-O-acetylcatechin 12, (þ)-3,5,7,30-tetra-O-acetylca-
techin 13, and (þ)-3,5,7,40-tetra-O-acetylcatechin 14. 3,5,7,30,40-Penta-O-
acetylcatechin 11 (2 g, 4 mmol) was dissolved in 10 mL of 1-methyl-2-
pyrrolidone (NMP). The solution was cooled to 0 °C in an ice bath, and
then 95 mg (1.4 mmol) of imidazole and 0.5 mL (4.89 mmol) of thiophenol
were added, successively. The medium was removed from the ice bath and
was left to react for 6 h at rt. It was diluted with 50 mL of CH2Cl2, and the
organic layer was washed with 50 mL of 1 N aqueous HCl and 3 ꢀ 50 mL
of H2O. The CH2Cl2 extract was purified by MPLC on silica using a
gradient of acetone in hexane as solvent. This yielded 910 mg (1.98 mmol;
50%) of a mixture of (þ)-3,7,30,40-tetra-O-acetylcatechin, (þ)-3,5,7,
30-tetra-O-acetylcatechin and (þ)-3,5,7,40-tetra-O-acetylcatechin (12-14),
as a colorless oil.
(3,7,30,40,200,300,400,600)-Octa-O-acetyl-(þ)-catechin 5-O-β-
D
-Glucoside
-Glucoside
D-Gluco-
15, 3,5,7,30,200,300,400,600-octa-O-acetyl-(þ)-catechin 40-O-β-
D
16, and 3,5,7,40,200,300,400,600-octa-O-acetyl-(þ)-catechin 30-O-β-
side 17. 398.8 mg (0.87 mmol assuming that the sample was composed
of tetra-O-acetylated catechins only) of
12-14 and 0.76 g (1.46 mmol) of 2,3,4,6-tetraacetyl-
a
mixture of compounds
-glucopyranosyl-
D
(N-phenyl)-2,2,2-trifluoroacetimidate 9 were dissolved in 15 mL of dry
CH2Cl2. 35 μL (0.28 mmol) of BF3 etherate was added, and the mixture
was stirred at rt overnight. The medium was diluted with 50 mL of CH2Cl2.
The CH2Cl2 extract was washed with 3 ꢀ 50 mL of H2O, concentrated
under reduced pressure, and purified by VLC on silica using a gradient of
EtOAc in hexane as solvent. This yielded 728.6 mg of a mixture of
3,7,30,40,200,300,400,600-octa-O-acetyl-(þ)-catechin 5-O-β-
D
-glucoside, 3,5,7,
-glucoside and 3,5,7,
-glucoside (15-17) as
30,200,300,400,600 octa-O-acetyl-(þ)-catechin 40-O- β-
D
40,200,300,400,600-octa-O-acetyl-(þ)-catechin 30-O- β-
D
a colorless oil.
(þ)-3-O-Acetylcatechin 30-O-β-
D-Glucoside 18. In a 250 mL round-
bottom flask, 1.4 g (1.77 mmol based on a Mr of 788.7 for acetylated
monoglucosylcatechins) of the mixture of compounds 15-17 was dis-
solved in 10 mL of MeOH and equilibrated under argon. The solution was
cooled to 0 °C (ice bath), which resulted in partial crystallization of the
substrate. 35 mL (35 mmol) of a 1 N solution of NaOH in MeOH was
added under inert atmosphere. This resulted in the dissolution of the
crystals and the appearance of a yellow coloration. The medium was
stirred for 5 min under inert atmosphere. The reaction was stopped by the
addition of 38 mL of 1 N aqueous HCl (the medium turned pale yellow),
and the acidic solution was removed from the ice bath and directly purified
by VLC on RP-18 using a gradient of MeOH in H2O as solvent, followed
by an MPLC on RP-18 using a similar gradient of solvents. This resulted in
(5,7,30,40,200,300,400,600)-Octa-O-acetyl-(þ)-catechin 3-O-β-
D-Glucoside
20. 5,7,30,40-Tetra-O-acetylcatechin 19 (287.3 mg, 0.63 mmol) and 495.4
the isolation of 35.5 mg of 3-O-acetylcatechin 30-O-β-
D
-glucoside 18, from
mg (0.95 mmol) of 2,3,4,6-tetra-O-acetyl-D-glucopyranosyl-(N-phenyl)-
a complex mixture of products. 1H NMR (360.13 MHz, CD3OD): δ 1.94
2,2,2-trifluoroacetimidate 9 were dissolved in 10 mL of dry CH2Cl2.

2142 J. Agric. Food Chem., Vol. 58, No. 4, 2010
Raab et al.
Table 3. Comparison of the NMR Data of Compound 4 (C5G)a
MeOH-d4
MeOH-d4 (33)
MeOH-d4
82.9
68.7
DMSO-d6 þ D2O (25)
MeOH-d4 (33)
H-2
H-3
4.59
3.97
4.58
3.97
C-2
C-3
80.9
65.9
82.8
68.5
H-4a
H-4b
H-6
2.57
3.00
2.56
3.01
C-4
C-5
28.4
158.0 or 158.1
96.9
27.3
28.4
156.4
94.9
154.7c
96.0
157.9
98.0b
157.8
96.7b
156.5
103.2
131.9
115.0
146.1
146.1
115.9
119.8
102.4
74.8
6.26
6.02
6.25
6.01
C-6
C-7
H-8
H-20
158.0 or 158.1
98.1
156.7
6.82
6.59
6.81
6.75
C-8
C-9
H-500
156.1c
100.3d
130.3
114.3
144.6
144.6
115.0
118.4
101.2d
73.1
H-600
H-100
H-200
H-3
6.70
4.82
6.66
4.83
C-10
C-10
C-20
C-30
C-40
C-500
C-600
C-1
C-200
C-300
C-400
C-500
C-600
103.4
132.1
3.35-3.47
3.35-3.47
3.35-3.47
3.35-3.47
3.71
3.42-3.90
3.42-3.90
3.42-3.90
3.42-3.90
3.42-3.90
3.42-3.90
115.23
146.3
146.3
H-400
H-500
H-600a
H-600b
116.1
120.0
3.90
102.6
74.9
78.2 or 78.3
71.3
76.3
69.4
78.1
71.2
78.2 or 78.3
62.6
76.7
60.4
78.1
62.5
a C5G was chemically synthesized here, previously isolated from Rhaphiolepis umbellata (25), or produced by cultured cells of Eucalyptus perriniana (33). b Assignments must
be reversed taking into account our HSQC and HMBC data for compound 4 (C5G). c The assignment between these two carbons probably needs to be reversed, taking into
account the HMBC data recorded for compound 4 (C5G). d The assignment between these two carbons probably needs to be reversed, taking into account the HSQC data we
obtained for compound 4 (C5G).
Table 4. 13C NMR Data of Compound 5 (C3G)a
MeOH-d4
acetone-d6 þ D2O (45)
MeOH-d4
acetone-d6 þ D2O (45)
C-2
C-3
C-4
C-5
C-6
C-7
C-8
C-9
C-10
C-10
C-20
80.0
76.1
79.3
75.6
C-30
C-40
C-500
C-600
C-1
C-200
C-300
C-400
C-500
C-600
146.2 or 146.3
146.2 or 146.3
116.3
145.6
145.7
116.0
119.1
103.5
74.6
26.0
157.4 or 157.8
96.4
28.4
156.9
95.3b
157.4
96.4b
156.0
100.0
131.7
114.6
119.5
103.8
157.4 or 157.8
95.6
156.6
75.1
77.7
77.3
71.1
71.6
78.0
100.7
132.2
77.1
62.5
62.8
114.8
a Comparison of C3G from this study and as previously isolated from the leaves Quercus miyagii (45). b The assignment between these two carbons probably needs to be
reversed in ref 45, taking into account the HSQC and HMBC data recorded for compound 5 C3G.
Twenty-five microliters (0.2 mmol) of BF3 etherate was added, and the
mixture was stirred at rt overnight. The solvent was evaporated under
reduced pressure, the residue was taken up in 5 mL of CH2Cl2, and the
solution was directly purified by MPLC on silica using a gradient of
EtOAc in hexane as solvent. This yielded 245.5 mg (0.31 mmol; 49%) of
Chemical Synthesis of Compound 5 (C3G). In a 100 mL 2-neck round-
bottom flask, 190 mg (3.52 mmol) of NaOMe was suspended in 15 mL of
dry MeOH. The solution was equilibrated under argon and cooled to 0 °C
(ice bath). 106.2 mg (0.135 mmol) of compound 20, dissolved in 1 mL of
dry CH2Cl2 þ 4 mL of dry MeOH, was added under inert atmosphere. The
medium was allowed to react for 4.5 h at 0 °C under stirring and argon
atmosphere. The reaction was stopped by the addition of 1.6 mL of 10%
aqueous HCl, and the medium was diluted with 40 mL of H2O (pH was
around 4), concentrated under reduced pressure (removal of MeOH) to a
volume of about 20 mL, and purified by VLC on RP-18 using a gradient of
MeOH in H2O as solvent. This yielded 57.2 mg (0.126 mmol; 93%) of
compound 5 (C3G) as a white powder. 1H NMR (360.13 MHz, CD3OD):
δ 2.74 (m, 2H, H4), 3.12 (m, 1H, H200), 3.18-3.20 (m, 1H, H500), 3.21-3.27
(m, 2H, H300 þ H400), 3.66 (dd, 1H, J = 11.8 and 5.4 Hz, H600), 3.86 (dd,
1H, J = 11.8 and 2.1 Hz, H600), 4.19 (d, 1H, J = 7.7 Hz, H100), 4.24 (dd,
1H, J = 11.4 and 5.9 Hz, H3), 4.94 (d, 1H, J = 5.9 Hz, H2), 5.90 (1H, d,
J = 2.3 Hz, H8), 5.94 (1H, d, J = 2.3 Hz, H6), 6.70 (dd, 1H, J = 8.1 and
1.9 Hz, H60), 6.75 (d, 1H, J = 8.1 Hz, H50), 6.82 (d, 1H, J = 1.9 Hz, H20).
13C NMR (90.56 MHz, CDCl3) chemical shifts are shown in Table 4.
Influence of pH on Stability. The stability of the catechin glucosides
1-5 was tested by incubating them individually at 37 °C (additionally
70 °C for pH 4 and 5) at concentrations of 40 μg/mL in 50 mM sodium
phosphate buffers at pH values of 4, 5, 6, 7, and 8. The samples were
analyzed by HPLC-FLD (gradient 3) several times over 50 h.
5,7,30,40,200,300,400,600-octa-O-acetyl-(þ)-catechin 3-O-β-
D-glucoside 20 as a
colorless oil. 1H NMR (360.13 MHz, CDCl3): δ 1.85 (s, 3H, sugar OAc),
1.95 (s, 3H, sugar OAc), 2.00 (s, 3H, sugar OAc), 2.09 (s, 3H, sugar OAc),
2.26 (s, 3H, catechin OAc), 2.30 (s, 9H, catechin OAc), 2.75 (dd, 1H, J =
16.7 and 9.2 Hz, H4), 3.06 (dd, 1H, J = 16.8 and 5.8 Hz, H4), 3.61 (m, 1H,
H500), 4.08 (1H, m, H3), 4.13 (dd, 1H, J = 12.4 and 2.5 Hz, H600), 4.22 (dd,
1H, J = 12.4 and 4.9 Hz, H600), 4.24 (d, 1H, J = 7.9 Hz, H100), 4.80 (d, 1H,
J = 8.7 Hz, H2), 4.88 (m, 1H, H200), 5.02 (m, 2H, H300 þ H400), 6.54
(1H, d, J = 2.2 Hz, H6), 6.56 (1H, d, J = 2.2 Hz, H8), 7.21-7.26 (m, 3H,
H20 þ H50 þ H60). 13C NMR (90.56 MHz, CDCl3): δ 20.46 (OAc),
20.55 (OAc), 20.58 (OAc), 20.63 (OAc), 20.69 (OAc), 20.73 (OAc), 21.05
(OAc), 28.46 (C4), 61.89 (C600), 68.31 (C400), 71.07 (C200), 71.79 (C500),
72.77 (C300), 74.44 (C3), 78.85 (C2), 100.19 (C100), 107.64 (C8), 108.89
(C6), 111.55 (C10), 122.34 (C20), 123.34 (C50), 125.89 (C60), 136.43
(C10), 142.24 (C30 or C40), 142.36 (C30 or C40), 149.35 (C5), 149.70 (C7),
154.75 (C9), 167.93 (OAc), 168.02 (OAc), 168.51 (OAc), 168.97 (OAc),
169.29 (OAc), 169.46 (OAc), 170.04 (OAc), 170.58 (OAc). Anal.
(C37H40O19 0.3 CH2Cl2) C: calcd, 55.02; found 55.04. H: calcd, 5.03;
found 5.29.

Article
J. Agric. Food Chem., Vol. 58, No. 4, 2010 2143
Figure 2. HPLC chromatogram after incubation of a desalted barley extract with UDP-glucose and (þ)-catechin.
Stability in a Physiological Buffer. The stability of the catechin
glucosides 1-5 was tested by incubating them individually at 37 °C at
concentrations of 15 μM in “perfusion buffer” (pH 6.7) of the following
composition: NaCl, 50 mM; KCl, 40 mM; K2HPO4, 2.5 mM; KH2PO4,
5 mM; tripotassium citrate, 10 mM; MgCl2, 2 mM; CaCl2, 2 mM; glucose
8 mM; PEG 4000, 2 g/L; taurocholate, 1 mM; NaHCO3, 5 mM. Aliquots
were withdrawn and analyzed by HPLC (gradient 2) after 0, 3.5, 18, 25,
and 42 h.
and 13.7 (major) min were formed according to analytical HPLC
with gradient 1 (Figure 2), both of which disappeared after
almond β-glucosidase treatment. The main peak was purified
by semipreparative HPLC, concentrated by evaporation under
reduced pressure, and analyzed by HPLC-ESI-MS. The product
had a molecular ion at m/z 451 = [M - H]- in the negative ion
mode and a fragment ion at m/z 289. The collision induced
dissociation (CID) spectrum of the parent ion at m/z 451
shows the presence of a daughter ion at m/z 289 resulting from
Intestinal Rat Perfusion Study. Intestinal perfusion of rat intestine
was carried out as previously described (38) using male white Wistar rats at
a maximum age of 16 weeks, purchased from Charles River (France), with
8 rats per compound. To avoid loss of the perfused compound due to
possible affinity of the compound for the PVC, the catheters of perfusion
were saturated with the tested solution for 5 min before starting the
experiment. 14C-PEG 4000 (2.5 μCi/mL) was added to the “perfusion
solution” (see above). Catechin glucoside solutions were prepared at
100 mM in DMSO and added to the perfusion solution to give a final
concentration of 30 μM, and pumped at a flow rate of 0.5 mL/min. The
influent and effluent were collected in plastic tubes containing EDTA
0.1%, vitamin C 20%, and NaH2PO4 0.4 M at pH 2.6 (20 μL/mL of
solution). All the samples were stored on ice during the experiment, then at
-20 °C until analysis.
1
the cleavage of one glucose molecule. On the basis of the H
and 13C NMR data and comparison with literature data (19,39),
compound 1 (C7G) was identified as (þ)-catechin 7-O-β-
D-
glucoside. A total of 40 mg of compound 1 (C7G) was
synthesized by incubation of 240 mg (þ)-catechin with UDP-
glucose and barley extract, with purification by semipreparative
HPLC.
Enzymatic Synthesis of Compound 2 (C30G) and Compound 3
(C40G) Using Lentil Extracts. As for barley above, the lentils were
tested for the presence of a glycosyltransferase activity accepting
catechin as a substrate. A desalted extract from lentils was
incubated with UDP-glucose and (þ)-catechin. HPLC-DAD
analysis revealed the formation of several conversion products
of catechin exclusively in the presence of UDP-glucose. Further
purification was carried out by semipreparative HPLC, which
gave 5 fractions (Figure 3), collected and concentrated by eva-
poration under reduced pressure and analyzed by HPLC-
ESI-MS. In fraction 2 (15.5 mg), a molecular ion at m/z 451 =
[M - H]- in the negative ion mode was detected. The collision
induced dissociation (CID) spectrum of the parent ion at m/z 451
showed the presence of a daughter ion at m/z 289 resulting from
the cleavage of one glucose molecule. This, together with the
NMR data (Table 2), identified fraction 2 as pure C40G 3. Indeed,
for the NOE-difference experiment on compound 3 (C40G), an
enhancement of the aglycone proton H-50 was obtained upon
irradiation of the anomeric proton H-100, indicating attachment
of the sugar in position C-40. Consistent with this assumption are
the chemical shifts observed within the B-ring of the aglycone.
Compared to the data for (þ)-catechin, the signals for C-10, C-30
and C-50 as well as of H-50 (Table 2) are shifted downfield.
Fraction 3 (6.2 mg) was C30G 2 based on NMR data (Table 2). In
the NOE-difference experiment on compound 2 (C30G), an
enhancement of the aglycone proton H-20 was obtained upon
irradiation of the anomeric proton H-100, indicating 30-glucosyla-
tion. Attachment of the sugar in position C-30 was confirmed by
the downfield shift of the signals for C-20, C-40 and C-60 as well as
of H-20 (Table 2).
For each animal, radioactivity of 14C-PEG 4000 was measured in
duplicate in the influent and effluent using scintillation counting, and used
to correct for water absorption. Correction of the concentration to
account for water absorption (Wabs) was calculated as the ratio of
concentration of the nonabsorbable marker, 14C-PEG 4000, in the inlet
and outlet perfusion solution.
Wabs ¼ Cinð14C-PEGÞ=Coutð
14C-PEGÞ
Catechin and catechin glucosides were directly analyzed, after centri-
fugation, by HPLC-MS (gradient 4). The fraction apparently absorbed
(Faa) of each compound was estimated as the fraction disappearing from
the intestinal lumen, correcting for a minor volume change by using
[
14C-PEG] as a nonabsorbable marker:
Faa ð%Þ ¼ ððCin -ðCoutWabsÞÞ=CinÞ ꢀ 100
where Cin and Cout are the concentrations of the compound of interest in
the inflow and outflow solutions respectively. The Faa corresponded to the
apparent absorption of the compound expressed in % of the concentration
of the compound in the inlet perfusion solution.
RESULTS AND DISCUSSION
Enzymic Synthesis of Compound 1 (C7G) Using Barley Extracts.
C7G 1 was found in barley (19), and so barley seeds were tested
for the presence of a glycosyltransferase activity accepting cate-
chin as a substrate. After incubation of (þ)-catechin with barley
extract, two new products with retention times of 12.6 (minor)

2144 J. Agric. Food Chem., Vol. 58, No. 4, 2010
Raab et al.
Figure 3. Semipreparative RP18-HPLC-UV chromatogram after incubation of a desalted lentil extract with UDP-glucose and (þ)-catechin. Fractions 1-5
collected as indicated by dashed lines.
Figure 4. Preparation of the glucosylation reagent 9: (a) tributyltin methoxide, THF, reflux 1.5 h, 94%; (b) K2CO3, tetrabutylammonium bromide, compound6,
CH2Cl2, H2O, 2 h at rt, 87%.
Chemical Synthesis of Catechin Glucosides. With the exception
of C3G, none of the other catechin glucosides have been chemi-
cally synthesized before. Glucosylation of catechin was performed
using a trifluoroacetimidoyl glucosyl donor (40), i.e. 2,3,4,6-tetra-
followed the order of acidity of the phenolic hydroxyls. Such an
approach, applied to polyacetylated catechin, was expected to
have little regioselectivity, due to the similar acidity of its four
phenolic hydroxyl groups (41). Nevertheless, we decided to
explore this method as a possible route to a mixture of mono-
deacetylated catechin derivatives. Thus polyacetylated catechin 11
was submitted to deacetylation in the presence of thiophenol
(Figure 5). The reaction was, as expected, not regioselective.
However, after 6 h of reaction at rt, the pool of monodeacetyl-
ated derivates 12-14 amounted to 50% of the reaction products,
while some 3,5,7,30,40-penta-O-acetylcatechin 11 remained un-
reacted, and a number of higher deacetylated compounds were
formed. Reducing the reaction time to 1 h efficiently reduced
the amount of higher deacetylated derivatives. However, this time,
unreacted 3,5,7,30,40-penta-O-acetylcatechin 11 was by far the
major constituent of the medium, and the amount of mono-
deacetylated compounds that we could isolate was less than 9%
(results not shown). Therefore the latter conditions were not
considered suitable. Purification of the reaction medium by
MPLC succeeded in the separation of the fraction of mono-
deacetylated products from other contaminants. However the
purification of individual monodeacetylated catechins was diffi-
cult and time-consuming.
O-acetyl-D-glucopyranosyl-(N-phenyl)-2,2,2-trifluoroacetimidate
9 (Figure 4). The synthesis of the glucosyl donor 9 involved on one
hand the preparation of N-phenyl-2,2,2-trifluoroacetimidoyl
chloride 6, and on the other the preparation of a 1-hydroxyglucose
protectedwith acetate groups (Figure 4, compound 8). Compound
6 was prepared from trifuoroacetic acid, aniline and carbon
tetrachloride in the presence of triphenylphosphine and triethyla-
mine, according to an established procedure (35). 2,3,4,6-Tetra-O-
acetyl-D-glucose 8 was synthesized by deacetylation of 1,2,3,4,
6-penta-O-acetyl-β-D-glucose 7 with tributyltin methoxide. Final-
ly compound 9 was synthesized by the base-catalyzed reaction of 6
and 8.
A direct glucosylation of (þ)-catechin was impossible since the
compound was not soluble in dichloromethane, the usual solvent
of the glucosylation reaction. In addition, assuming that some
glucosylated compounds could be produced, it was unlikely that
the reaction would have stopped at monoglucosylation. Thus
glucosylation of catechin required access to a panel of catechin
derivatives protected on all hydroxyl groups, except the one to be
glucosylated. This approach would involve a total synthesis of the
catechin skeleton. However, a number of studies have demon-
strated the feasibility ofthepartial and regioselectivedeacetylation
of fully acetylated polyphenols under various conditions (40). In
fact, the removal of acetate groups from quercetin pentaacetate
followed the order of susceptibility of the corresponding acetate
toward nucleophilic removal, i.e. 7 > 40 > 3 > 30 > 5. This also
Therefore we decided to directly perform the two next gluco-
sylation and deprotection steps (Figure 5) on the mixture of
monodeacetylated catechins 12-14. While the glucosylation step,
giving rise to the acetylated glucosides 15-17, proceeded
smoothly, the deprotection of the latter compounds was more
difficult. In a first trial, we attempted to deacetylate the mixture of
compounds 15-17 using dilute sodium hydroxide in aqueous

Article
J. Agric. Food Chem., Vol. 58, No. 4, 2010 2145
Figure 5. Preparation of compound 2 (C30G), and compound 4 (C3G): (a) Ac2O, pyridine, 6 days at rt, 99%; (b) imidazole, thiophenol, NMP, 6 h at rt; (c)
mixture of compounds 12-14, compound 9, BF3 etherate, dry CH2Cl2, 16 h at rt; (d) mixture of compounds 15-17, aqueous NaOH þ MeOH, 5 min under
argonatmosphere at0 °C;(e) tannase, 0.1M sodium acetate buffer pH5;(f) mixture of compounds15-17, NaOMe, dry MeOHþ dry CH2Cl2, 4.5 h at0° and
argon atmosphere.
methanol under argon atmosphere. This resulted in an extensive
degradation of the products, and only 3-O-acetylcatechin 30-O-
glucoside 18 could be isolated from this experiment, albeit in
extremely low yield. In the HMBC spectrum of compound 18, a
correlation was observed between H-100 at 4.75 ppm and the
carbon signal at 146.52 ppm, assignable to either C-30 or C-40. The
final demonstration of the position of glucosylation was obtained
by means of a NOE-difference experiment. In fact, irradiation of
H-100 at 4.75 ppm induced a strong NOE effect on the signal of
H-20 at 7.14 ppm. This unequivocally identified compound 18 as
either pig liver esterase (43) or lipase (37) has been reported to be
unsuccessful, no data was available so far concerning the use of
tannase. Tannase cleaves the 3-O-gallate ester bond in com-
pounds such as epigallocatechin gallate, one of the major con-
stituents of green tea extracts. In fact, on incubation of compound
18 with tannase, compound 2 (C30G) was formed in fair 67%
yield.
The deprotection of acetylated glucosides 15-17 was also
achieved using methanolic sodium methoxide for 4.5 h. This
resulted in complete deprotection, including hydrolysis of the 3-
acetyl group, and the already isolated compound 2 (C30G), aswell
as the new compound 4 (C5G), could be isolated in pure form
from the mixture. In the HMBC spectrum of compound 4 (C5G),
no correlation could be seen between the anomeric H-100 sugar
proton and one of the carbons of the catechin ring. Therefore, the
position of attachment of the glucose was again demonstrated by
means of a NOE-difference experiment. In fact, irradiation of
H-100 at 4.85 ppm only produced a significant NOE effect on the
signal at 6.26 ppm, assignable to H-6. This unequivocally
demonstrated that 4 was the 5-O-glucoside of (þ)-catechin.
Indeed, in the case of 7-O-glucosylation, both the signals of H-
6 and H-8 would have been affected. Furthermore, the 13C NMR
3-O-acetylcatechin 30-O- β-
D-glucoside. This experiment demon-
strated that the 3-O-protected catechin derivatives were more
stableinaqueous alkalineconditions, suggestingthat the presence
of a free 3-hydroxyl group favored the degradation of other
compounds. We subsequently explored the possibility to prepare
compound 2 (C30G) from its 3-O-acetyl analogue 18. An alkaline
hydrolysis of the 3-acetate was of course out of question, con-
sidering the above-mentioned stability problems, despite the fact
that the production of catechin from 3,5,7,30,40-penta-O-acetyl-
catechin had been previously carried out using an aqueous
solution of sodium hydroxide (42). Furthermore, the 3-O-acetyl
group is resistant to various chemical and enzymatic conditions.
Indeed, our treatment of 3,5,7,30,40-penta-O-acetylcatechin with
sodium methoxide for 0.5 h at 0 °C resulted in the formation of
3-O-acetylcatechin as sole product (results not shown). On the
other hand, although enzymatic hydrolyses of the 3-acetate using
of 4 was similar to published data for (þ)-catechin 5-O-β-
D-
glucoside in DMSO-d6 þ D2O (Table 3), taking into account a
difference of about 2 ppm due to the solvent effect (MeOH-d4 vs
DMSO-d6). However, in the HSQC spectrum of compound 4

2146 J. Agric. Food Chem., Vol. 58, No. 4, 2010
Raab et al.
Figure 6. Preparation of compound 5 (C3G): (a) triethylamine, acetyl
chloride, DMF, 1 h at rt, 35%; (b) compound 9, BF3 etherate, dry CH2Cl2,
16 h atrt, 49%; (c) NaOMe, dry MeOH þ dry CH2Cl2, 4.5h at 0° and argon
atmosphere, 93%.
(C5G), of the two 102.56 and 103.41 13C signals, only that at
102.56 ppm was correlated to a proton signal. This unambigu-
ously assigned the 102.56 ppm peak to C-100 (CH), and its
neighboring 103.41 signal to C-10 (quaternary). Thus the assign-
ments of C-10 and C-100 of the previously published spectrum in
DMSO-d6 þ D2O (25) must be reversed (Table 3). Similarly, for
the HMBC spectrum of compound 4 (C5G), a 3J correlation was
visible between H-2 at 4.59 ppm and the 13C signal at 156.71 ppm,
assigning the latter to C-9. Thus the assignments of C-7 and C-9
of the previously published spectrum in DMSO-d6 þ D2O (25)
needed to be exchanged as well (Table 3). Finally both the 1H and
the 13C NMR data of compound 4 (C5G) in MeOH-d4 were
identical to previously published data for (þ)-catechin 5-O-β-
D-
glucoside in the same solvent (Table 3) (33), except that the HSQC
and HMBC spectra of 4 demonstrated that the assignments of
C-6 and C-8 must be reversed in the previously published
spectrum (Table 3) (33).
Figure 7. Time course of the degradation of C3G (9), C30G (0), C40G
(ꢀ), C5G (2), C7G (b) in sodium phosphate buffer at pH 8, pH 7 and pH
6 compared to the degradation of catechin ([).
The synthetic route for the synthesis of catechin 3-O-β-D-
of catechin at pH 6, 7, and 8. At pH 8, a distinct increase in
stability of ∼70% was observed for C30G, while all other isomers
were slightly less stable than catechin. At pH 7, the stability was
significantly increased by glucosylation of position 30 (∼200%)
and 40, and to a lesser extent also for position 5. While C7G had
the same stability as catechin at this pH, C3G was again less
stable. At pH 6, all glucosides showed increased stability when
compared to the free form. Again C40G and C30G were the two
most stable isomers, followed by C5G.
At pH 4 and 5, no substantial decrease could be observed over
48 h at 37 °C even for (þ)-catechin. Therefore the temperature
was increased from 37 to 70 °C for these pH values. After 48 h at
pH 5, onlylessthan 10% ofcatechin remainedwhile 80% ofC40G
and 60% of C30G could still be detected, respectively (Figure 8a).
An increase in stability was also observed for C5G and C7G,
whereas the stability of C3G was comparable with that of
catechin. At pH 4, C40G was again the most stable isomer.
Surprisingly, C30G was the least stable form under these condi-
tions. Note however that the overall lossesat this pHare relatively
small (Figure 8b).
glucoside 5 is shown in Figure 6. In this case the protection of
catechin was based on the differences in reactivity between the
3-alcoholic group and all the remaining phenolic groups. Com-
pound 19 has been synthesized previously by a similar method,
but only its proton NMR spectrum was published (44). Thus this
study reports the first full NMR characterization of compound
19. Lewis acid-catalyzed reaction of compound 19 with the
glucosyl donor 9 afforded the acetylated glucoside 20 in 49%
yield. Finally alkaline deacetylation of 20gave rise tocompound 5
(C3G) in excellent yield. 3-Glucosylation of both compound 20
3
and 5 was confirmed by the presence of a strong J correlation
between H100 and C3 in their HMBC 2D NMR spectra. In the
HMBC spectrum of compound 5 (C3G), an additional correla-
tion between H3 and C100 was present, further confirming
3-glucosylation. Furthermore, although recorded in different
solvents, the 13C NMR data of compound 5 (C3G) (MeOH-d4),
and of (þ)-catechin 3-O-β- -glucoside (acetone-d6 þ D2O), pre-
D
viously isolated from the leaves Quercus miyagii (45), were similar
(Table 4). However the analysis of the HSQC and HMBC spectra
of compound 5 (C3G) suggested that the assignment of
C-6 and C-8 must be exchanged in the previously published
spectrum (Table 4).
Stability in a Physiological Buffer. The stability of the indivi-
dual catechin glucosides (15 μM) was tested in the “perfusion
buffer” (Figure 9a). For all catechin glucosides, an increase in
stability was observed compared to the aglycone. After 42 h
incubation, >60% of catechin was degraded while ∼90% of
the catechin 30-glucoside was still intact. Figure 9b shows the
increase of stability for the individual glucosides compared to the
Influence of pH on the Stability. Stabilities of catechin gluco-
sides were tested in a sodium phosphate buffer in the pH range of
4 to 8. In general, the stability of catechins is poor at neutral and
slightly basic conditions and increases with decreasing pH.
Figure 7 shows the degradation of the catechin glucosides and

Article
J. Agric. Food Chem., Vol. 58, No. 4, 2010 2147
Figure 10. Apparent absorption of catechin and catechin glucosides after
intestinal perfusion calculated as the difference between the influent
concentration and the sum of catechin and catechin glucoside concentra-
tion of the effluent. Under control conditions, all of the compounds were
>95% stable.
Figure 8. Catechin and catechin glucosides remaining after storage in
sodium phosphate buffer at pH 5 (a) or pH 4 (b) at 70 °C for 48 h.
Figure 11. Degree of intestinal hydrolysis of the different catechin gluco-
sides during the perfusion experiment calculated from the catechin/
catechin glucoside ratios in the effluent samples as determined by HPLC
analysis.
that, during the relatively short time of perfusion, all the com-
pounds tested were >95% stable.
Intestinal Perfusion of Catechin Glucosides. The gut lumen
metabolism of the catechin glucosides was assessed and compared
with that of catechin in a rat intestinal perfusion study. The
“absorption” (difference between influent and effluent con-
centrations) for C30G (28%) was significantly higher (p =
0.005) than for catechin (16%). For C3G, C40G and C7G, a
significant decrease in “absorption” was observed as compared to
the aglycone, and the percentage absorption for C5G was un-
changed (Figure 10). If the results for “absorption” are compared
with the degree of intestinal hydrolysis of the catechin glucosides
(Figure 11), there is a direct correlation. The two glucosides (C30G
and C7G) which were intensively hydrolyzed are apparently
better “absorbed” than the glucosides which stay intact during
intestinal transit. This shows that the intestinal enzyme respon-
sible for the hydrolysis, probably lactase phloridzin hydrolase,
seems to be highly regioselective for hydrolysis of the catechin
glucosides. This is different fromquercetin (12) where the position
of sugar attachment did not influence absorption of two quercetin
4- and 30-glucosides in human volunteers, since the specificity of
brush border lactase phlorizin hydrolase is similar for both
glucosides (14).
Figure 9. (a) Time course of the degradation of C3G (9), C30G (0),
C40G (ꢀ), C5G (2), C7G (b) compared to the degradation of catechin
([) in perfusion buffer (pH 7) at 37 °C for 42 h.
(b) Percentage of the increase in stability by glucosylation based on the
data at 42 h.
aglycone based on the degradation observed after 42 h. While
most glucosides are approximately twice as stable as catechins,
there is a 5-fold increase of stability for C30G. These data show

2148 J. Agric. Food Chem., Vol. 58, No. 4, 2010
Raab et al.
ABBREVIATIONS USED
(15) Miyase, T. Preparation of catechin derivatives with increased
stability. JP Patent A2 2001253879.
Ac, acetyl; Bn, benzyl; BuOH, butanol; CH2Cl2, dichloro-
(16) Lam, W. H.; Kazi, A.; Kuhn, D. J.; Chow Larry, M. C.; Chan Albert,
S. C.; Dou, Q. P.; Chan, T. H. A potential prodrug for a green tea
polyphenol proteasome inhibitor: evaluation of the peracetate ester
of (-)-epigallocatechin gallate [(-)-EGCG]. Bioorg. Med. Chem.
2004, 12, 5587–5593.
(17) Lopez-Toledano, A.; Mayen, M.; Merida, J.; Medina, M. Yeast-
induced inhibition of (þ)-catechin and (-)-epicatechin degradation
in model solutions. J. Agric. Food Chem. 2002, 50, 1631–1635.
(18) Kim, H. J.; Yeom, S. H.; Kim, M. K.; Shim, J. G.; Lim, H. W.; Lee,
M. W. Nitric oxide and prostaglandin E2 synthesis inhibitory
activities of flavonoids from the barks of Ulmus macrocarpa. Nat.
Prod. Sci. 2004, 10, 344–346.
(19) Friedrich, W.; Galensa, R. Identification of a new flavanol glucoside
from barley (Hordeum vulgare L.) and malt. Eur. Food Res. Technol.
2002, 214, 388–393.
(20) Tanaka, T.; Kataoka, M.; Tsuboi, N.; Kouno, I. New monoterpene
glycoside esters and phenolic constituents of Paeoniae Radix, and
increase of water solubility of proanthocyanidins in the presence of
paeoniflorin. Chem. Pharm. Bull. 2000, 48, 201–207.
methane; C7G, (þ)-catechin 7-O-β-
D-glucoside (compound 1);
C30G, (þ)-catechin 30-O-β-
D
-glucoside (compound 2); C40G, (þ)-
catechin 40-O-β-
D
-glucoside (compound 3); C5G, (þ)-catechin 5-
O-β-
D
-glucoside (compound 4); C3G, (þ)-catechin 3-O-β-
D-glu-
coside (compound 5); calcd, calculated; cpd, compound; DBU,
1,8-diazabicyclo[5.4.0]undec-7-ene; DCM, dichloromethane;
DMAP, 4-dimethylaminopyridine; DMF, dimethylformamide;
DMSO, dimethyl sulfoxide; EDCI, 1-(3-dimethylaminopropyl)-
3-ethylcarbodiimide hydrochloride; EtOAc, ethyl acetate;
HCOOH, formic acid; iPr2Net, N,N-diisopropylethylamine;
MeCN, acetonitrile; Me2CO, acetone; NMP, 1-methyl-2-pyrro-
lidone; NOE, nuclear Overhauser effect; MPLC, medium
pressure liquid chromatography; MOM, methoxymethyl; rt,
room temperature; THF, tetrahydrofuran; VLC, vacuum liquid
chromatography.
LITERATURE CITED
(21) Watanabe, M. Catechins as antioxidants from Buckwheat
(Fagopyrum Esculentum Moench) groats. J. Agric. Food Chem.
1998, 46, 839–845.
(22) Lee, M. W.; Tanaka, T.; Nonaka, G. I.; Hahn, D. R. Phenolic
compounds of the leaves of Betula platyphylla var. latifolia. Arch.
Pharm. Res. 1992, 15, 211–214.
(23) Foo, L. Y.; Karchesy, J. J. Polyphenolic glycosides from Douglas fir
inner bark. Phytochemistry 1989, 28, 1237–1240.
(24) Kashiwada, Y.; Nonaka, G.; Nishioka, I. Tannins and related
compounds. XLV. Rhubarb. (5). Isolation and characterization of
flavan-3-ol and procyanidin glucosides. Chem. Pharm. Bull. 1986, 34,
3208–3222.
(25) Nonaka, G. I.; Ezaki, E.; Hayashi, K.; Nishioka, I. Tannins and
related compounds. Part 11. Flavanol glucosides from rhubarb and
Rhaphiolepis umbellata. Phytochemistry 1983, 22, 1659–1661.
(26) Nonaka, G.; Nishizawa, M.; Yamagishi, T. Chemical components
of rhubarb (Oriental drug). Wakan Iyaku Gakkaishi 1985, 2,
89–92.
(1) Su, Y. L.; Leung, L. K.; Huang, Y.; Chen, Z. Y. Stability of tea
theaflavins and catechins. Food Chem. 2003, 83, 189–195.
(2) Higdon Jane, V.; Frei, B. Tea catechins and polyphenols: health
effects, metabolism, and antioxidant functions. Crit. Rev. Food Sci.
Nutr. 2003, 43, 89–143.
(3) Manach, C.; Williamson, G.; Morand, C.; Scalbert, A.; Remesy, C.
Bioavailability and bioefficacy of polyphenols in humans. I. Review
of 97 bioavailability studies. Am. J. Clin. Nutr. 2005, 81, 230S–242S.
(4) Donovan, J. L.; Crespy, V.; Manach, C.; Morand, C.; Besson, C.;
Scalbert, A.; Remesy, C. Catechin is metabolized by both the small
intestine and liver of rats. J. Nutr. 2001, 131, 1753–1757.
(5) Crespy, V.; Morand, C.; Manach, C.; Besson, C.; Demigne, C.;
Remesy, C. Part of quercetin absorbed in the small intestine is
conjugated and further secreted in the intestinal lumen. Am. J.
Physiol. 1999, 277, G120–G126.
(6) Kuhnle, G.; Spencer, J. P.; Schroeter, H.; Shenoy, B.; Debnam, E. S.;
Srai, S. K.; Rice-Evans, C.; Hahn, U. Epicatechin and catechin are
O-methylated and glucuronidated in the small intestine. Biochem.
Biophys. Res. Commun. 2000, 277, 507–512.
(7) Yang, C. S.; Chen, L. S.; Lee, M. J.; Balentine, D.; Kuo, M. C.;
Schantz, S. P. Blood and urine levels of tea catechins after ingestion
of different amounts of green tea by human volunteers. Cancer
Epidem. Biomarker. Prev. 1998, 7, 351–354.
(8) van het Hof, K. H.; Kivits, G. A. A.; Weststrate, J. A.; Tijburg, L. B.
M. Bioavailability of catechins from tea: the effect of milk. Eur. J.
Clin. Nutr. 1998, 52, 356–359.
(9) Chow, H. H.; Cai, Y.; Alberts, D. S.; Hakim, I.; Dorr, R.; Shahi, F.;
Crowell, J. A.; Yang, C. S.; Hara, Y. Phase I pharmacokinetic study
of tea polyphenols following single-dose administration of epigallo-
catechin gallate and polyphenon E. Cancer Epidemiol., Biomarkers
Prev. 2001, 10, 53–58.
(10) Zhu, Q. Y.; Holt, R. R.; Lazarus, S. A.; Ensunsa, J. L.; Hammerstone,
J. F.; Schmitz, H. H.; Keen, C. L. Stability of the flavan-3-ols
epicatechin and catechin and related dimeric procyanidins derived
from cocoa. J. Agric. Food Chem. 2002, 50, 1700–1705.
(11) Hollman, P. C.; van Trijp, J. M.; Mengelers, M. J.; de Vries, J. H.;
Katan, M. B. Bioavailability of the dietary antioxidant flavonol
quercetin in man. Cancer Lett. 1997, 114, 139–140.
(27) Kitagawa, I.; Wang, H. K.; Saito, M.; Yoshikawa, M. Saponin and
sapogenol. XXXI. Chemical constituents of the seeds of Vigna
angularis (Willd.) Ohwi et Ohashi. (1). Triterpenoidal sapogenols
and 3-furanmethanol β-D-glucopyranoside. Chem. Pharm. Bull.
1983, 31, 664–673.
(28) Takani, M.; Ohya, K.; Takahashi, K. Studies on constituents of
medicinal plants. XXII. Constituents of Schizandra nigra Max. (4).
Chem. Pharm. Bull. 1979, 27, 1422–1425.
(29) Pan, H.; Lundgren, L. N. Phenolics from inner bark of Pinus
sylvestris. Phytochemistry 1996, 42, 1185–1189.
(30) Bae, Y. S.; Burger, J. F. W.; Steynberg, J. P.; Ferreira, D.; Hemingway,
R. W. Flavan and procyanidin glycosides from the bark of blackjack
oak. Phytochemistry 1994, 35, 473–478.
(31) Duenas, M.; Sun, B.; Hernandez, T.; Estrella, I.; Spranger, M. I.
Proanthocyanidin composition in the seed coat of lentils. (Lens
culinaris L.). J. Agric. Food Chem. 2003, 51, 7999–8004.
(32) Funayama, M.; Nishino, T.; Hirota, A.; Murao, S.; Takenishi, S.;
Nakano, H. Enzymic synthesis of (þ)catechin-a-glucoside and its
effect on tyrosinase activity. Biosci., Biotechnol. Biochem. 1993, 57,
1666–1669.
(33) Otani, S.; Kondo, Y.; Asada, Y.; Furuya, T.; Hamada, H.; Nakajima,
N.; Ishihara, K.; Hamada, H. Biotransformation of (þ)-catechin by
plant cultured cells of Eucalyptus perriniana. Plant Biotechnol. 2004, 21,
407–409.
(12) Olthof, M. R.; Hollman, P. C.; Vree, T. B.; Katan, M. B. Bioavail-
abilities of quercetin-3-glucoside and quercetin-40-glucoside do not
differ in humans. J. Nutr. 2000, 130, 1200–1203.
(13) Day, A. J.; DuPont, M. S.; Ridley, S.; Rhodes, M.; Rhodes, M. J. C.;
Morgan, M. R. A.; Williamson, G. Deglycosylation of flavonoid and
isoflavonoid glycosides by human small intestine and liver β-gluco-
sidase activity. FEBS Lett. 1998, 436, 71–75.
(14) Day, A. J.; Canada, F. J.; Diaz, J. C.; Kroon, P. A.; McLauchlan, W. R.;
Faulds, C. B.; Plumb, G. W.; Morgan, M. R. A.; Williamson, G. Dietary
flavonoid and isoflavone glycosides are hydrolysed by the lactase site of
lactase phlorizin hydrolase. FEBS Lett. 2000, 468, 166–170.
(34) Weinges, K.; Seiler, D. Phenolic natural substances. IX. Diastereo-
meric catechin 3-glucosides and 3-gallates. Justus Liebigs Ann. Chem.
1968, 714, 193–204.
(35) Tamura, K.; Mizukami, H.; Maeda, K.; Watanabe, H.; Uneyama,
K. One-pot synthesis of trifluoroacetimidoyl halides. J. Org. Chem.
1993, 58, 32–35.
(36) Nudelman, A.; Herzig, J.; Gottlieb, H. E.; Keinan, E.; Sterling, J.
Studies in sugar chemistry. Part IV. Selective deacetylation of

Article
J. Agric. Food Chem., Vol. 58, No. 4, 2010 2149
anomeric sugar acetates with tin alkoxides. Carbohydr. Res. 1987,
162, 145–152.
(42) Chattopadhyay, S. K.; Banerjee, S.; Agarwal, S.; Sashidhara, K. V.;
Tripathi, V.; Kukreja, A. K.; Kumar, S.; Kulshrestha, M.; Sharma,
R. P.; Mehta, V. K. Process for the production of a compound (þ)
catechin pentaacetate useful as a precursor for the production of (þ)
catechin. US Patent 6,365,757, March 28, 2000.
(37) Lambusta, D.; Nicolosi, G.; Patti, A.; Piatelli, M. Enzyme-mediated
regioprotection-deprotection of hydroxyl groups in (þ)-catechin.
Synthesis 1993, 11, 1155–1158.
(43) Basak, A.; Mandal, S.; Bandyopadhyay, S. Regioselective hydrolysis
of pentaacetyl catechin and epicatechin by porcine liver esterase.
Bioorg. Med. Chem. Lett. 2003, 13, 1083–1085.
(44) Patti, A.; Piattelli, M.; Nicolosi, G. Use of Mucor miehei lipase in the
preparation of long chain 3-O-acylcatechins. J. Mol. Catal. B:
Enzym. 2000, 10, 577–582.
(45) Ishimaru, K.; Nonaka, G.; Nishioka, I. Tannins and related com-
pounds. Part 53. Flavan-3-ol and procyanidin glycosides from
Quercus miyagii. Phytochemistry 1987, 26, 1167–1170.
(38) Donovan, J. L.; Crespy, V.; Oliveria, M.; Cooper, K. A.; Gibson,
B. B.; Williamson, G. (þ)-catechin is more bioavailable than (-)-
catechin: Relevance to the bioavailability of catechin from cocoa.
Free Radical Res. 2006, 40, 1029–1034.
(39) Pan, C. Y.; Kao, Y. H.; Fox, A. P. Enhancement of inward Ca(2þ)
currents in bovine chromaffin cells by green tea polyphenol extracts.
Neurochem. Int. 2002, 40, 131–137.
(40) Li, M.; Han, X. W.; Yu, B. Facile synthesis of flavonoid 7-O-
glycosides. J. Org. Chem. 2003, 68, 6842–6845.
(41) Cren-Olive, C.; Wieruszeski, J. M.; Maes, E.; Rolando, C. Catechin
and epicatechin deprotonation followed by 13C NMR. Tetrahedron
Lett. 2002, 43, 4545–4549.
Received for review September 29, 2009. Revised manuscript received
December 23, 2009. Accepted January 14, 2010.