Chemical synthesis and characterization of epicatechin glucuronides and sulfates: Bioanalytical standards for epicatechin metabolite identification

Article abstract of DOI:10.1021/np300568m

The monoglucuronides and sulfates of epicatechin, 3′- Omethylepicatechin, and 4′-O-methylepicatechin, respectively, were synthesized as authentic bioanalytical standards. Reversed-phase HPLC methods capable of baseline separation of the glucuronides and sulfates have been developed. Both the epicatechin glucuronides and sulfates were stable in the solid state when stored under ambient conditions and in aqueous solution when stored refrigerated. These results should prove invaluable to the research community as analytical standards as well as in future studies of the biological and pharmacological effects of epicatechin in humans.

Full text of DOI:10.1021/np300568m

Article
pubs.acs.org/jnp
Chemical Synthesis and Characterization of Epicatechin
Glucuronides and Sulfates: Bioanalytical Standards for Epicatechin
Metabolite Identification
Mingbao Zhang,*,† G. Erik Jagdmann, Jr.,† Michael Van Zandt,† Ryan Sheeler,† Paul Beckett,†
and Hagen Schroeter‡
†The Institutes for Pharmaceutical Discovery, 23 Business Park Drive, Branford, Connecticut 06405, United States
‡MARS Symbioscience, 20425 Seneca Meadows Parkway, Germantown, Maryland 20876, United States
S
* Supporting Information
ABSTRACT: The monoglucuronides and sulfates of epicatechin, 3′-O-
methylepicatechin, and 4′-O-methylepicatechin, respectively, were
synthesized as authentic bioanalytical standards. Reversed-phase
HPLC methods capable of baseline separation of the glucuronides
and sulfates have been developed. Both the epicatechin glucuronides and
sulfates were stable in the solid state when stored under ambient
conditions and in aqueous solution when stored refrigerated. These
results should prove invaluable to the research community as analytical
standards as well as in future studies of the biological and
pharmacological effects of epicatechin in humans.
picatechin and its C-2 epimer catechin are members of the
In order to accurately assess the systemic levels of
epicatechin in humans, the chemical structures of the
metabolites must be elucidated and their individual systemic
exposures determined. Furthermore, some of the metabolites
may themselves be biologically active and, thus, responsible for
some of the benefits observed after ingestion of epicatechin.22,23
In this context, authentic standards of various epicatechin
metabolites prepared in a chemically unambiguous fashion are
essential. In the past, and due to the absence of such authentic
standards, the structure elucidation of epicatechin metabolites
has mainly relied on various NMR techniques such as COSY,
HSQC, HMBC, and NOE. However, since the structurally
related epicatechin metabolites are mainly regioisomers, NMR
spectroscopic techniques alone do not always yield unequivocal
elucidation of their structures due to weak correlations.24 In
addition, the availability of authentic flavan-3-ol metabolite
standards would greatly enhance the development of sample
preparation and analytical methods.14 Various attempts have
been made to synthesize authentic standards of some of the
metabolites either enzymatically25 or chemically26,27 from
epicatechin. However, due to the nonregioselective nature of
these earlier syntheses, mixtures of regioisomers are unavoid-
ably obtained, which further requires labor-intensive chromato-
graphic separations, an approach that is intrinsically ambiguous,
especially when considering the absence of authentic standards
for method validation in the first place. Ironically, so-called
1
Eflavan-3-ol family.
Together with their oligomeric
derivatives, the procyanidins, flavan-3-ols are found in many
plants that are part of the human diet.2 There is accumulating
scientific evidence that diets rich in flavan-3-ols may provide
important health benefits, including a reduced risk of
cardiovascular disease.3−10 Contrary to a widespread assump-
tion, procyanidins do not contribute directly to the systemic
pool of flavan-3-ols in humans.11 Furthermore, the absorption,
metabolism, and biological activity of dietary flavan-3-ols are
strongly influenced by stereochemistry, and the predominant
effects are attributed to epicatechin and its metabolites.12
Indeed, a recent study has shown that oral administration of
epicatechin to human subjects resulted in vascular effects
similar to those observed with flavan-3-ol-rich cocoa, thus
providing evidence for a causality chain directly linking
epicatechin to the reported vascular effects observed after
consumption of flavan-3-ols.9
Consequently, there has been a great deal of interest in
elucidating the full metabolic profile of epicatechin after human
ingestion.13−15 It has been documented that ingested
epicatechin undergoes rapid phase-II metabolism to form a
wide range of metabolites including structurally related
epicatechin metabolites (SREM) containing an intact C-ring
and seco-metabolites from the breakdown of epicatechin by the
gut microbiome.14,16,17 It is further established that ingested
epicatechin is extensively metabolized into SREM by O-
methylation, sulfation, O-glucuronidation, and combinations
thereof.14,18−21
Received: August 20, 2012
Published: January 28, 2013
© 2013 American Chemical Society and
American Society of Pharmacognosy
157
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
Chart 1. Structures of Orthogonally Protected Epicatechin Intermediates 1a−10a (red), the Corresponding O-β-D-Glucuronides
1G−10G (green), and Sulfates 1S−10S (blue)
“authentic standards” obtained in this fashion are further prone
to structural ambiguities, because the site(s) of the methylation,
glucuronidation, and sulfation cannot always be readily
ascertained.
remaining hydroxy groups are protected as benzyl ethers. This
strategy allows the regiospecific glucuronidation/sulfation of a
single hydroxy group while the rest remain protected. With this
synthetic approach, the site of glucuronidation/sulfation of
epicatechin is unambiguous. Using these chemically unambig-
uous authentic standards of epicatechin glucuronides and
sulfates, a method to quantify epicatechin metabolites in human
plasma after ingestion of epicatechin has been successfully
developed.14
In a recent preliminary communication, some of us have
reported the preparation of the orthogonally protected
epicatechin intermediates (4a, 7a−10a) and the final synthesis
of five O-β-D-glucuronides (4G, 7G−10G, Chart 1) derived
from epicatechin, 3′-O-methylepicatechin, and 4′-O-methylepi-
catechin, respectively.30 In the present paper, we advance on
previous work and describe the chemical synthesis of the
remaining five monoglucuronides, 1G−3G, 5G, and 6G, and all
10 monosulfates 1S−10S (Chart 1) from the orthogonally
protected epicatechin intermediates 1a−10a. Also presented in
Recently, we have focused our attention on the chemical
synthesis of all theoretically possible monoglucuronides and
sulfates28 of epicatechin, 3′-O-methyl-, and 4′-O-methylepica-
techin as authentic bioanalytical standards (see also ref 29).
The advantage of this approach is that not only are we making
NMR assignments to a predetermined structure, but we are also
able to compare assignments directly with all of the other
possible isomers and show that they are distinct and that the
data are internally consistent. The key to our synthetic strategy
is the availability of a set of orthogonally protected epicatechin
intermediates, 1a−10a (Chart 1), prepared de novo. In this
approach, the specific phenolic hydroxy group to be
glucuronidated or converted to the sulfate group is selectively
protected with a methoxymethyl (MOM) or p-methoxybenzyl
(PMB) group (highlighted in red in Chart 1), while all of the
158
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
a
Scheme 1. Synthesis of Intermediates 1c−10c
a
Reagents and conditions: (i) NaH, BnBr, DMF, 0 °C; (ii) 4 N HCl in dioxane, DCM/MeOH, room temperature for 1c and 3c−10c; (iii)
pyridinium p-toluenesulfonate, 2-propanol/toluene, reflux for 2c.
Scheme 2. Epimerization at C-2 during the Removal of the MOM Protective Group in 5b
a
Scheme 3. Synthesis of Epicatechin-5-O-β-D-glucuronide (3G)
a
Reagents and conditions: (i) BF3·OEt2, DCM, 0 °C; (ii) 1 N NaOH, THF/MeOH, 0 °C; (iii) Pd(OH)2/C, H2, MeOH.
1
this paper are complete assignments of H and 13C NMR
signals of 1G−10G and 1S−10S. HPLC methods capable of
baseline separation of 1G−10G and 1S−10S are also reported,
together with assessments of their stability in both the solid
state and aqueous solution.
orthogonally protected epicatechin intermediates, 1a−10a
(Chart 1), described elsewhere.31 Protection of the 3-OH
group by benzylation afforded the orthogonally protected
epicatechin intermediates 1b−10b. The phenolic OH group to
be glucuronidated was then selectively unmasked by the
removal of the MOM or the PMB group. The MOM group was
removed by treatment with 4 N HCl/dioxane in DCM at room
temperature. Under these conditions, intermediate 5b afforded
a 1:1 mixture of epicatechin 5c and ent-catechin 5c′ (Scheme
2), while the rest of the intermediates bearing an MOM group
RESULTS AND DISCUSSION
■
Epicatechin-O-β-D-glucuronides. Chemical Synthesis.
The synthesis shown in Scheme 1 relies on a series of
159
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
Table 1. NMR Spectroscopic Data for Epicatechin Glucuronides 1G−4G in Acetone-d6 Unless Specified Otherwise
3′-glucuronide 1G
δH (J in Hz)
4′-glucuronide 2G
δH (J in Hz)
5-glucuronide 3G
7-glucuronide 4G
δH (J in Hz)
a
b
a
a
c
position
δC
δC
δC
δH (J in Hz)
δC
2
3
4
78.9
66.4
28.7
5.04, s
79.2
66.8
29.1
4.94, s
79.2
66.3
28.6
4.84, s
79.6
66.7
29.1
4.90, s
4.32, m
4.25, br, s
4.19, br, s
4.22, m
2.92, dd (16.8, 4.4)
2.72, dd (17.0, 2.8)
2.88, d (16.5)
2.75, d (16.5)
2.88, d (16.8)
2.81, d (16.8)
2.89, dd (16.8, 4.6)
2.76, d (16.8, 3.1)
5
157.4
96.2
157.6
96.3
157.5
96.3
157.5
97.5
6
6.10, d (2.4)
6.08, d (2.4)
6.03, s
5.94, s
6.25, s
6.07, s
6.21, d (2.3)
6.16, d (2.3)
7
157.2
95.4
157.6
95.8
156.5
98.0
158.0
96.9
8
9
156.6
99.5
157.0
99.8
157.2
102.1
131.6
115.0
145.0
145.2
115.5
119.0
101.6
73.8
157.1
102.9
132.1
115.3
145.4
145.4
115.5
119.5
102.0
74.4
10
1′
2′
3′
4′
5′
6′
1″
2″
3″
4″
5″
COOH
131.9
117.8
145.4
147.5
116.4
123.3
103.7
73.8
137.2
116.0
148.6
145.3
119.3
119.0
104.8
74.4
7.26, d (1.9)
7.10, s
7.00, s
7.06, d (1.8)
6.99, d (8.3)
7.11, dd (8.6, 1.9)
5.11, d (7.4)
3.62, m
7.13, d (8.0)
6.94, d (8.0)
4.89, d (7.3)
3.58, m
6.75, d (8.6)
6.79, d (8.6)
4.99, d (7.2)
3.54, t (8.6)
3.58, t (8.7)
3.66, t (8.6)
6.79, d (8.1)
6.84, dd (8.1, 1.8)
5.01, d (7.7)
3.48, t (7.9)
3.58, t (9.0)
3.70, t (9.0)
4.07, d (9.7)
76.3
3.63, m
76.8
3.62, m
76.6
77.2
72.3
3.67, m
72.6
3.72, t (9.2)
4.08. d (9.6)
72.2
72.7
d
76.3
3.96, d (9.6)
76.2
75.9
4.04
76.1
e
170.9
170.2
171.1
170.2
a
b
c
d
In acetone-d6 containing 10% D2O. In D2O. The two phenolic OHs were visible as broad singlets (δ 7.8 and 8.4). Overlapping with H2O peak.
e
Observed in H−C HMBC spectrum (correlation with H-5″).
Table 2. NMR Spectroscopic Data for 3′-O-Methylepicatechin Glucuronides 5G−7G in Acetone-d6 Unless Specified Otherwise
3′-OMe-4′-glucuronide 5G
δH (J in Hz)
3′-OMe-5-glucuronide 6G
3′-OMe-7-glucuronide 7G
δH (J in Hz)
a
a
b
position
δC
δC
δH (J in Hz)
δC
2
3
4
78.0
65.7
27.4
4.83, s
79.6
66.8
29.2
4.92, s
79.7
66.8
29.3
4.95, s
4.27, br, s
4.21, br, s
4.25, m
2.85, dd (17.1, 4.3)
2.73, d (17.1)
2.93, d (16.8)
2.85, d (16.8)
2.91, dd (16.8, 4.5)
2.80, dd (16.8, 2.8)
5
155.6
96.0
158.0
96.6
157.5
97.6
6
6.08, d (2.3)
6.07, d (2.3)
6.30, s
6.09, s
6.22, d (2.3)
6.18, d (2.3)
7
155.3
95.5
157.6
98.3
158.0
97.1
8
9
155.1
99.7
156.8
102.5
131.9
111.8
147.9
146.9
115.3
120.6
102.2
74.4
157.1
102.9
131.9
111.9
147.9
146.9
115.2
120.6
102.1
74.4
10
1′
2′
3′
4′
134.0
110.9
148.6
144.9
116.1
119.1
100.6
72.6
7.08, s
7.17, s
7.19, d (1.8)
5′
6′
7.13, d (8.4)
6.99, d (8.4)
5.09, d (7.1)
3.64, m
6.80, d (7.8)
6.96, d (7.8)
5.01, d (7.4)
3.56, t (8.5)
3.60, t (8.8)
3.72, t (9.2)
4.07, d (9.6)
6.81, d (8.1)
6.97, dd (8.1, 1.8)
5.01, d (7.7)
3.48, t (7.8)
3.58, t (9.0)
3.70, t (9.0)
1″
2″
3″
4″
5″
COOH
75.2
3.64, m
77.3
77.2
71.3
3.64, m
72.7
72.7
4.04, d (8.3)
76.1
76.1
170.2
56.3
170.2
56.3
OMe
55.9
3.83, s
3.83, s
3.84, s
a
b
In D2O. The two phenolic OHs were visible as broad singlets (δ 7.5 and 8.4).
(1a, 3a, 4a, and 6a−10a) afforded the epicatechin products
without significant ent-catechin byproducts. The formation of
the ent-catechin byproduct 5c′ was the result of the 4′-OH-
induced C-2 epimerization of epicatechin.32 In order to avoid a
similar acid-catalyzed epimerization in the synthesis of
epicatechin-4′-O-β-D-glucuronide (2G), removal of the PMB
group in 2b was achieved by treatment with pyridinium p-
toluenesulfonate in 2-propanol/toluene at reflux. Under these
conditions, a smaller amount (ca. 17%) of the ent-catechin was
1
detected by H NMR. The ent-catechin byproducts are readily
160
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
Table 3. NMR Spectroscopic Data for 4′-O-Methylepicatechin Glucuronides 8G−10G in Acetone-d6
4′-OMe-3′-glucuronide 8G
δH (J in Hz)
4′-OMe-5-glucuronide 9G
4′-OMe-7-glucuronide 10G
position
δC
δC
δH (J in Hz)
δC
δH (J in Hz)
2
3
4
79.2
66.6
29.1
4.94, s
79.4
66.7
29.0
4.92, s
79.5
4.94, s
4.25, br, s
4.21, br, s
66.7
29.2
4.25, br, s
a
2.86, dd (16.5, 4.1)
2.76, d (16.5)
2.88, m
2.91, d (16.8)
2.78, d (16.8)
a
2.88, m
5
157.5
96.2
157.9
96.5
157.5
97.6
6
6.03, d (2.2)
5.93, d (2.2)
6.29, s
6.09, s
6.21, s
6.18, s
7
157.5
95.8
157.6
98.1
158.0
96.9
8
9
157.0
99.7
156.8
102.5
133.5
114.9
146.9
147.7
112.0
118.9
102.1
74.3
157.0
102.9
133.5
115.1
147.0
147.8
112.0
119.0
102.0
74.4
10
1′
2′
3′
4′
5′
6′
1″
2″
3″
4″
5″
COOH
OMe
133.4
118.1
147.0
150.5
113.3
122.8
102.9
74.5
7.39, s
7.06, s
7.07, s
6.99, d (8.2)
7.20, d (8.5)
5.09, d (7.3)
3.54, t (8.8)
3.57, t (9.3)
3.71, t (9.2)
3.99, d (9.5)
6.94, d (8.5)
6.91, d (8.1)
5.02, d (7.2)
3.55, t (8.7)
3.59, t (9.2)
3.72, t (8.5)
4.08, d (9.6)
6.95, d (8.2)
6.91, d (8.0)
5.02, d (7.3)
3.49, t (7.9)
3.59, t (8.8)
3.70, t (9.0)
4.07, d (9.5)
77.1
77.2
77.2
72.5
72.6
72.7
b
b
76.1
76.1
76.1
b
bc
,
170.1
56.6
170.1
56.3
170.2
3.84, s
3.84, s
56.4
3.84, s
a
b
c
Overlapping signals. Not observable in 13C NMR when ca.10% D2O was present. Observed in H−C HMBC spectrum (correlation with H-5″) in
acetone-d6 with 10% D2O.
1
distinguishable from the corresponding epicatechin precursors
NMR Characterization. The H and 13C NMR spectra of
1
3
the epicatechin glucuronides 1G−10G were unambiguously
assigned using COSY, HSQC, and HMBC 2D NMR methods,
and the data are collated in Tables 1−3.
by H NMR based on the J2,3 coupling constants of 8.3 Hz.
The phenolic intermediates 1c−10c were glucuronidated by
coupling with methyl 2,3,4-tri-O-acetyl-1-O-(trichloroacetimi-
doyl)-α-D-glucuronate. For clarity, only the synthesis of
epicatechin-5-O-β-D-glucuronide (3G) is shown in Scheme 3.
The glucuronidation of the 5-OH group of epicatechin 3c was
carried out under known reaction conditions (BF3 etherate, 4 Å
molecular sieves, DCM, 0 °C).33,34 BF3 etherate was dissolved
in anhydrous DCM and added to the reaction slowly. Although
this coupling reaction ran smoothly most of the time, the
manner in which the Lewis acid was added is important.35
The coupling product 3d was purified by flash chromatog-
raphy (silica gel, heptane/DCM/EtOAc, 5:4:1, v/v/v). Alkaline
ester hydrolysis/O-deacetylation followed by hydrogenolysis to
remove the benzyl protecting groups afforded the crude
glucuronide, which was purified by preparative reversed-phase
HPLC to give epicatechin-5-O-β-D-glucuronide (3G). Com-
pound 3G was obtained as a white powder in 59% yield after
lyophilization.
The synthesis of 5G was carried out with the mixture of 5c
and 5c′ obtained from the removal of the MOM protective
group shown in Scheme 2 in the same fashion as described in
Scheme 3. The crude product was isolated as a 1:1 mixture of
3′-O-methylepicatechin-4′-O-β-D-glucuronide (5G) and 3′-O-
methyl-ent-catechin-4′-O-β-D-glucuronide (5G′) after unmask-
ing all the OH groups. Pure 5G and the byproduct 5G′36 were
obtained after separation by preparative reversed-phase HPLC.
Compounds 1G, 2G, 4G, and 6G−10G were prepared in a
similar fashion to that described for 3G starting from the
corresponding orthogonally protected epicatechin intermedi-
ates 1a, 2a, 4a, and 6a−10a, respectively.
1H NMR Assignments. The 1H NMR spectra of the
epicatechin glucuronides exhibit three key features: (a) H-2
appears as a singlet ca. δ 4.9, indicating that it is in the syn
relationship with H-3 (δ 4.3). H-3 is coupled to H-4a and in
some cases to H-4b as well with very small coupling constants
(ca. 3−4 Hz), thus appearing as partially resolved multiplets or
broad singlets; (b) the anomeric H-1″ of the glucuronides
appears at ca. δ 5.0 as a doublet with a coupling constant
around 7 Hz, characteristic of the β-anomer;33,34 (c) the two C-
4 protons H-4a and H-4b are diastereotopic and generally
appear as AB quartets around 2.7−2.9 ppm with JAB between 16
and 17 Hz and ΔνAB between 25 and 50 Hz (reported as two
distinct doublets in Tables 1−3 for convenience). There are
two exceptions however. In epicatechin-3′-O-β-D-glucuronide
(1G), H-4a and H-4b exhibit an AX splitting pattern with a
similar coupling constant (ca. 17 Hz); in 4′-O-methylepica-
techin-5-O-β-D-glucuronide (9G), H-4a and H-4b appear as an
unresolved multiplet centered around δ 2.88. Complete
assignments of the protons of the B-ring, C-ring, and the
glucuronic acid unit were accomplished using 2D-COSY. The
two A-ring protons H-6 and H-8 were assigned on the basis of
analogy with those of epicatechin, which have been assigned
unambiguously according to NOE difference spectroscopy.37,38
When one of the epicatechin phenolic hydroxy groups is
glucuronidated, significant chemical shift changes are observed
only for those protons in the ortho- and para-positions. Little
change in chemical shift is observed for protons in the meta-
position or those on the non-glucuronidated aromatic ring
(Table 4). The effects are generally larger on the ortho protons
161
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
XDB-Phenyl analytical column. The Eclipse XDB-Phenyl
column was chosen because of its ability to handle mobile
phases with aqueous content as high as 100%. With a mobile
phase consisting of 25 mM Na3PO4 buffer pH 2.5 and MeCN,
the epicatechin glucuronides exhibited adequate retention on
the Eclipse XDB-Phenyl column. Shown in Figure 1 are HPLC
Table 4. Changes in Chemical Shifts (Δδ, ppm) of the A-
and B-Ring Protons of Epicatechin Glucuronides 1G−4G
a
Relative to Epicatechin
H-6
H-8
H-2′
H-5′
H-6′
b
c
3′-glucuronide 1G
4′-glucuronide 2G
5-glucuronide 3G
0.00
0.01
0.24
0.19
0.01
0.02
0.16
0.24
0.34
0.050
−0.01
0.06
0.34
0.00
0.00
0.29
0.10
0.00
0.00
b
c
7-glucuronide 4G
0.01
a
1
Chemical shift changes are calculated using H chemical shifts of
b
epicatechin in the same solvent. In acetone-d6 containing 10% D2O.
c
In acetone-d6.
than the para ones. For example, when comparing epicatechin-
3′-O-β-D-glucuronide (1G) to the parent, epicatechin, H-2′ and
H-6′ were downfield shifted by 0.34 and 0.29 ppm, respectively,
while the chemical shifts of the other aromatic protons
remained virtually unchanged. For epicatechin-4′-O-β-D-glucur-
onide (2G), H-5′ showed a large downfield shift of 0.34 ppm.
For epicatechin-5-O-β-D-glucuronide (3G), H-6 showed a
downfield shift of 0.24 ppm. The proton H-8 was also
downfield shifted, but to a lesser degree (0.16 ppm). For
epicatechin-7-O-β-D-glucuronide (4G), H-6 and H-8 under-
went a downfield shift by a similar degree (0.19 and 0.24 ppm,
respectively). No significant changes were observed for the
three B-ring protons.
13C NMR Assignments. The assignments of 13C NMR
spectra for proton-bearing carbons were achieved using HSQC.
The anomeric carbon of the glucuronic acid motif appears
around δ 102 ppm, consistent with literature data.33,34 The
assignments of the A- and B-ring ipso carbons C-5, C-7, C-9, C-
10, C-1′, C-3′, and C-4′ were accomplished using HMBC
spectroscopy. The ipso carbon to which the OGluA unit is
attached was assigned on the basis of its three-bond long-range
coupling with the anomeric H-1″. The B-ring ipso carbons C-1′
and C-3′ were assigned on the basis of their three-bond
coupling with H-5′, and C-4′ was assigned on the basis of its
three-bond coupling with H-2′ and H-6′. When attached to an
OGluA unit, C-3′ and C-4′ also showed three-bond coupling
with the anomeric H-1″. The A-ring ipso carbons C-5, C-7, and
C-9 were assigned on the basis of their two-bond correlations
with H-6 and H-8. The ipso carbon C-10 showed three-bond
correlations with H-6 and H-8 and in some cases two-bond
correlations with H-4a and H-4b. For the methylated
epicatechin glucuronides, the ipso carbon to which the OMe
group is attached showed strong three-bond coupling with the
methyl protons. Thus, by the combination of HSQC and
HMBC, all carbons could be assigned unambiguously.
LC/MS Analysis. The four epicatechin glucuronides 1G−4G,
in the negative ionization mode, exhibited a molecular ion of
465 as the base peak resulting from the loss of a proton [M −
H]. The m/z 289 ion resulting from the loss of the glucuronic
acid unit was also detected. Peaks from dimerization of the
glucuronides (m/z 931) were detected in the negative
ionization mode as well. The six methylated analogues 5G−
10G showed a molecular ion of 479 [M − H]. Like the
nonmethylated epicatechin glucuronides, a fragment from the
loss of the glucuronic acid unit (m/z 303) was detected for all
methylated analogues. Dimers (m/z 959) were also observed.
The MS/MS spectra of 1G−3G, 5G, 6G, and 10G appear in
the Supporting Information.
Figure 1. HPLC traces of epicatechin glucuronides 1G−10G on an
Agilent Eclipse XDB-Phenyl column (3.5 μm, 4.6 × 150 mm). Flow
rate: 1 mL/min. UV detection: 230 nm. Temperature: ambient.
Mobile phase: solvent A, 25 mM Na3PO4 buffer pH 2.5; solvent B,
MeCN. (a) Isocratic conditions (85% A/15% B). (b) Gradient
conditions (0−8% B in 40 min).
traces of the glucuronides 1G−10G. A mixture of 1G−10G was
artificially created by mixing aqueous solutions of the individual
components. The peaks were assigned by comparing the
retention times to those of the individual components under
the same HPLC conditions. Under isocratic conditions (85% of
25 mM Na3PO4 buffer/15% MeCN), baseline separation was
achieved for nearly all the compounds, except 2G, 4G, 8G, and
10G, in less than 10 min run time (Figure 1a). Epicatechin
itself co-eluted with compound 8G under these conditions.
Baseline separation of all 10 glucuronides was achieved using
gradient conditions by slowly increasing MeCN content from 0
to 8% in 40 min (Figure 1b). Under these conditions,
epicatechin co-eluted with epicatechin-3′-O-β-D-glucuronide
(1G).
Stability Assessment. The stability of epicatechin glucur-
onides under storage conditions was assessed by appearance,
1H NMR, and HPLC. All the glucuronides were white solids
after lyophilization. No change in appearance was observed
HPLC Analysis. The 10 epicatechin glucuronides 1G−10G
were analyzed by reversed-phase HPLC on an Agilent Eclipse
1
after 30 days at ambient temperature. H NMR spectra were
162
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
a
Scheme 4. Synthesis of 4′-O-Methylepicatechin-7-sulfate Ammonium Salt (10S)
a
Reagents and conditions: (i) DMAP (1 equiv), Et3N (2 equiv), Cl3CCH2OSO2Cl (2 equiv), THF; (ii) Pd(OH)2/C, H2, EtOAc; (iii) HCO2NH4,
Zn dust, MeOH.
Table 5. NMR Spectroscopic Data for Epicatechin Sulfate Ammonium Salts 1S−4S (DMSO-d6)
3′-sulfate 1S
δH (J in Hz)
4′-sulfate 2S
δH (J in Hz)
5-sulfate 3S
δH (J in Hz)
7-sulfate 4S
δH (J in Hz)
a
position
δC
δC
δC
δC
2
3
4
77.7
64.7
28.3
4.79, s
77.8
64.7
28.2
4.81, s
78.2
64.8
28.7
4.73, s
78.2
64.7
28.5
4.75, s
4.03, br, s
4.04, br, s
4.00, br, s
4.03, br, s
2.69, dd (16.8, 4.6)
2.70, dd (16.4, 3.8)
2.80, dd (16.8, 4.2)
2.59, dd (16.8, 2.5)
2.72, dd (16.8, 3.8)
2.55, d (17.0)
b
b
2.50
2.49, d (16.4)
5
156.6
95.2
156.6
95.2
152.7
101.1
155.7
97.8
156.1
99.8
6
5.89, d (2.2)
5.73, d (2.2)
5.91, s
5.74, s
6.54, s
5.95, s
6.32, s
6.12, s
7
156.3
94.1
156.3
94.0
152.2
99.3
8
9
155.6
98.4
155.5
98.4
155.1
103.5
130.5
115.1
144.6
144.6
114.9
117.9
155.0
102.7
130.4
115.1
144.85
144.90
115.1
117.8
10
1′
2′
3′
4′
5′
6′
131.0
122.3
140.1
148.5
116.5
123.9
136.5
116.1
148.5
139.9
122.3
117.9
7.17, d (1.5)
6.93, s
6.89, s
6.89, s
6.78, d (8.4)
7.05, d (8.4)
7.07, d (8.3)
6.66 (s)
6.66 (s)
6.66, s
6.66, s
6.79, d (8.2)
a
b
The 3-OH proton was also visible at δ 4.70 as a broad singlet. Partially overlapping with the DMSO peak.
identical to those obtained prior to storage. In aqueous
solution, the compounds were stable when stored in a
refrigerator for 2 days, as no change in HPLC peak area was
detected for any of the compounds from their initial values.
However, when its aqueous solution was stored at ambient
temperature, compound 3G turned yellow overnight. The
solution stability of the ammonium salt of 4′-O-methylepica-
techin-7-O-β-D-glucuronide (10G) in D2O was also assessed by
1H NMR after 4 weeks of storage at ambient temperature.
There was no change in 1H NMR except disappearance of H-6
and H-8 due to H−D exchange.39 These observations suggest
that epicatechin glucuronides are stable in the solid state at
ambient conditions and aqueous solutions when refrigerated.
Epicatechin Sulfates. Chemical Synthesis. The epicate-
chin sulfates 1S−10S were synthesized from the epicatechin
intermediates 1c−10c. The synthesis of 4′-O-methylepicate-
chin-7-sulfate (10S) is shown in Scheme 4. Reaction of the
phenolic compound 10c with chlorosulfuric acid 2,2,2-
trichloroethyl (TCE) ester in the presence of Et3N and
DMAP in THF afforded the TCE-protected epicatechin-7-
sulfate intermediate 10f in 44% isolated yield. Debenzylation
via catalytic hydrogenolysis cleanly afforded the TCE-protected
4′-O-methylepicatechin-7-sulfate 10g. Removal of the TCE
moiety was accomplished using Zn-ammonium formate in
MeOH40 to give 4′-O-methylepicatechin-7-sulfate ammonium
salt (10S). The crude was purified by preparative reversed-
phase HPLC. The pure 10S was obtained as an off-white
powder in 42% yield over two steps after lyophilization.
Epicatechin sulfates 1S−9S were prepared and isolated in the
same fashion as described for 10S starting from the
corresponding epicatechin intermediates 1c−9c.
1
NMR Characterization. The H and 13C NMR spectra of
the epicatechin sulfates 1S−10S were unambiguously assigned
using COSY, HSQC, and HMBC 2D NMR methods, and the
data are collated in Tables 5−7.
1
1H NMR Assignments. The H NMR spectra of epicatechin
sulfates 1S−10S in DMSO-d6 exhibit three key features: (a) H-
2 appears as a singlet (ca. δ 4.8), indicating its syn relationship
(J < 1 Hz) with H-3; (b) H-3 (ca. δ 4) is coupled to H-4a and
in some cases to H-4b as well with small coupling constants (ca.
3−4 Hz), thus appearing as partially resolved multiplets or
broad singlets; (c) H-4a and H-4b are diastereotopic, exhibiting
an AX splitting pattern with coupling constants in the range
16−17 Hz. The assignments of H-6 and H-8 of epicatechin
163
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
Table 6. NMR Spectroscopic Data for 3′-O-Methylepicatechin Sulfate Ammonium Salts 5S−7S (DMSO-d6)
3′-OMe-4′-sulfate 5S
δH (J in Hz)
3′-OMe-5-sulfate 6S
δH (J in Hz)
3′-OMe-7-sulfate 7S
δH (J in Hz)
a
position
δC
δC
δC
2
3
4
78.0
64.7
28.0
4.84, s
78.2
64.7
28.7
4.79, s
78.1
64.7
28.6
4.83, s
4.07, br, s
4.03, br, s
4.07, br, s
2.71, dd (16.6, 4.3)
2.80, d (16.4)
2.62, d (17.2)
2.74, dd (17.4, 3.9)
2.58, d (16.7)
b
2.53, d (16.6)
5
156.6
95.2
152.8
101.1
155.6
97.8
155.8
99.9
6
5.91, s
5.75, s
6.55, s
5.95, s
6.32, s
6.15, s
7
156.3
94.1
152.3
99.4
8
9
155.6
98.4
155.0
103.5
130.5
111.7
146.9
145.7
114.7
119.6
55.6
155.0
102.5
130.6
111.5
146.9
145.7
114.7
119.5
55.6
10
1′
2′
3′
4′
5′
6′
OMe
134.9
111.9
149.9
141.9
120.8
118.6
55.7
7.06, s
7.03, s
7.06, s
7.40, d (8.4)
6.90, d (8.4)
3.73, s
6.72, d (7.8)
6.83, d (8.1)
3.74, s
6.73, d (7.9)
6.85, d (8.2)
3.76, s
a
b
The 3-OH proton was visible at δ 4.70 as a broad singlet. Overlapping with the DMSO peak.
Table 7. NMR Spectroscopic Data for 4′-O-
Methylepicatechin Sulfate Ammonium Salts 8S−10S
(DMSO-d6)
Table 8. Changes in A- and B-Ring Proton Chemical Shifts
of Epicatechin Sulfates 1S−4S Relative to Epicatechin in
a
DMSO-d6
4′-OMe-7-sulfate
10S
H-6
H-8
H-2′
H-5′
H-6′
4′-OMe-3′-sulfate 8S 4′-OMe-5-sulfate 9S
epicatechin-3′-sulfate (1S)
epicatechin-4′-sulfate (2S)
epicatechin-5-sulfate (3S)
epicatechin-7-sulfate (4S)
0.00
0.02
0.65
0.43
0.02
0.03
0.24
0.41
0.28
0.04
0.00
0.00
0.12
0.41
0.00
0.00
0.39
0.13
0.00
0.00
δH (J in
Hz)
δH (J in
a
δH (J in
Hz)
position
δC
δC
Hz)
δC
2
3
4
77.9
64.7
28.4
4.78, s
78.0
64.8
28.6
4.78, s
78.0
64.7
28.4
4.81, s
4.03, m
4.01, m
4.05, br, s
a
1H NMR data for epicatechin (DMSO-d6, 400 MHz): δ 6.89 (H-2′),
6.66 (s, H-5′, H-6′), 5.89 (s, H-6), 5.71 (s, H-8), 4.73 (s, H-2), 4.00 (s,
H-3), 2.68 (d, J = 16.4 Hz, H-4a), 2.47 (d, J = 16.4 Hz, H-4b).
2.72, dd
(16.6,
4.1)
2.80, dd
(16.8,
4.5)
2.74, dd
(16.8,
4.1)
2,53, d
(16.6)
2.58, dd
(16.8,
3.0)
2.56, d
(16.2)
b
H-6′ of epicatechin-3′-sulfate (1S) underwent downfield shifts
of 0.28 and 0.39 ppm, respectively, while the chemical shift of
H-5′ remained relatively unaffected. For epicatechin-4′-sulfate
(2S), H-5′ was shifted downfield by 0.41 ppm, while H-2′ and
H-6′ remained relatively unaffected. For both 1S and 2S, the
chemical shifts of the A-ring protons were unaffected. For
epicatechin-5-sulfate (3S), H-6 underwent a downfield shift by
as much as 0.65 ppm compared with a downfield shift of 0.24
ppm for the H-8. For epicatechin-7-sulfate (4S), both H-6 and
H-8 underwent a similar downfield shift of ca.0.4 ppm. In both
instances, the chemical shifts of the B-ring protons remained
unchanged.
5
6
156.7
95.2
152.7
156.1
99.8
5.91, s
5.73, s
101.1 6.55, d
(2.0)
6.32, s
6.12, s
7
8
156.3
94.1
155.7
152.3
99.3
97.8
5.94, d
(2.0)
9
155.6
98.4
155.0
103.4
132.3
154.9
102.7
132.4
10
1′
2′
3′
4′
5′
131.6
120.4 7.51, s
142.0
114.8 6.93, s
145.9
114.7 6.96, s
146.0
150.0
146.8
146.9
112.1 6.92, d
(8.4)
111.7 6.85, d
(8.2)
111.8 6.86, d
(8.4)
13C NMR Assignments. The assignment of proton-bearing
carbons was achieved using HSQC. The assignments of the A-
ring and B-ring ipso carbons C-5, C-7, C-9, C-10, C-1′, C-3′,
and C-4′ were accomplished using HMBC. Specifically, the B-
ring ipso carbons C-1′ and C-3′ were assigned on the basis of
their three-bond coupling with H-5′, and C-4′ was assigned on
the basis of correlations with H-2′ and H-6′. The A-ring ipso
carbons C-5, C-7, and C-9 were assigned on the basis of their
two-bond correlations with H-6 and H-8. The nonoxygenated
ipso carbon C-10 showed three-bond correlations with H-6 and
H-8 and in some cases two-bond correlations with H-4a and H-
4b. For the methylated epicatechin sulfates, the ipso carbon to
which the OMe group was attached showed strong three-bond
coupling with the methyl protons. Thus, by the combination of
6′
122.1 7.10, d
(8.2)
117.7 6.78, d
(8.3)
117.5 6.79, d
(8.4)
OMe
55.8
3.73, s
55.7
3.74,s
55.7
3.75, s
a
The 3-OH proton was visible at δ 4.67 as a doublet (J = 3.6 Hz).
Overlapping with the DMSO peak.
b
sulfates 1S−10S were made on the basis of analogy to those of
epicatechin.37,38 Complete assignments of all H NMR signals
1
were accomplished using 2D-COSY.
The sulfate group causes the ortho and para protons to shift
downfield but has little effect on the chemical shifts of the meta
protons relative to epicatechin (Table 8). Specifically, H-2′ and
164
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
a
Table 9. Changes in 13C NMR Chemical Shifts of Epicatechin Sulfates 1S−10S Relative to Epicatechin
a
13C NMR of epicatechin (DMSO-d6, 100 MHz): δ 78.0 (C-2), 64.9 (C-3), 28.2 (C-4), 156.5 (C- 5), 95.1 (C-6), 156.2 (C-7), 94.1 (C-8), 155.7 (C-
9), 98.5 (C-10), 130.6 (C-1′), 114.9 (C-2′), 144.4 (C-3′), 144.5 (C-4′), 114.7 (C-5′), 117.9 (C-6′).
Figure 2. HPLC traces of epicatechin sulfates 1S−10S and the parent epicatechin (EC) on an Agilent Eclipse XDB-Phenyl 3.5 μm, 4.6 × 150 mm
column. Mobile phases: solvent A, 25 mM Na3PO4 buffer pH 2.5; solvent B, MeCN. Flow rate: 1 mL/min. UV detection: 230 nm. Temperature:
ambient. (a) Isocratic conditions (25 mM Na3PO4 buffer pH 2.5/MeCN = 80:20). (b) Gradient conditions (0−10% B in 40 min).
HSQC and HMBC, all carbons could be assigned unambigu-
ously.
As can be seen from the data, sulfation had significant effects
on the ortho and para carbons in addition to the ipso carbon of
the sulfated hydroxyl group, but little or no effect on the meta
carbons. As might be expected, sulfation had virtually no effect
on the 13C chemical shifts of carbons on the other
(nonsulfated) aromatic ring or on the aliphatic C-ring.
Specifically, sulfation of an A- or B-ring OH group caused
the ipso carbon to move upfield (yellow shading), whereas the
Effects of the sulfate group on the 13C NMR spectrum of
epicatechin were also examined. Listed in Table 9 are changes
in 13C NMR chemical shifts of epicatechin sulfates relative to
epicatechin. The comparison data in Table 9 reveal some
interesting trends which are only apparent because of the
unequivocal assignment of each signal.
165
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
ortho and para carbons moved downfield (blue shading).
Similar effects of replacing a phenolic OH with a sulfate group
on the 13C NMR shifts of the aromatic ring have been
previously reported.41 For the six methylated epicatechin
sulfates 5S−10S, the unsubstituted ortho carbon next to the
OMe group always showed an upfield shift of approximately 3
ppm (lavender shading). In addition, the ipso carbon bearing
the OMe was deshielded relative to epicatechin as expected
(green shading).
LC/MS Analysis. The epicatechin sulfate ammonium salts
1S−4S exhibited a base peak of m/z 369, in the negative
ionization mode, corresponding to a loss of NH4, and a smaller
peak at m/z 289 from the loss of SO3NH4. Under the same
conditions, the methylated epicatechin sulfate ammonium salts
5S−10S exhibited a base peak of m/z 383 [M − NH4] and a
second peak at m/z 303 [M − SO3NH4].
HPLC Analysis. The epicatechin sulfates 1S−10S were
analyzed by reversed-phase HPLC on an Agilent Eclipse XDB-
Phenyl analytical column. Owing to the highly polar nature, the
sulfates are not retained very well under typical reversed-phase
conditions (aqueous content <50%). Using a mobile phase that
consisted of 25 mM Na3PO4 buffer pH 2.5 and MeCN, the
epicatechin sulfates exhibited adequate retention on the Eclipse
XDB-Phenyl column. A mixture of sulfates 1S−10S was
artificially created by mixing aqueous solutions of the individual
sulfates. Shown in Figure 2 are HPLC traces of the 10-sulfate
mixture under isocratic conditions (80% of 25 mM Na3PO4
buffer pH 2.5/20% MeCN). In less than 5 min, all 10 sulfates
were separated, with five having baseline separation (Figure
2a). Under gradient conditions (from 0 to 10% MeCN in 40
min) the same column provided baseline separation of all 10
sulfates plus epicatechin (Figure 2b). The HPLC peaks were
assigned on the basis of retention times of the individual
sulfates under the same HPLC conditions.
hydroxy groups remain protected as benzyl ethers during the
sulfate/glucuronide formation step, there is no ambiguity with
regard to the site of glucuronidation/sulfation. The glucur-
1
onides and sulfates have been characterized by H NMR, 13C
NMR, and LC/MS. HPLC methods capable of baseline
separation of all the glucuronides and sulfates were developed.
The epicatechin glucuronides and sulfates are stable in the solid
state at ambient temperature and in aqueous solution when
stored refrigerated. These results should prove invaluable to the
research community as analytical standards as well as in future
studies of the biological and pharmacological effects of
epicatechin in humans.
EXPERIMENTAL SECTION
■
General Experimental Procedures. 1H NMR, 13C NMR, and all
2D NMR spectra were obtained on a Bruker Avance III 400 MHz
1
spectrometer. H NMR spectra were recorded at 400 MHz, and 13C
NMR spectra were recorded at 100 MHz. Samples were dissolved in
CDCl3, DMSO-d6, acetone-d6, and D2O. Chemical shifts are expressed
in ppm relative to TMS. Coupling constants are expressed in Hz. LC/
MS data were obtained on a PE SCIEX API 100 LC/MS system
equipped with a Perkin-Elmer 200 Series autosampler and LC pump.
MS/MS analyses were performed on a PE SCIEX API 4000 QTrap
system. The samples were introduced into the system by infusion.
Preparative HPLC purifications were performed on a Gilson system
consisting of a liquid handler, a syringe pump, an injection module, a
binary pump, and a UV detector on a Phenomenex preparative column
(Synergy 4 μ Polar-RP 80A, 250 × 21.20 mm, part no. 00G-4336-P0).
Analytical HPLC was carried out on an HP 1100 series system
equipped with a photodiode array detector on an Agilent Eclipse XDB-
Phenyl 3.5 μm, 4.6 × 150 mm column (part no. 963967-912). The
mobile phase for the preparative epicatechin glucuronide runs was a
binary gradient system consisting of H2O with 0.1% TFA by volume
(solvent A) and MeCN with 0.1% TFA by volume (solvent B). The
mobile phase for preparative epicatechin sulfate runs consisted of H2O
with 10 mM NH4OAc (solvent A) and MeCN with 10 mM NH4OAc
(solvent B). The amount of solvent B was increased from 10% to 40%
in 14 min or from 0 to 20% in 20 min. The flow rate was 20 mL/min.
Detection by UV was set at 230 nm. The crude products were
dissolved in H2O. Injection volume was 2 mL. The mobile phases for
the analytical runs were binary systems (isocratic or gradient)
consisting of 25 mM Na3PO4 buffer pH 2.5 (solvent A) and MeCN
(solvent B). Flow rate was 1 mL/min and UV detection was at 230
nm. The experiments were performed at ambient temperature.
Reagents and solvents were obtained from commercial sources and
used without further purification. Specifically, methyl 2,3,4-tri-O-
acetyl-1-O-(trichloroacetimidoyl)-α-D-glucuronate was purchased from
LC Scientific Inc. (Ontario, Canada). HPLC grade MeCN and H2O
were purchased from Burdick & Jackson (Honeywell). Orthogonally
protected epicatechin intermediates 1a−10a were prepared in house
using a synthesis strategy outlined in a preliminary communication by
Mull et al.30 A full report will be published elsewhere.31 All reactions
were run under a nitrogen atmosphere at room temperature unless
specified otherwise.
Stability Assessments. The chemical stability of the
epicatechin sulfates in both solid state and aqueous solution
1
was assessed by appearance, LC/MS, HPLC, and H NMR.
Solid samples of epicatechin sulfates 3S, 5S, 6S, and 10S stored
in tightly closed vials in a refrigerator for more than 13 months
did not exhibit any significant changes in appearance. LC/MS,
1H NMR, and HPLC showed no major degradation products
after the storage. In dilute aqueous solutions (<1 mg/mL), the
epicatechin sulfates 1S−10S are stable when kept refrigerated
for at least three days, as determined by HPLC. For selected
sulfates, aqueous HPLC samples kept in a refrigerator for more
than a year were reinspected by visual appearance and
reanalyzed by HPLC. 3′-O-Methylepicatechin-4′-sulfate (5S),
4′-O-methylepicatechin-3′-sulfate (8S), and 4′-O-methylepica-
techin-5-sulfate (9S) showed no change in appearance and no
decomposition products when reanalyzed by HPLC. Epica-
techin-4′-sulfate (2S), epicatechin-5-sulfate (3S), 3′-O-methyl-
epicatechin-5-sulfate (6S), and 3′-O-methylepicatechin-7-sul-
fate (7S) showed no change in appearance, suggesting no major
oxidation of the phenols during storage. However, when
reanalyzed by HPLC, the one-year-old aqueous samples of 3S,
6S, and 7S all showed a new peak (11−35%) in addition to the
main compound.42
Representative Procedure for Benzylation of the 3-OH
Group of Protected Epicatechin Intermediates 1a−10a. To a
suspension of NaH 60% dispersion (280 mg, 7 mmol) in anhydrous
DMF (10 mL) cooled in an ice-H2O bath was added a solution of 3a
(3.0 g, 5 mmol) in anhydrous DMF (20 mL) with a syringe. After the
addition, the cooling bath was removed and the reaction was stirred at
ambient temperature for 30 min. The reaction was recooled in the ice-
H2O bath, and BnBr (1.2 g, 7 mmol) was added. The reaction was
stirred at ambient temperature for 3 h, poured onto crushed ice (100
g), and extracted with EtOAc (150 mL × 3). The combined organic
layers were washed with H2O (100 mL) and brine (100 mL) and dried
over anhydrous Na2SO4. Removal of solvent in vacuo afforded the
crude product as an oil, which was purified by flash chromatography
(silica gel, heptane then heptane/EtOAc, 3:1, v/v) to afford the
In summary, we have synthesized all monoglucuronides and
sulfates of epicatechin, 3′-O-methylepicatechin, and 4′-O-
methylepicatechin, respectively, from the corresponding
orthogonally protected epicatechin intermediates prepared de
novo. Since only the phenolic hydroxy group to be
glucuronidated/sulfated is unmasked while all the remaining
166
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
product 3b as a white solid (2.5 g, 73% yield): 1H NMR (CDCl3, 400
MHz) δ 7.52−7.27 (15H, m), 7.22−7.17 (4H, m), 7.06−7.01 (2H,
m), 6.95 (2H, s, H-5′, H-6′), 6.42 (1H, s, H-6), 6.32 (1H, s, H-8),
5.26−5.00 (8H, m, 3 × OCH2Ph and OCH2OCH3), 4.97 (1H, s, H-
2), 4.42 (2H, ABq, JAB = 12.4 Hz, ΔνAB = 48.6 Hz, 3-OCH2Ph), 3.95
(1H, m, H-3), 3.49 (3H, s, OCH3), 2.99 (1H, dd, J = 17.2, 2.7 Hz, H-
4a), 2.79 (1H, dd, J = 17.4, 4.5 Hz, H-4b); 13C NMR (CDCl3,100
MHz) δ 158.8, 156.7, 155.8, 149.0, 148.6, 138.3, 137.6, 137.5, 137.2,
132.4, 128.7, 128.6, 128.5, 128.3, 128.0, 127.9, 127.8, 127.7, 127.6,
127.5, 120.0, 115.0, 114.1, 102.1, 96.0, 95.6, 94.7, 78.2, 72.7, 71.6, 71.4,
71.3, 70.2, 56.2, 24.7; LC/MS m/z 695 [M + H]+, 717 [M + Na]+.
Representative Procedure for Selective Removal of the
MOM Protective Group. To a solution of 3b (2.4g, 3.45 mmol) in a
1:1 mixture of MeOH and DCM (41 mL) was added 4 N HCl in
dioxane (20.5 mL). The reaction was stirred at ambient temperature
for 2 h and concentrated in vacuo without heating. The orange residue
thus obtained was dissolved in EtOAc (200 mL) and washed with
H2O (50 mL × 3) and brine (50 mL) until the aqueous phase was
close to neutral (pH 5−6). The organic layer was dried over
anhydrous Na2SO4. Removal of solvent in vacuo afforded the crude
product as a light brown oil, which was purified by flash
chromatography (silica gel, heptane/EtOAc, 3:1, v/v) to afford the
400 MHz) δ 7.53−6.92 (23H, m), 6.38 (1H, s, H-6), 6.28 (1H, s, H-
8), 5.38−5.32 (3H, m), 5.21 (2H, s, OCH2Ph), 5.18 (1H, d, J = 7.5
Hz, H-1″), 5.08 (2H, s, OCH2Ph), 5.02 (2H, s, OCH2Ph), 4.92 (1H, s,
H-2), 4.33 (2H, ABq, JAB = 12.3 Hz, ΔνAB = 65.6 Hz, 3-OCH2Ph),
4.20−4.14 (1H, m), 3.91 (1H, s, br, H-3), 3.75 (3H, s, OCH3), 2.89
(1H, d, J = 17.2 Hz, H-4a), 2.79 (1H, dd, J = 17.1, 3.7 Hz, H-4b), 2.08
(9H, s, 3 × OC(O)CH3); 13C NMR (CDCl3,100 MHz) δ 170.1,
169.5, 169.3, 166.9, 158.6, 156.0, 149.0, 148.6, 138.0, 137.5, 136.9,
132.2, 128.8, 128.6, 128.5, 128.3, 128.2, 127.9, 127.7, 127.6, 127.5,
120.0, 115.0, 114.0, 103.0, 99.0, 97.4, 96.3, 78.4, 72.7, 72.3, 72.0, 71.6,
71.3, 71.2, 70.4, 69.3, 68.8, 68.7, 53.2, 53.1, 24.7, 20.8, 20.7, 20.6; LC/
MS m/z 967 [M + H]+, 989 [M + Na]+.
Representative Procedure for Methyl Ester Hydrolysis and
Removal of Acetyl Groups of the Glucuronic Acid Unit. The
glucuronidation product 3d (390 mg, 0.4 mmol) was dissolved in a
mixture of THF (18 mL) and MeOH (3 mL). The mixture was cooled
in an ice-H2O bath. A solution of 1 N NaOH (6 mL) was added slowly
with a syringe. The reaction was stirred for 2 h in an ice-H2O bath and
partially concentrated in vacuo (no heat) to remove the organic
solvents. The aqueous residue (a white suspension) was acidified to
pH 1 with 1 N HCl. The white solid formed was collected by filtration,
rinsed with H2O, and dried by suction to afford the product 3e (200
1
1
product 3c as a pale pink solid (2.1 g, 94% yield): H NMR (CDCl3,
mg, 60% yield): H NMR (acetone-d6, 400 MHz) δ 7.61−7.04 (23H,
400 MHz) δ 7.55−7.30 (15H, m), 7.20 (4H, m), 7.07−7.01 (2H, m),
6.94 (2H, s, H-5′, H-6′), 6.27 (1H, d, J = 1.7 Hz, H-6), 6.10 (1H, d, J =
1.9 Hz, H-8), 5.26−4.95 (6H, m, 3 × OCH2Ph), 4.73 (1H, s, H-2),
4.42 (2H, ABq, JAB = 12.4 Hz, ΔνAB = 47.0 Hz, 3-OCH2Ph), 3.97 (1H,
m, H-3), 2.92 (1H, dd, J = 16.9, 3.0 Hz, H-4a), 2.78 (1H, dd, J = 16.7,
4.2 Hz, H-4b); 13C NMR (CDCl3,100 MHz) δ 158.7, 156.1, 155.1,
148.9, 148.5, 138.1, 137.5, 137.4, 137.2, 132.3, 128.7, 128.6, 128.5,
128.3, 128.0, 127.9, 127.8, 127.6, 127.5, 119.9, 115.0, 114.0, 100.0,
96.1, 95.5, 78.1, 72.5, 71.6, 71.4, 71.2, 70.1, 24.3; LC/MS m/z 651 [M
+ H]+, 673 [M + Na]+.
m), 6.51 (1H, s, H-6), 6.28 (1H, s, H-8), 5.24−5.02 (8H, m, 3 ×
OCH2Ph plus H-1 and H-1″), 4.43 (2H, ABq, JAB = 11.9 Hz, ΔνAB
=
101.9 Hz, 3-OCH2Ph), 3.82−3.70 (1H, m, H-4″), 4.09 (1H, m, H-3),
4.15 (1H, d, J = 9.7 Hz, H-4″), 3.66−3.55 (2H, m, H-2″, H-3″), 3.28
(1H, dd, J = 17.1, 2.6 Hz, H-4a), 2.84 (1H, dd, J = 16.9, 3.9 Hz, H-4b);
13C NMR (acetone-d6, 100 MHz) δ 170.3, 159.4, 158.0, 156.8, 149.6,
149.5, 139.6, 138.8, 138.7, 138.5, 133.7, 129.3, 129.2, 128.9, 128.6,
128.5, 128.4, 128.0, 121.0, 115.3, 115.2, 103.9, 102.7, 97.4, 97.0, 78.8,
77.3, 76.0, 74.4, 73.8, 72.6, 71.7, 71.6, 70.6, 24.9; LC/MS m/z 827 [M
+ H]+, 849 [M + Na]+.
Procedure for Selective Removal of PMB Protective Group.
Intermediate 2b (0.5 g, 0.65 mmol) and pyridinium p-toluenesulfonate
(0.082 g, 0.325 mmol) were heated in a mixture of 2-propanol (5 mL)
and toluene (3 mL) at reflux for 24 h. The reaction was cooled and
concentrated in vacuo. The residue was dissolved in EtOAc (10 mL)
and washed with H2O (5 mL). The aqueous layer was further
extracted with EtOAc (5 mL). The combined EtOAc layers were
washed with H2O (5 mL × 2) and dried over anhydrous Na2SO4.
Removal of solvent afforded the crude product as an oil, which was
purified by flash chromatography (silica gel, heptane/EtOAc, 3:1, v/v)
to obtain the epicatechin product 2c as a waxy solid containing ca. 17%
of the ent-catechin byproduct 2c′ by 1H NMR (0.28 g, 66% yield). The
Representative Procedure for the Removal of the Benzyl
Groups by Hydrogenolysis to Afford the Final Glucuronides. A
solution of intermediate 3e (109 mg, 0.132 mmol) in MeOH (10 mL)
containing 20% Pd(OH)2/C (wet, 44 mg) was stirred under an
atmosphere of H2 for 1 h. The flask was purged with N2, and the
reaction mixture was filtered through a pad of Celite to remove
catalyst. The Celite pad was rinsed with MeOH (20 mL). The
combined filtrate was concentrated to dryness in vacuo (no heat) and
purified by preparative HPLC in five injections. The purity of the
product fractions was checked by analytical HPLC. Fractions with
purity >95% were combined and partially evaporated in vacuo (no
heat) to a volume of ca. 5 mL. The aqueous residue was lyophilized
overnight to afford epicatechin-5-O-β-D-glucuronide (3G) as a white
solid (36 mg, 59% yield).
1
mixture was inseparable by TLC. NMR and LC/MS data for 2c: H
NMR (CDCl3, 400 MHz) δ 7.39−6.79 (23H, m), 6.21 (1H, s, H-6),
6.19 (1H, s, H-8), 5.58 (1H, s, phenolic OH), 4.97−4.83 (7H, m, 3 ×
OCH2Ph and H-2), 4.35 (2H, ABq, JAB = 12.4 Hz, ΔνAB = 69.8 Hz, 3-
OCH2Ph), 3.88 (1H, m, H-3), 2.96 (1H, dd, J = 17.3, 2.5 Hz, H-4a),
2.72 (1H, dd, J = 17.2, 4.4 Hz, H-4b); 13C NMR (CDCl3,100 MHz) δ
158.8, 158.2, 155.8, 145.7, 145.6, 138.3, 137.3, 137.2, 136.5, 131.0,
128.8, 128.7, 128.5, 128.3, 128.1, 128.0, 127.8, 127.7, 127.6, 127.4,
120.3, 114.1, 111.5, 101.7, 95.0, 94.0, 78.4, 73.1, 71.4, 71.2, 70.3, 70.1,
24.7; LC/MS m/z 651 [M + H]+, 673 [M + Na]+.
Representative Procedure for the Preparation of the 2,2,2-
Trichloroethyl-Protected Epicatechin Sulfates. The 7-OH
epicatechin intermediate 10c (780 mg, 1.36 mmol) and DMAP
(166 mg, 1.36 mmol) were dissolved in anhydrous THF (15 mL)
under N2. To the reaction were added Et3N (275 mg, 2.72 mmol) and
chlorosulfuric acid 2,2,2-trichloroethyl ester (506 mg, 2.04 mmol) with
a syringe. The mixture was stirred overnight at room temperature. The
resulting white suspension was diluted with EtOAc (50 mL) and
washed with H2O (50 mL), 0.5 N HCl (50 mL), and 5% K2CO3 (50
mL), successively. The organic layer was further washed with brine (50
mL each) until the aqueous wash became neutral. The organic layer
was separated and dried over anhydrous MgSO4. Removal of solvent in
vacuo afforded the crude as an oil, which was dissolved in a minimal
amount of DCM and purified by flash column (silica gel, heptane/
EtOAc, 3:1, v/v) to afford the epicatechin sulfate 2,2,2-trichloroethyl
Representative Procedure for Glucuronidation. The inter-
mediate 3c (325 mg, 0.5 mmol) and methyl 2,3,4-tri-O-acetyl-1-O-
(trichloroacetimidoyl)-α-D-glucuronate (275 mg, 0.575 mmol) were
dissolved in anhydrous DCM (10 mL) in the presence of 4 Å
molecular sieves. The reaction was stirred at ambient conditions for 30
min and then cooled in an ice-H2O bath for 15 min. A solution of BF3-
etherate (35.5 mg, 0.25 mmol) in anhydrous DCM (2 mL) was added
slowly with a syringe. The reaction was stirred for 1.5 h in the ice-H2O
bath and quenched by saturated NaHCO3 solution (15 mL). The
organic layer was separated, washed with brine (5 mL), and dried over
anhydrous Na2SO4. The crude product, obtained after removal of
solvent, was purified by flash chromatography (silica gel, heptane/
DCM/EtOAc, 5:4:1, v/v/v) to afford the product 3d as a white solid
(390 mg, 81% yield). 1H NMR data indicated the presence of ca. 10%
1
ester 10f (0.47g, 44% yield): H NMR (CDCl3, 400 MHz) δ 7.60−
7.01 (18H, m), 6.79 (1H, s, H-6), 6.70 (1H, s, H-8), 5.27−5.16 (4H,
m, 2 × OCH2Ph), 5.13 (1H, s, H-2), 4.94 (2H, s, OSO3CH2CCl3),
4.51 (2H, ABq, JAB = 12.2 Hz, ΔνAB = 54.4 Hz, 3-OCH2Ph), 4.11 (1H,
s, br, H-3), 4.07 (3H, s, 3′-OCH3), 3.19 (1H, d, J = 17.6 Hz, H-4a),
2.96 (1H, dd, J = 17.8, 4.1 Hz, H-4b); 13C NMR (CDCl3,100 MHz) δ
158.1, 156.0, 149.6, 149.3, 148.2, 138.0, 137.3, 136.4, 130.8, 128.8,
128.6, 128.4, 128.3, 127.9, 127.8, 127.7, 127.5, 119.7, 113.1, 111.5,
1
ent-catechin 3d′. NMR and LC/MS data for 3d: H NMR (CDCl3,
167
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
108.8, 102.9, 97.7, 92.6, 80.6, 78.5, 72.3, 71.6, 71.1, 70.6, 56.3, 25.1;
LC/MS m/z 785 [M + H]+, 807 [M + Na]+.
(8) Curtis, P. J.; Sampson, M.; Potter, J.; Dhatariya, K.; Kroon, P. A.;
Cassidy, A. Diabetes Care 2012, 35, 226−232.
Representative Procedure for Debenzylation of the 2,2,2-
Trichloroethyl-Protected Epicatechin Sulfates. The epicatechin
sulfate 2,2,2-trichloroethyl ester 10f (470 mg, 0.60 mmol) was
dissolved in EtOAc (25 mL) in the presence of Pd(OH)2/C (211 mg)
and placed under an atmosphere of H2 for 30 min. The catalyst was
removed by filtration over a pad of Celite. The filtrate was
concentrated in vacuo (with no heat) to obtain the crude 2,2,2-
trichloroethyl-4′-O-methylepicatechin-7-sulfate (10g) as an off-white
foam (0.29 g): 1H NMR (DMSO-d6, 400 MHz) δ 6.96 (1H, s, H-2′),
6.89 (1H, d, J = 8.3 Hz, H-5′), 6.81 (1H, d, J = 8.2 Hz, H-6′), 6.49
(1H, s, H-6), 6.48 (1H, s, H-8), 5.32 (2H, s, OSO3CH2CCl3), 4.92
(1H, s, H-2), 4.09 (1H, br, s, H-3), 3.76 (3H, s, OCH3), 2.81 (1H, dd,
J = 17.0, 3.4 Hz, H-4a), 2.65 (1H, d, J = 16.8 Hz, H-4b); 13C NMR
(DMSO-d6, 100 MHz) δ 156.9, 155.9, 148.1, 147.0, 145.9, 131.7,
117.7, 114.7, 111.6, 107.8, 100.1, 99.7, 92.9, 80.0, 78.4, 64.0, 55.7, 28.6;
LC/MS m/z 515 [M + H]+, 537 [M + Na]+. The product was used
directly for next step.
Representative Procedure for the Removal of the 2,2,2-
Trichloroethyl Group to Generate the Epicatechin Sulfates.
The crude sulfate ester 10g (270 mg, 0.52 mmol) was treated with
ammonium formate (198 mg, 3.14 mmol) and Zn dust (67 mg, 1.02
mmol) in MeOH (10 mL) at room temperature. After completion of
the reaction (TLC-silica gel, heptane/EtOAc, 1:2, v/v), the mixture
was filtered through a pad of Celite. The filtrate was concentrated in
vacuo (no heat). The oily residue, thus obtained, was mixed with H2O
(6 mL). The precipitated solid was removed by filtration, and the
filtrate was purified by preparative HPLC in three injections using the
conditions described earlier (vide supra). Fractions containing pure
product were pooled and concentrated in vacuo to a volume of about 5
mL. The aqueous residue was lyophilized to obtain the 4′-O-
methylepicatechin-7-sulfate ammonium salt (10S) as a white powder
(101 mg, 42% over two steps).
(9) Schroeter, H.; Heiss, C.; Balzer, J.; Kleinbongard, P.; Keen, C. L.;
Hollenberg, N. K.; Sies, H.; Kwik-Uribe, C.; Schmitz, H. H.; Kelm, M.
Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 1024−1029.
(10) Mink, P. J.; Scrafford, C. G.; Barraj, L. M.; Harnack, L.; Hong,
C. P.; Nettleton, J. A.; Jacobs, D. R., Jr. Am. J. Clin. Nutr. 2007, 85,
895−909.
(11) Ottovani, J. I.; Kwik-Uribe, C.; Keen, C. L.; Schroeter, H. Am. J.
Clin. Nutr. 2012, 95, 851−858.
(12) Ottaviani, J. I.; Momma, T. Y.; Heiss, C.; Kwik-Uribe, C.;
Schroeter, H.; Keen, C. L. Free Radical Biol. Med. 2011, 50, 237−244.
(13) Natsume, M.; Osakabe, N.; Yasuda, A.; Osawa, T.; Terao, J. J.
Clin. Biochem. Nutr. 2008, 42, 50−53.
(14) Ottovani, J. I.; Momma, T. Y; Kuhnle, G. K.; Keen, C. L.;
Schroeter, H. Free Radical Biol. Med. 2012, 52, 1403−1412.
(15) Actis-Goretta, L.; Leveques, A.; Giuffrida, F.; Michailidis, F. R.;
Viton, F.; Barron, D.; Duenas-Paton, M.; Gonzalez-Manzano, S.;
Santos-Buelga, C.; Williamson, G.; Dionisi, F. Free Radical Biol. Med.
2012, 53, 787−95.
(16) Li, C.; Meng, X.; Winnik, B.; Lee, M. J.; Lu, H.; Sheng, S.;
Buckley, B.; Yang, C. S. Chem. Res. Toxicol. 2001, 14, 702−707.
(17) Meng, X.; Sang, S.; Zhu, N.; Lu, H.; Sheng, S.; Lee, M. J.; Ho, C.
T.; Yang, C. S. Chem. Res. Toxicol. 2002, 15, 1042−1050.
(18) Piskula, M. K.; Terao, J. J. Nutr. 1998, 128, 1172−1178.
(19) Spencer, J. P.; Chowrimootoo, G.; Choudhury, R.; Debnam, E.
S.; Srai, S. K.; Rice-Evans, C. FEBS Lett. 1999, 458, 224−230.
(20) Kuhnle, G.; Spencer, J. P.; Schroeter, H.; Shenoy, B.; Debnam,
E. S.; Srai, S. K.; Rice-Evans, C.; Hahn, U. Biochem. Biophys. Res.
Commun. 2000, 277, 507−512.
(21) Donovan, J. L.; Crespy, V.; Manach, C.; Morand, C.; Besson, C.;
Scalbert, A.; Remesy, C. J. Nutr. 2001, 131, 1753−1757.
(22) Spencer, J. P.; Schroeter, H.; Kuhnle, G.; Srai, S. K.; Tyrrell, R.
M.; Hahn, U.; Rice-Evans, C. Biochem. J. 2001, 354, 493−500.
(23) Harada, M.; Kan, Y.; Naoki, H.; Fukui, Y.; Kageyama, N.; Nakai,
M.; Miki, W.; Kiso, Y. Biosci. Biotechnol. Biochem. 1999, 63, 973−977.
(24) Natsume, M.; Osakabe, N.; Oyama, M.; Sasaki, M.; Baba, S.;
Nakamura, Y.; Osawa, T.; Terao, J. Free Radical Biol. Med. 2003, 34,
840−849.
(25) Blount, J. W.; Ferruzzi, M.; Raftery, D.; Pasinetti, G. M.; Dixon,
R. A. Biochem. Biophys. Res. Commun. 2012, 417, 457−461.
(26) Gonzalez-Manzano, S.; Gonzalez-Paramas, A.; Santos-Buelga,
C.; Duenas, M. J. Agric. Food Chem. 2009, 57, 1231−1238.
(27) Duenas, M.; Gonzalez-Manzano, S.; Surco-Laos, F.; Gonzalez-
Paramas, A.; Santos-Buelga, C. J. Agric. Food Chem. 2012, 60, 3592−
3598.
(28) The O-methylation of epicatechin and catechin primarily occurs
at 3′- and 4′-OH via the action of catechol-O-methyltransferase, and
only one OH group is methylated, limiting the total number of
epicatechin glucuronides/sulfates that could be formed to 10 each (4
nonmethylated and 6 methylated forms). For references on
enzymology of methylation of epicatechin and catechin, see:
(a) Kuhnle, G.; Spencer, J. P.; Schroeter, H.; Shenoy, B.; Debnam,
E. S.; Srai, S. K.; Rice-Evans, C.; Hahn, U. Biochem. Biophys. Res.
Commun. 2000, 277, 507. (b) Lu, H.; Meng, X.; Yang, C. S. Drug.
Metab. Dispos. 2003, 31, 572−579.
(29) During the preparation of this article, a paper describing the
synthesis of four B-ring conjugates using a different synthetic approach
has appeared: Romanov-Michailidis, F.; Viton, F.; Fumeaus, R.;
Leveques, A.; Actis-Coretta, L.; Rein, M.; Williamson, G.; Barron, D.
Org. Lett. 2012, 14, 3902−3905.
(30) Mull, E. S.; Van Zandt, M.; Golebiowski, A.; Beckett, P.;
Sharma, P. K.; Schroeter, H. Tetrahedron Lett. 2012, 53, 1501−1503.
(31) Zhang, M.; Jagdmann, G. E., Jr.; Van Zandt, M.; Beckett, P.;
Schroeter, H. Tetrahedron: Asymmetry, submitted.
ASSOCIATED CONTENT
■
S
* Supporting Information
1
Copies of H NMR, 13C NMR, COSY, HSQC, and HMBC
spectra of 1G−10G and 1S−10S and MS/MS spectra of 1G−
3G, 5G, 6G, and 10G. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Tel: 1 203 570 8879. Fax: 1 203 653 2803. E-mail: nengren@
yahoo.com.
■
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) In this paper, the common names epicatechin and catechin
designate the (2R, 3R)- and (2R, 3S)-epimers, respectively.
(2) Porter, L. J. In The Flavonoids, Advances in Research since 1986;
Harborne, J. B., Ed.; Chapman & Hall: Landon, 1994; pp 23−56.
(3) Ellinger, S.; Reusch, A.; Stehle, P.; Helfrich, H. P. Am. J. Clin.
Nutr. 2012, 95, 1365−1377.
(4) Hooper, L.; Kay, C.; Abdelhamid, A.; Kroon, P. A.; Cohn, J. S.;
Rimm, E. B.; Cassidy, A. Am. J. Clin. Nutr. 2012, 95, 740−751.
(5) Schroeter, H.; Heiss, C.; Spencer, J. P.; Keen, C. L.; Lupton, J. R.;
Schmitz, H. H. Mol. Aspects Med. 2010, 31, 546−557.
(6) Van Praag, H.; Lucero, M. J.; Yeo, G. W.; Stecker, K.; Heivand,
N.; Zhao, C.; Yip, E.; Afanador, M.; Schroeter, H.; Hammerstone, J.;
Gage, F. H. J. Neurosci. 2007, 27, 5869−5878.
(7) Heiss, C.; Jahn, S.; Taylor, M.; Real, W. M.; Angeli, F. S.; Wong,
M. L.; Amabile, N.; Prasad, M.; Rassaf, T.; Ottaviani, J. I.; Mihardja, S.;
Keen, C. L.; Springer, M. L.; Boyle, A.; Grossman, W.; Glantz, S. A.;
Schroeter, H.; Yeghiazarians, Y. J. Am. Coll. Cardiol. 2010, 56, 218−
224.
(32) Geissman, T. A. The Chemistry of Flavonoid Compounds;
MacMillan: New York, 1962; p 446.
(33) Pearson, A. G.; Kiefel, M. J.; Ferro, V.; Von Itzstein, M.
Carbohydr. Res. 2005, 340, 2077−2085.
168
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169

Journal of Natural Products
Article
(34) Needs, P. W.; Kroon, P. A. Tetrahedron 2006, 62, 6862−6868.
(35) In one instance, adding BF3·OEt2 neat to the reaction resulted
in significant C-2 epimerization of the epicatechin core to form the
corresponding ent-catechin epimer. After removal of the acetyl, methyl,
and benzyl protecting groups, the product was isolated as a mixture of
ent-catechin-5-O-β-D-glucuronide (3G′) and epicatechin-5-O-β-D-
glucuronide (3G) in a ratio of 2.5:1 as determined by HPLC. The
two products were separated by preparative reversed-phase HPLC.
(36) Characterization data of 3′-O-methyl-ent-catechin-4′-O-β-D-
1
glucuronide (5G′): H NMR (acetone-d6 with 10% D2O, 400 MHz)
δ 7.13 (1H, d, J = 8.2 Hz, H-5′), 7.08 (1H, s, H-2′), 6.96 (1H, d, J =
8.3 Hz, H-6′), 6.01 (1H, s, H-6), 5.86 (1H, s, H-8), 5.08 (1H, d, J =
7.0 Hz, H-1″), 4.58 (1H, d, J = 8.3 Hz, H-2), 4.06−3.93 (2H, m, H-5″,
H-3), 3.84 (3H, s, OCH3), 3.70−3.62 (1H, m, H-3″), 3.62−3.53 (2H,
m, H-4″, H-2″), 2.93 (1H, dd, J = 16.1, 5.7 Hz, H-4a), 2.49 (1H, dd, J
= 15.9, 9.0 Hz, H-4b); 13C NMR (acetone-d6 with 10% D2O, 100
MHz) δ 157.5, 157.1, 156.4, 149.7, 146.8, 135, 121.1, 116.5, 112.6,
101.8, 100.4, 96.2, 95.1, 82.3, 76.5, 73.9, 72.3, 68.1, 56.4, 29.0; LC/MS
(negative mode) m/z 479 [M − H]−.
(37) Abd El-Razek, M. H. Asian J. Chem. 2007, 19, 4867−4872.
(38) Shen, C.-C.; Chang, Y.-S.; Ho, L.-K. Phytochemistry 1993, 34,
843−845.
(39) Kiehlmann, E.; Lehto, N.; Cherniwchan, D. Can. J. Chem. 1988,
66, 2431−2439.
(40) Liu, Y.; Lien, I.-F.; Ruttgaizer, S.; Dove, P.; Taylor, S. D. Org.
Lett. 2004, 6, 209−212.
(41) Ragan, M. A. Can. J. Chem. 1978, 56, 2681−2685.
(42) LC/MS of these samples revealed no new molecular ions,
suggesting that the new peaks were likely from the corresponding ent-
catechin sulfates formed as a result of the 4′-OH-induced C-2
epimerization during storage (see ref 32).
169
dx.doi.org/10.1021/np300568m | J. Nat. Prod. 2013, 76, 157−169