Theadibenzotropolone A, a new type pigment from enzymatic oxidation of (-)-epicatechin and (-)-epigallocatechin gallate and characterized from black tea using LC/MS/MS

Article abstract of DOI:10.1016/S0040-4039(02)01707-0

Theaflavins and thearubigins are major pigments of black tea. In the course of studies on the oxidation mechanism of tea polyphenols, a new type of tea pigment, theadibenzotropolone A, together with theaflavin 3-gallate were formed by the reaction of (-)-epicatechin (EC) and (-)-epigallocatechin gallate (EGCG) with horseradish peroxidase in the presence of H₂O₂. The structure of theadibenzotropolone A was elucidated on the basis of MS and 2D NMR spectroscopic analyses. The observation that galloyl ester groups of theaflavins can be oxidized to form dibenzotropolone skeletons strongly implied that this type of oxidation as an important pathway to extend the molecular size of thearubigins. The existence of this compound in black tea was characterized by LC/ESI-MS/MS. Theadibenzotropolone A is the first theaflavin type trimer of catechins in black tea.

Full text of DOI:10.1016/S0040-4039(02)01707-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7129–7133
Theadibenzotropolone A, a new type pigment from enzymatic
oxidation of (−)-epicatechin and (−)-epigallocatechin gallate and
characterized from black tea using LC/MS/MS
Shengmin Sang,a,* Shiying Tian,b Xiaofeng Meng,c Ruth E. Stark,b Robert T. Rosen,a
Chung S. Yangc and Chi-Tang Hoa
aDepartment of Food Science and Center for Advanced Food Technology, Rutgers University, 65 Dudley Road, New Brunswick,
NJ 08901-8520, USA
bDepartment of Chemistry, Graduate Center and College of Staten Island, City University of New York, 2800 Victory Boulevard,
Staten Island, NY 10314-6600, USA
cLaboratory for Cancer Research, College of Pharmacy, Rutgers University, 164 Frelinghuysen Road, Piscataway,
NJ 08854-8020, USA
Received 24 July 2002; revised 13 August 2002; accepted 15 August 2002
Abstract—Theaflavins and thearubigins are major pigments of black tea. In the course of studies on the oxidation mechanism of
tea polyphenols, a new type of tea pigment, theadibenzotropolone A, together with theaflavin 3-gallate were formed by the
reaction of (−)-epicatechin (EC) and (−)-epigallocatechin gallate (EGCG) with horseradish peroxidase in the presence of H2O2.
The structure of theadibenzotropolone A was elucidated on the basis of MS and 2D NMR spectroscopic analyses. The
observation that galloyl ester groups of theaflavins can be oxidized to form dibenzotropolone skeletons strongly implied that this
type of oxidation as an important pathway to extend the molecular size of thearubigins. The existence of this compound in black
tea was characterized by LC/ESI-MS/MS. Theadibenzotropolone A is the first theaflavin type trimer of catechins in black tea.
© 2002 Elsevier Science Ltd. All rights reserved.
Tea is one of the most popular beverages in the world.
Three types of tea are produced from the leaves of
Camellia sinensis—green tea, oolong tea, and black tea.
Green tea (non-fermented) and oolong tea (semi-fer-
mented) are more popular in China, Japan, Korea and
some African countries, whereas black tea (fermented)
is preferred in India and the Western countries.
Theaflavins and thearubigins are major pigments in
black tea, and it is generally accepted that these are
produced by oxidation of flavan-3-ols (catechins) dur-
ing tea fermentation.1 Theaflavins, which are orange or
orange–red in color, possess a benzotropolone skeleton
that is formed from co-oxidation of appropriate pairs
of catechins, one having a vic-trihydroxy structure, and
the other having an ortho-dihydroxy group.2,3 It is
known that theaflavins make important contributions
to the properties of black tea, such as color,4
‘mouthfeel’5 and extent of tea cream formation.6 Their
structures are well studied.7–10 On the other hand,
thearubigins, which are red–brown or dark brown, are
heterogeneous polymers.11 So far information about
their formation, structures and contribution to black
tea quality is very limited. A partial structure of thearu-
bigins from black tea was elucidated by using chemical
degradation, which indicated them to be heterogeneous
polymers of flavan-3-ols and flavan-3-ol gallates having
bonds at C-4, C-6, C-8, C-2%, C-5%, and C-6% in the
flavan-3-ol units.12 In addition, the possible participa-
tion of theaflavins in the formation of thearubigins has
been suggested.1 It was shown in a recent study that
horseradish peroxidase would oxidize theaflavins into
thearubigins in the presence of H2O2.13 While the con-
tribution of peroxidase to the metabolism of theaflavin
and thearubigin in tea fermentation remains to be
clarified. In the course of our studies on the functions
of peroxidase in tea fermentation, we examined the
oxidation of two tea catechins, EC (1) and EGCG (2)
by treatment with horseradish peroxidase in the pres-
ence of H2O2. In addition to theaflavin 3-gallate, a new
type pigment, theadibenzotropolone A (4), was isolated.
This paper reports the isolation, structural elucidation
and characterization from black tea by LC/ESI-MS/
MS. of this new pigment.
Keywords: theadibenzotropolone A; black tea; peroxidase; H2O2;
enzyme oxidation.
* Corresponding author.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01707-0

7130
S. Sang et al. / Tetrahedron Letters 43 (2002) 7129–7133
with the 1H NMR spectrum of 3, compound 4 is
distinguished by the absence of galloyl ester signals, a
large downfield shift of H-3%, an additional set of A and
C ring signals from flavan-3-ol, and three more olefinic
proton signals (l 7.54 brs H-c; 7.79 s H-g; 8.30 brs
H-e). In the 13C NMR spectrum of 4, besides the A and
C ring signals, there were observed 23 carbon signals
including two carbonyls (l 186.5 s C-a; 185.5 s C-a%),
one ester carbonyl (l 167.7 s C-l) and 20 olefinic
carbons (Table 1). All of these spectral features support
the presence of two benzotropolone groups in com-
pound 4. Thus, the galloyl ester group on 3 can react
with the B-ring of EC to form another benzotropolone.
This assertion is supported by the 2D HMBC NMR
spectrum. HMBC spectral analysis (Fig. 2) yielded cor-
relation peaks between H-c (l 7.54) and C-a (l 186.5),
C-b (l 155.1), C-d (l 123.9), C-e (l 132.4), C-l (l
167.7); H-e (l 8.30) and C-c (l 114.9), C-d (l 123.9),
C-f (l 134.8), C-l (l 167.7), C-j (l 122.7); H-g (l 7.79)
and C-2 (l 76.8), C-h (l 149.7), C-i (l 152.0), C-k (l
126.3). Thus, the correlation peaks of H-c and H-e with
the ester carbonyl carbon C-l, and H-g with C-2 of the
flavan-3-ol unit all indicate the presence of a benzotro-
polone group formed by the galloyl ester group on 3
with the B-ring of EC. Another three olefinic protons (l
7.28 brs H-c%; 7.56 brs H-e%; 7.71 s H-g%) show similar
correlation patterns to those of H-c, H-e, and H-g. H-c%
has cross peaks with C-a% (l 185.5), C-b% (l 155.6), C-d%
(l 133.4), C-e% (l 125.9), and C-2% (l 79.9); H-e% shows
correlation peaks with C-c% (l 117.7), C-d% (l 133.4),
C-f% (l 131.2), C-j% (l 121.8) and C-2% (l 79.9); H-g% has
correlations with C-f% (l 131.2), C-h% (l 146.9), C-i% (l
EC (1) (1.0 g) and EGCG (2) (1.0 g) were dissolved in
a mixture of acetone–pH 5 buffer (1:10, v/v, 50 mL),
which contained 5 mg of horseradish peroxidase. While
being stirred, 2 mL of 3.13% H2O2 was added four
times over 45 min. One major reaction product, com-
pound 3, was formed. After the addition of 0.5 mL of
H2O2, stirring was continued for an additional 30 min,
the amount of compound 3 decreased, while another
reaction product, compound 4 was formed. Compound
3 (120 mg) and 4 (18 mg), the two major reaction
products, and the recovered EC (180 mg) and EGCG
(70 mg) were isolated using a combination of Sephadex
LH-20 (40% acetone/water) column and RP C-18 (50%
MeOH/water) column chromatography.
Compound 3 was identified as theaflavin 3-gallate by
comparison with the standard sample isolated from
1
black tea. They showed the same APCI MS, H, 13C
NMR spectra data.
The molecular formula of 4 was determined to be
C50H38O21 by negative-ion ESI-MS ([M−H]− at m/z
973) as well as from its 13C NMR data, which indicate
that this compound consists of three flavan-3-ol units.
1
This was confirmed by the H and 13C NMR data. The
13C NMR spectrum of 4 displayed 50 carbon signals, 27
of which were assigned to the A and C rings of flavan-
3-ols. In addition, the 1H NMR spectrum exhibited
three sets of signals due to protons at the 2-, 3-, 4-, 6-,
and 8- positions of the flavan-3-ol nucleus. These obser-
vations also indicate that the A and C rings of 4 did not
undergo any change during oxidation. In comparison
Table 1. lH (600 MHz) and lC (150 MHz) NMR spectra data of compound 4 (CD3OD) (l in ppm, J in Hz)
lH
lC
lH
lC
2
3
4
5.11 brs
3.94 brs
2.62 d 17.4
2.55 dd 17.4, 1.8
76.8 d
66.3 d
29.6 t
8¦
9¦
10¦
a
5.52 brs
96.1 d
157.8 s
99.9 s
186.5 s
155.1 s
114.9 d
123.9 s
132.4 d
134.8 s
123.5 d
149.7 s
152.0 s
122.7 s
126.3 s
167.7 s
185.5 s
155.6 s
117.7 d
133.4 s
125.9 d
131.2 s
124.1 d
146.9 s
151.6 s
121.8 s
128.2 s
5
6
7
8
156.5 s
95.9 d
158.5 s
97.3 d
156.8 s
99.2 s
79.9 d
70.9 d
26.0 t
b
c
d
e
f
g
h
i
j
k
l
a%
b%
c%
d%
e%
f%
g%
h%
i%
6.05 brs
6.02 brs
7.53 brs
8.30 brs
7.79 s
9
10
2%
3%
4%
5.21 brs
5.86 brs
3.13 dd 17.4, 4.2
3.03 dd 17.4, 3.6
5%
6%
7%
8%
157.5 s
97.1 d
158.1 s
96.3 d
157.0 s
99.8 s
76.9 d
66.0 d
29.5 t
5.91 brs
5.82 d 1.2
7.28 brs
7.56 brs
7.71 s
9%
10%
2¦
3¦
4¦
5.38 brs
4.12 brs
2.84 17.4, 3.6
2.72 d 17.4
5¦
6¦
7¦
157.5 s
96.9 d
157.8 s
j%
k%
5.89 d 1.2

S. Sang et al. / Tetrahedron Letters 43 (2002) 7129–7133
7131
isolated from the 80% acetone aqueous extraction of
black tea by Sephadex LH-20 column chromatography
(40% acetone/water). Fig. 3 shows the respective LC/
MS/MS chromatograms and fragment ion mass spectra
of compound 4 and the theaflavin fraction of black tea.
As shown in this figure, compound 4 and the peak in
the theaflavin fraction of black tea exhibited almost not
only the same chromatographic retention time and
molecular masses, but also had the same fragment ion
mass spectra. This proved the presence of theadiben-
zotropolone A in black tea.
It is well known that the major oxidizing enzyme in tea
is polyphenol oxidase. It is the action of this enzyme
that mainly produces the theaflavins. So, the formation
in black tea of theadibenzotropolone A may come from
the reaction between theaflavin 3-gallate and the free
catechin EC by the action of peroxidase in the presence
of H2O2. This was confirmed by the reaction of
theaflavin 3-gallate and EC using the same conditions
as that of EC and EGCG. All of the theaflavin 3-gallate
can change to theadibenzotropolone A if there is
enough free EC. Thus, the synthesis of theadibenzotro-
polone A via enzymatic oxidation of EC (1) and EGCG
(2) and EC and theaflavin 3-gallate (3) is highly signifi-
cant. It is generally accepted that theaflavins possess a
benzotropolone skeleton that is formed from co-oxida-
tion of appropriate pairs of catechins, one having a
vic-trihydroxyl structure and the other having an ortho-
dihydroxy group. The formation of theaflavin type
compounds involves the oxidation of the B rings to the
quinones followed by a Michael addition of the gallo-
catechin quinone to the catechin quinone, prior to
carbonyl addition across the ring and subsequent decar-
boxylation.7,14 Since theaflavin 3-gallate has a vic-trihy-
droxy structure, and EC has an ortho-dihydroxy group,
they can form a benzotropolone skeleton. The forma-
tion of theadibenzotropolone A confirmed this hypothe-
sis. Theadibenzotropolone A provides the first evidence
to suggest that the galloyl ester group of theaflavin
3-gallate is as reactive as the B-ring (vic-trihydroxy) of
EGCG or EGC (epigallocatechin) and the galloyl ester
group of ECG (epicatechin gallate) in the same mecha-
nism as the formation of theaflavins and theaflavate A
and B15,16 (Fig. 4). The observation that the galloyl
ester group of theaflavins can be oxidized to form
dibenzotropolone skeletons strongly implied that this
type of oxidation is an important pathway to extend
the molecular size of thearubigins. Theadibenzotro-
polone A is the first theaflavin type trimer of catechins
in black tea. In addition, it may be a good method to
study the minor constituents of black tea by combining
the enzymatic oxidation reaction and LC/MS/MS anal-
ysis.
Figure 1. Structures of compounds 1–4.
Figure 2. Significant HMBC (H“C) correlations of com-
pound 4.
151.6), C-k% (l 128.2) and C-2¦ (l 76.9) (Fig. 2). These
through-bond connectivities proved that H-c%, H-e% and
H-g% belong to the benzotropolone of the theaflavin
part of 4. Thus, the structure of 4 was deduced as
shown (Fig. 1) and named theadibenzotropolone A.
The complete interpretation of the NMR data was
based on the results of COSY, HMQC and HMBC
experiments (Table 1).
Acknowledgements
Efforts were made to establish the presence of theadi-
benzotropolone A in black tea using selected-ion mon-
itoring (SIM) chromatograms of the standard
(compound 4) and the theaflavin fraction of black tea
This work was supported by NIH Grant PO1
CA88961. Funds to purchase the 600 MHz NMR
spectrometer at CUNY College of Staten Island were
provided by the New York State Dormitory Authority

7132
S. Sang et al. / Tetrahedron Letters 43 (2002) 7129–7133
Figure 3. Negative ion LC/ESI-MS/MS chromatograms and fragment ion mass spectra of compound 4 (A) and the peak in the
theaflavin fraction of black tea (B).
Figure 4. Proposed mechanism for the formation of compound 4.
and the New York State Higher Education Applied
Technology Program.
1999; pp. 697–724.
2. Geissman, T. A. Chemistry of Flavonoid Compounds;
Pergamon Press: Oxford, UK, 1962; pp. 468–512.
3. Runeckles, V. C.; Tso, T. C. Recent Advances in Phyto-
chemistry; Academic Press: New York, 1972; Vol. 5, pp.
247–316.
References
1. Gross, G. G.; Hemingway, R. W.; Yoshida, T. Plant
Polyphenols 2: Chemistry, Biology, Pharmacology, Ecol-
ogy; Kluwer Academic/Plenum Publishers: New York,
4. Roberts, E. A. H. Two and A Bud 1962, 9, 3–8.
5. Millin, D. J.; Crispin, D. J.; Swaine, D. J. Agric. Food
Chem. 1969, 17, 717–722.

S. Sang et al. / Tetrahedron Letters 43 (2002) 7129–7133
7133
6. Powell, C.; Clifford, M. N.; Opie, S.; Robertson, A.;
Gibson, C. J. Sci. Food Agric. 1992, 63, 77–80.
7. Takino, Y.; Imagawa, H.; Horikawa, H.; Tanaka, A.
Agric. Biol. Chem. 1964, 28, 64–71.
8. Takino, Y.; Imagawa, H. Agric. Biol. Chem. 1964, 28,
125–130.
12. Ozawa, T.; Kataoka, M.; Morikawa, K.; Negishi, C.
Biosci. Biotechnol. Biochem. 1996, 60, 2023–2027.
13. Subramanian, N.; Venkatesh, P.; Ganguli, S.; Sinkar, V.
P. J. Agric. Food Chem. 1999, 47, 2571–2578.
14. Tanaka, T.; Mine, C.; Inoue, K.; Matsuda, M.; Kouno, I.
J. Agric. Food Chem. 2002, 50, 2142–2148.
15. Wan, X. C.; Nursten, H. E.; Cai, Y.; Davis, A. L.;
Wilkins, J. P. G.; Davies, A. P. J. Sci. Food Agric. 1997,
74, 401–408.
9. Nakagawa, M.; Torii, H. Agric. Biol. Chem. 1965, 29,
278–284.
10. Davis, A. L.; Cai, Y.; Davies, A. P. Magn. Reson. Chem.
16. Lewis, J. R.; Davis, A. L.; Cai, Y.; Davies, A. P.;
Wilkins, J. P. G.; Pennington, M. Phytochemistry 1998,
49, 2511–2519.
1995, 33, 549–552.
11. Roberts, E. A. H. J. Sci. Food Agric. 1958, 9, 212–
216.