Journal of Medicinal Plants Research Vol. 3(10), pp.

736-757, October, 2009

Available online at http://www.academicjournals.org/JMPR
ISSN 1996-0875© 2009 Academic Journals

Full Length Research Paper

Assessment of the DPPH and -glucosidase inhibitory

potential of gambier and qualitative identification of
major bioactive compound
F. B. Apea-Bah1*, M. Hanafi2, R. T. Dewi2, S. Fajriah2, A. Darwaman2, N. Artanti2, P. Lotulung2,
P. Ngadymang2 and B. Minarti2
1
Biotechonology and Nuclear Agriculture Research Institute, Ghana Atomic Energy Commission, P. O. Box LG 80,
Legon, Ghana.
2
Research Centre for Chemistry, Indonesian Institute of Sciences, Kawasan Puspiptek, Serpong, Tangerang, Indonesia.
Accepted 3 July, 2009

Gambir (Uncaria gambir) and other plants belonging to the genus Uncaria have been used in traditional
medicine in southeastern Asia, Africa and South America and they have been studied widely over the
past century. Gambier, the dried leaf extract from gambir is known to have antioxidant properties and
some studies have attributed it to the presence of tannins and condensed tannins. The objective of this
study was to investigate the potential of commercial gambier on the Indonesian market as a scavenger
of reactive free radicals, evaluate its ability to inhibit -glucosidase and determine the bioactive
compound responsible for these activities. An ethanolic extract of commercial gambier was extracted
with ethyl acetate. The ethanol and ethyl acetate extracts as well as the aqueous extract after ethyl
acetate extraction and residue from ethanol extraction were tested for free radical scavenging activity
using 2,2-diphenyl-1-picrylhydrazyl (DPPH). They were also tested for -glucosidase inhibitory activity.
The extracts were then studied using reverse phase HPLC, LCMS and NMR to identify the bioactive
compound. It was observed that all the extracts had high activity for DPPH inhibition but moderate
activity for inhibiting -glucosidase in vitro. Apart from the aqueous extract, 92% DPPH inhibition by the
extracts was achievable at 30 g/ml. The ethanol and ethyl acetate extracts had significantly higher (p <
0.01) DPPH inhibitory activity than the aqueous extract. IC50 of the organic extracts and residue ranged
between 13.8 to 16.2 g/ml for DPPH inhibition while that of the aqueous extract was 27.4 g/ml. With
regards to -glucosidase inhibition, however, IC50 range of 15.2 and 49.5 g/ml was recorded. Catechin
was identified as the major bioactive compound present.

Key words: Uncaria gambir, gambier, DPPH, alpha-glucosidase, HPLC, LCMS, NMR

INTRODUCTION

Gambir (Uncaria gambir) is a vine belonging to the family
Rubiaceae and genus Uncaria (Heitzman et al., 2005).
Species of the genus Uncaria have many tradi-tional
medicinal uses including treatments for wounds and
ulcers, fevers, headaches, gastrointestinal illnesses and
bacterial and fungal infections (Aquino et al., 1991; Cerri
et al., 1988; Chang et al., 1989; Lemaire et al., 1999;
Rizzi et al., 1993; Wurm et al., 1998). In Malaysia,
Singapore, Borneo and Sumatra where U. gambir is com-
mon or indigenous, aqueous extracts of the leaves are
*Corresponding author. E-mail: franklinapeabah@yahoo.com.
processed and used for medicinal purposes (Ahmed et
al., 1978; Das and Griffiths, 1967; Phillipson et al., 1978).
The aqueous extracts of leaves and young twigs of U.
gambir are used traditionally for the treatment of diarrhea
and dysentery. They may also be used as gargle for
treatment of sore throat (Taniguchi et al., 2007). Dried
hooks of some Uncaria species have also been integral
components in traditional oriental medicines, where they
are used as spasmolytics, analgesics and sedatives for
symptoms associated with nervous disorders and in the
treatment of hypertension (Heitzman et al., 2005).
Gambier, the dried leaf extract from U. gambir is
believed to have antioxidant properties which are attri-
buted to the presence of tannins and condensed tannins

(Desmarchelier et al., 1997; Wirth and Wagner, 1997).
Gambir was formerly cultivated extensively in Malaysia
and Indonesia as a commercial source of tanning mate-
rials (Das and Griffiths, 1967) and catechin is the most
abundant polyphenolic constituent in the plant (Taniguchi
et al., 2007; Das and Griffiths, 1967). Other polyphenols
extracted and identified in gambir include gambiriins
which are chalcane-flavan dimers, (+)-epicatechin and
dimeric proanthocyanidins (Taniguchi et al., 2007). The
presence of catechin in green tea and fermented tea has
been associated with health protective and cancer pre-
ventive properties in animal models due to their antioxi-
dant activity (Sang et al., 2002). Of particular importance
in cancer prevention is the ability of catechin to scavenge
reactive oxygen species that play a role in carcinogenesis
(Sang et al., 2002).
Diabetes mellitus is a serious chronic metabolic disor-
der characterized by high blood glucose levels (Corry and
Tuck, 2000). Worldwide, the number of patients is rapidly
growing with increasing obesity (King et al., 1998) and
ways to control postprandial hyperglycemia involve medi-
cation with dietary restriction and body exercise (Goke
and Herrmann-Rinke, 1998). Among the several thera-
peutic approaches is retarding glucose absorption using
inhibitors of carbohydrate-hydrolyzing enzymes such as
-amylase and -glucosidase (Holman et al., 1999; Saito
et al., 1998; Toeller, 1994). Although Acarbose (Schmidit
et al., 1977), miglitol (Standl et al., 1999), voglibose
(Saito et al., 1998) and nojirimycin (Hansawasdic et al.,
2000) are known to be potent inhibitors of -glucosidase,
further screening for -glucosidase inhibitors from natural
sources such as plants and microorganisms will help in
reducing overdependence on synthetic drugs. Pine bark
extract which is rich in polyphenols such as catechin,
quercetin, dihyhydroquercetin, taxifolin and phenolic
acids (Markham and Porter, 1973; Packer et al., 1999), is
reportedly effective in suppressing postprandial hyper-
glycemia in diabetics (Kim et al., 2005). Even though
gambir is also rich in polyphenols, it is not used
traditionally in treating diabetes mellitus. There is the
need to study its potential as an inhibitor of -glucosidase.
In the present study, we investigated the free radical
scavenging and -glucosidase inhibitory potential of
commercial gambier sold in a local Indonesian market in
Tangerang. The major compound responsible for the
bioactivity was qualitatively identified using reversed-
phase HPLC, LCMS and NMR.
MATERIALS AND METHODS
Materials
Distilled water, methanol, deuterated methanol (CD3OD), ethanol,
ethyl acetate, hexane, chloroform, sulphuric acid in methanol, 2,2-
diphenyl-1-picrylhydrazyl, catechin standard (Sigma-Aldrich Chemi-
cal Co.), -glucosidase (Sigma-Aldrich Chemical Co.), p-
nitrophenyl- -D-glucopyranoside (Sigma-Aldrich Chemical Co.),
Na2CO3 (E. Merck), Na2HPO4.2H2O (E.Merck), KH2PO4 (E.Merck),
bovine serum albumin (E.Merck), dimethyl sulphoxide (E. Merck),
Apea-Bah et al. 737
phosphate buffer pH 7.0, analytical balance (Mettler Toledo AT400),
rotary evaporator (Buchi Rotavapor R-124), column chromatograph,
thin layer chromatograph plates, forced convection oven (Memmert
854 Schwabach, Germany), UV-Visible spectrophotometer (Hitachi
U-2000, Japan), HPLC (Shimadzu, Japan), liquid chromatograph-
mass spectrometer (LC: Hitachi L 6200; Mariner Biospectrometry,
Electrospray Ionisation System), micropipettes, glassware, water
bath, stopwatch and other supplies.
Sample preparation
A dried leaf extract of gambir was purchased from a local market in
Serpong, a surburb of Tangerang on the Java Island of Indonesia.
10 g of the gambier (dried leaf extract of gambir) was dispersed in
100ml of 95% aqueous ethanol and stirred. It was then filtered
through Whatmann No.1 filter paper into a conical flask. The resi-
due from the ethanol filtration was dried for further analysis while a
portion of the ethanol filtrate was evaporated on a rotary evaporator
(50°C, 90 -100 rpm) to obtain the ethanol extract. The other portion
of the ethanol filtrate was further extracted with ethyl acetate,
resulting in ethereal and aqueous layers which were separated
using a separating funnel. The ethereal and aqueous layers were
concentrated using rotary evaporator (45°C, 90 - 100 rpm for
ethereal layer and 53°C, 90 – 100 rpm for aqueous fraction) to
obtain the ethyl acetate extract and the aqueous extract (after ethyl
acetate extraction). The samples were dried in an oven at
temperature of about 45°C overnight (24 h).
DPPH radical scavenging activity
The DPPH radical scavenging activity was determined using the
method of Hatano et al. (1988). 4 mg of each sample was weighed
and added to 4 ml of methanol. 4 concentrations, 10, 50, 100 and
200 g/ml, were prepared from each sample solution using
methanol as solvent and 500 l DPPH was added giving a total
volume of 2.5 ml per sample concentration. Standards were also
prepared at similar concentrations using standard catechin while
the blank was prepared using methanol and DPPH without any
sample. They were then incubated in a water bath at 37°C for 30
min after which absorbance of samples and standards were read
against the blank at 517 nm. The percent inhibition was estimated
as % inhibition = [(C - S)/C] x 100%, where S is the sample absor-
bance and C is absorbance of the blank. Scatter diagrams were
plotted and linear regression estimated using the equation y = ax+b,
where y is % inhibition and x is concentration ( g/ml). IC50 was
calculated as the concentration that caused 50% inhibition of DPPH.
-Glucosidase inhibitory activity
The -glucosidase inhibitory activity was determined using the
method described by Sutedja (2005). The enzyme solution was
prepared by dissolving 0.5 mg -glucosidase in 10 ml phosphate
buffer (pH 7.0) containing 20 mg bovine serum albumin. It was
diluted further to 1:10 with phosphate buffer just before use. Sam-
ple solutions were prepared by dissolving 4 mg sample extract in
400 l DMSO. 3 concentrations; 6.25, 12.5 and 25 g/ml were
prepared and 5 l each of the sample solutions or DMSO (sample
blank) was then added to 250 l of 20 mM p-nitrophenyl- -D-
glucopyranoside and 495 l of 100 mM phosphate buffer (pH 7.0). It
was pre-incubated at 37°C for 5 min and the reaction started by
addition of 250 l of the enzyme solution, after which it was
incubated at 37°C for exactly 15 min. 250 l of phosphate buffer
was added instead of enzyme for blank. The reaction was then
stopped by addition of 1000 l of 200 mM Na2CO3 solution and the
amount of p-nitrophenol released was measured by reading the
738 J. Med. Plant. Res.

absorbance of sample against sample blank (containing DMSO with 120

no sample) at 400 nm using UV-visible spectrophotometer. Stan-
dard solutions of same concentrations were also prepared using
100
‘koji’, an extract from Aspergillus terreus. Scatter diagrams were
plotted and linear regression estimated. IC50 was calculated as the
concentration that caused 50% inhibition of enzyme activity. 80

% Inhibition
60
HPLC analysis

The samples were analyzed using reverse phase HPLC to 40

y = 21.41Ln(x) - 9.2035
ascertain presence of catechin. Wavelength for maximum absorp- 2
tion of catechin was determined using the UV-visible spectropho- R = 0.8046
20
tometer. The standards and samples were then analyzed using
HPLC at 210 nm. Water and methanol (30:70) was used as the
eluent and was set to a flow rate of 1 ml/min and elution time of 20 0
min. 0 50 100 150 200 250
Concentration (ug/ml)
Identification and structure elucidation of active component
Figure 1. Percent inhibition of DPPH against catechin
A sample of the ethyl acetate extract was purified using column concentration.
chromatography with hexane, ethyl acetate and methanol in
different proportions as eluent. The purified extract and other sam-
ples were analyzed using liquid chromatograph-mass spectrometer 120
(LC: Hitachi L 6200; Mariner Biospectrometry; Electrospray
Ionisation System) to determine the molecular weights (m/z values) 100
of the major compounds. 20 l of the samples were injected and
run at a flow rate of 1 ml/min using acetonitrile: water (70:30) as % Inhibition 80
eluent and C18 column (Supelco: 150 mm x 2 mm x 5 m). Nuclear
magnetic resonance spectroscopy (JEOL) with radio frequency of 60
500 MHz was then used to elucidate the structure of the active
y = 21.614Ln(x) - 11.089
compound in the purified sample which was prepared using 40 2
deuterated methanol (CD3OD) as solvent. One and two dimensional R = 0.7863
NMR spectra were developed and analyzed. 20

0
Data analysis
0 50 100 150 200 250

Two way analysis of variance (Montgomery, 1991) was used to Concentration (ug/ml)
determine the effects of concentration and type of extraction on the
DPPH and -glucosidase inhibitory activities. Where significant
differences existed (p < 0.05), least significant difference (LSD) was Figure 2. Percent inhibition of DPPH against concentration of
used for mean comparison. Statgraphics centurion XV (Statgra- ethanol extract
phics, 2005) statistical software was used for all statistical analysis.

120
RESULTS AND DISCUSSION
DPPH Radical scavenging activity 100

Figures 1 to 4 show the linear regression curves of

80
concentration of catechin standard and gambier extracts
% Inhibition

plotted against percent inhibition of DPPH. The regres- 60

2
sion coefficient (R ) of the logarithmic trend line obtained
for the extracts and standard was between 0.79 - 0.80,
implying that 80% of the variability in DPPH inhibition can 40
y = 26.941Ln(x) - 35.809
be attributed to the concentration of the extracts. Analysis 2
R = 0.7981
of variance (ANOVA) showed concentration to signifi- 20
cantly (p < 0.01) affect percent inhibition. From the mean
comparison using least significant difference, it was 0
observed that inhibition of DPPH was significantly low at 0 50 100 150 200 250
10 g/ml concentration (Table 1) compared to all the Concentration (ug/ml)
other concentrations. The extracts from gambier were not
significantly different (p > 0.05) from the catechin stan- Figure 3. Percent inhibition of DPPH against concentration
dard with regards to DPPH inhibition. The IC50 for the of Residue after Ethanol extraction.
 Apea-Bah et al. 739

Table 1. Concentration of Catechin and extracts from Gambier and their IC50 for DPPH inhibition.

% Inhibition
Concentration
Catechin Ethyl acetate Ethanol Residue from *Mean
( g/ml)
standard extract extract ethanol extraction
a
200 92.41 91.39 90.67 91.61 91.52
a
100 92.26 92.33 91.75 92.19 92.13
a
50 92.48 92.62 92.48 92.48 92.51
b
10 31.14 21.59 29.11 14.72 24.14
IC50 ( g/ml) 15.88 20.70 16.88 24.17
*Means in a column with same superscript are not significantly different from each other (p<0.01)

120 to 10) from the logarithmic trend line. The IC50 of the ethyl
acetate extract was least and closest to that of catechin
100
standard while that of the aqueous extract was highest
80
(Table 2) indicating that the active compound is very
soluble both in ethanol and ethyl acetate.
% Inhibition

From the ANOVA, both sample type or extract and

60
y = 24.495Ln(x) - 24.218
40
2 not significantly different from each other but resulted in
R =0.794
20
0
0 50 100 150 200
Concentration (ug/ml)
Figure 4. Percent inhibition of DPPH
concentration of ethyl acetate extract.
catechin standard was 15.9 g/ml while it was in the
range of 16.9 to 24.2 g/ml for the extracts (Table 1).
Since IC50 was lowest for ethanol extract and highest for
residue, it implies that the ethanol extract had the highest
free radical scavenging activity at concentrations
between 10 and 200 g/ml, while that of the residue was
least. This suggests that the compounds present in
gambier that are respon-sible for scavenging DPPH free
radicals are very soluble in ethanol. It is worth noting from
the linear regression curves (Figures 1 to 4) that
increasing sample concen-tration above 50 g/ml did not
cause a corresponding increase in percent inhibition of
DPPH. There was there-fore the need to study
concentrations between 10 and 50 g/ml.
It was observed from studies using lower concentra-
tions of standard catechin and gambier extracts that
maximum DPPH inhibition was measured at 30 g/ml
(Table 2), except for the aqueous extract obtained after
ethyl acetate extraction of gambier, which showed per-
cent DPPH inhibition to increase correspondingly with
concentration until 50 g/ml. At the lower concentrations,
2
a coefficient of regression (R ) of 0.87 for catechin and
0.94 - 0.97 for gambier extracts were obtained (Figures 5
concentration significantly affected (p < 0.01) percent
DPPH inhibition. Concentrations of 30 to 50 g/ml were
significantly higher DPPH inhibition than 5 to 20 g/ml
(Table 2). The lower concentrations of 5, 10 and 20 g/ml
were all significantly different from each other. This
means therefore that for optimum free radical scavenging
250
using organic extracts of commercial gambier, a
concentration of 30 g/ml is adequate.
With regards to the samples, however, ethanol and
against ethyl acetate extracts were statistically not different (P <
0.01) from the catechin standard. They were also not
significantly different from the residue from ethanol ex-
traction. They all, however, showed significantly higher
potential for DPPH inhibition than the aqueous extract
(Table 2). This suggests that the major bioactive
compound has lower solubility in water than in ethanol
and ethyl acetate.
-Glucosidase inhibitory activity
Figures 11 to 16 show the linear regression curves of
concentration of koji extract and gambier extracts plotted
against percent inhibition of -glucosidase. The regres-
2
sion coefficient (R ) of the trend lines obtained for the
extracts and standard was between 0.94 - 0.99, implying
that 94 to 99% of the variability in -glucosidase inhibition
can be attributed to the concentration of the extracts.
Since koji, an extract from Aspergillus terreus, is known
to have high activity for -glucosidase inhibition, the
gambier extracts were compared to it for -glucosidase
inhibitory activity. Three concentrations, 6.25, 12.5 and 25
g/ml, were studied. For all extracts, percent -glucosi-
dase inhibition increased with increasing concentration.
Koji extract showed a mean of 74% inhibition while
ethanol, ethyl acetate and the aqueous extraction caused
740 J. Med. Plant. Res.

Table 2. Percent DPPH inhibition by catechin and Gambier Extracts at lower concentrations.

% Inhibition
Concentration
Catechin Ethanol Ethyl acetate Residue after Aqueous
( g/ml) Mean
Standard Extract extract ethanol extraction extract
h
50 91.89 91.77 92.63 92.24 92.44 92.20
h
40 92.48 91.50 92.44 92.63 73.28 88.47
h
30 92.99 91.93 92.56 92.36 67.80 87.53
i
20 92.60 68.07 70.66 54.24 33.01 63.71
j
10 35.71 27.68 33.32 17.22 14.16 25.62
k
5 14.32 5.90 6.72 4.45 0.61 6.40
m mn mn n p
Mean 70.00 62.81 64.72 58.86 46.88
IC50 ( g/ml) 11.62 14.54 13.78 16.20 27.35
*Means with same superscript are not significantly different from each other (p < 0.01).

100 100
90 90
80 80
% Inhibition

70 70

% Inhibition
60 60
y = 37.549Ln(x) - 42.087
50 50
40
2
R = .8690 40 y = 41.554Ln(x) - 61.231
30 30
20
2
R = .9571
20
10 10
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Concentration (ug/ml) Concentration (ug/ml)

Figure 5. Linear regression curve of percent DPPH Figure 7. Linear regression curve of percent DPPH
inhibition at lower concentration of catechin standard. inhibition at lower concentration of ethanol extract.

100
90 100
80 90
70 Catechin 80
% Inhibition

60 70
EtOH ext
% Inhibition

50 60
40 EtOAc ext 50
40 y = 44.23Ln(x) - 73.169
30 Residue
20 30 2
Water ext R = .9393
10 20
10
0
0 10 20 30 40 50 60 0
0 10 20 30 40 50 60
Concentration (ug/ml)
Concentration (ug/ml)
Figure 6. Linear Regression curve of percent DPPH inhibition
lower concentrations of catechin standard and gambier Figure 8. Linear Regression Curve of Percent DPPH
extracts. inhibition at lower concentration of residue from ethanol
extraction.

extracts caused below 50% inhibition of -glucosidase moderate -glucosidase inhibitory activity in vitro. 2 way
activity (Table 3). This suggests that the extracts have ANOVA showed both the extracts and their concentrations
 Apea-Bah et al. 741

100 100
90 90
80 80
70
70
% Inhibition

60
60
y = 40.723Ln(x) - 56.835

% Inhibition
50
50
40 2
R = .9607 40
30
20 30 y = 1.1567x + 32.46
10 20 2
R = .9953
0 10
0 10 20 30 40 50 60 0
Concentration (ug/ml) 0 5 10 15 20 25 30

Concentration (ug/ml)
Figure 9. Linear regression curve of percent DPPH
inhibition at lower concentration of ethyl acetate extract. Figure 12. Linear regression curve of percent -
glucosidase inhibition at lower concentration of ethanol
extract.
100
90
80
100
70
60 90
% Inhibition

50 y = 2.0549x - 6.2024 80
40 70
R2 = .9709
30 60 y = 15.416Ln(x) - 26.974
% Inhibition

20 50 R2 = 0.9981
10 40
0 30
0 10 20 30 40 50 60 20
Concentration (ug/ml) 10
0
Figure 10. Linear regression curve of percent DPPH
inhibition at lower concentration of water extract. 0 5 10 15 20 25 30
concentration (ug/ml)

100 Figure 13. Linear Regression curve of percent -

glucosidase inhibition at lower concentration of
90
residue of ethanol extract
80
70
% inhibition

60
50 to significantly (p < 0.01) affect percent inhibition.
y = 22.368Ln(x) + 18.546 Inhibition of -glucosidase was significantly highest for
40
2
R = .9383 koji and lowest for residue after ethanol extraction. The
30
other solvent extracts did not differ significantly from each
20 other. Percent inhibition was also significantly highest at
10 25 g/ml concentration compared to the lower
0 concentrations of 6.25 and 12.5 g/ml, which did not
0 5 10 15 20 25 30 significantly differ from each other (Table 3).
concentration (ug/ml) Increasing the concentrations of the extracts up to 50
g/ml did not improve their -glucosidase inhibitory
Figure 11. Linear regression curve of percent -glucosidase activity (Figure 17). The IC50 for koji extract was 4.08
inhibition at lower concentration of ‘koji’ extract from Aspergillus g/ml while it was in the range of 15.16 g/ml to 49.48
terreus extract. g/ml for the extracts (Table 3). Since IC50 was lowest for
 742 J. Med. Plant. Res.

Table 3. Concentration of Koji and Extracts from Gambier and their IC50 for -glucosidase inhibition.

Concentration Ethyl acetate Residue after Ethanol Aqueous

Koji extract Ethanol extract *Mean
( g/ml) extract extraction extract
c
25.0 88.71 61.09 45.77 22.38 66.73 56.94
d
12.5 80.85 47.78 43.15 12.50 38.51 44.56
d
6.25 53.43 39.11 40.73 1.01 35.08 33.87
e g g f g
*Mean 74.33 49.33 43.21 11.96 46.77
IC50 ( g/ml) 4.08 15.16 40.65 49.48 16.41
*Means with same superscripts are not significantly different from each other (p>0.01). ¥Koji is an extract from Aspergillus terreus.

100 100
90 90
Koji
80 80
70 Residue EtOH
70 ext
60

% Inhibition
Water ext
60
% Inhibition

50
50 ext EtOH
40
40 30 EtOAC ext
30 y = 3.6358Ln(x) + 34.029 20
20 10
R2 = 0.9995
0
10
0 10 20 30
0
Concentration (ug/ml)
0 5 10 15 20 25 30

Concentration (ug/ml) Figure 16. Linear regression curve of percent -

glucosidase inhibition at lower concentrations of gambier
extracts.
Figure 14. Linear regression curve of percent -
glucosidase inhibition at lower concentration of ethyl
acetate extract
100
90
80
100 70 Pure Gambier
% Inhibition

90 60 EtOAc ext
80 50 Residue
70 40
EtOH ext
30
60 Water ext
20
50
% Inhibition

10
40 0
30 y = 1.7696x + 20.968 0 10 20 30 40 50 60
2
20 R = 0.9459
Concentration (ug/ml)
10
Figure 17. Linear regression curve of percent -glucosidase
0 inhibition at higher concentrations of gambier extracts.
0 5 10 15 20 25 30
Concentration (ug/ml)
ethanol extract and highest for residue from ethanol
Figure 15. Linear regression curve of percent - extract, it implies that the ethanol extract had the highest
glucosidase inhibition at lower concentration of water -glucosidase inhibitory activity in vitro while that of the
extract. residue from ethanol extraction was least.
 Apea-Bah et al. 743

TIC=>NR(2.00)
100 T1.5 2.8E+4

T1.3

T1.9
90
% Intensity

T6.5

80

700
2.4 4.8 7.2 9.6 12.0

Retention time (min)

Figure 18. LCMS spectrum for ethanol extract showing three peaks at T1.3, T1.5, T1.9.

Mariner Spec /34:34 (T /1.26:1.26) -28:28 (T -1.26:1.26) ASC=>MC=>NR(2.00)[BP = 145.6, 143]

145.63 142.8
100

90

80

70

60
% Intensity

329.10
50

40 147.62
30

20 330.13
619.67
10 185.64 626.60
114.80 187.59 260.33 331.10 403.10 478.36 775.39 911.93
551.16
0 0
108.0 298.2 488.4 678.6 868.8 1059.0
Mass (m/z)

Figure 19. m/z spectral lines of compounds present in peak T1.3 for ethanol extract.

HPLC analysis
Preliminary work on the wavelength for maximum absorp-
tion using the UV-visible spectrophotometer resulted in 5
ppm standard catechin giving best results at 209.4 nm.
All HPLC analyses were therefore done at 210 nm wave-
length. Catechin standard gave maximum peak at reten-
tion time range of 3.258 to 3.271 min (Table 4) for the
various concentrations studied.
At sample concentrations of 500 and 1000 g/ml, for all
the extracts, retention times were similar (Table 5) to that
of standard catechin indicating that the bioactive
compound could be catechin.
LCMS analysis
The extracts upon analysis using liquid chromatograph-
744 J. Med. Plant. Res.

Table 4. Retention time of standard catechin at 210 nm using HPLC.

Retention time (min) Concentration ( g/ml) Area

3.266 5 1650628
3.263 10 3495729
3.264 25 7766748
3.271 50 15226023
3.266 100 28116464
3.263 500 70398534
3.258 1000 83927104

Table 5. HPLC Retention time of Samples at higher concentration at 210nm

Sample Concentration ( g/ml) Retention time Area

500 3.299 60923005
Ethyl Acetate
1000 3.275 70239090
500 3.274 62442591
Ethanol
1000 3.266 75125886
500 3.256 65592871
Water
1000 3.227 92219895
500 3.277 79735339
Residue
1000 3.265 92755082

Mariner Spec /34:34 (T /1.26:1.26) -28:28 (T -1.26:1.26) ASC=>MC=>NR(2.00)[BP = 145.6, 143]

145.63 142.8
100

90

80

70

60
% Inte nsity

329.10
50

40 147.62
30

20 330.13
619.67
10 185.64 626.60
114.80 187.59 260.33 331.10 403.10 478.36 775.39 911.93
551.16
0 0
108.0 298.2 488.4 678.6 868.8 1059.0
Mass (m/z)

Figure 19. m/z spectral lines of compounds present in peak T1.3 for ethanol extract.

mass spectrometer (LCMS) showed 2 or 3 board peaks NMR analysis

(Figures 18 to 37) each comprising different compounds
and hence different m/z ratios. Notable among the m/z
values was 291.23 with its major peak, which is indicative
of catechin. NMR was used to then elucidate and confirm
the structure of the major bioactive compound.
1 13
Plates 1 to 5 show the H, C, DEPT, HMQC and HMBC
spectra generated from the NMR analysis at 500MHz.
1
From the H NMR spectra (Plate 1), 2 double doublet
1
peaks each with one proton ( H) are observed at chemi-
 Apea-Bah et al. 745

Mariner Spec /39:40 (T /1.45:1.49) -35:36 (T -1.45:1.49) ASC=>MC=>NR(2.00)[BP = 291.2, 104]

291.23 103.5
100

90

80

70

60
% In te n s ity

50

40

30
292.24
20

10
214.47 293.18 333.12 391.04 448.45 532.14 579.88 619.15 667.95
0 0
214.0 319.4 424.8 530.2 635.6 741.0
Mass (m/z)

Figure 20. m/z spectral lines present in peak T1.5 for ethanol extract

Mariner Spec /49:51 (T /1.84:1.91) -44:46 (T -1.84:1.91) ASC=>MC=>NR(2.00)[BP = 291.2, 24]

291.23 23.7
100

90

80

70

60
% Intensity

50

40

30 292.19
20

10 241.40 293.29 390.98

242.12 351.20 438.36 479.34 522.21 577.88 620.97 661.98 700.86
0 0
214.0 319.4 424.8 530.2 635.6 741.0
Mass (m/z)

Figure 21. m/z spectral lines present in peak T1.9 for ethanol extract.

1
cal shifts ( ), 2.48 - 2.53 ppm (J-coupling, 7.95Hz) and The Hs are in axial and equatorial positions at the
2.82 - 2.87 ppm (J-coupling 5.50 Hz). These peaks repre- bonded site (Plate 1). At 3.98 - 4.0ppm, a quartet peak
1
sent an aliphatic group having a single near H and (J-coupling, 7.95Hz, 2.45 Hz) is observed indicating 3
1 1
another longer range H coupling per each double doublet. neighboring H-atoms. At 4.45 - 4.58 ppm, there is a
746 J. Med. Plant. Res.

TIC=>NR(2.00)
T1.5 2.6E+4
100

T1.9
% Intensity

90
T23.6

80
0 6 12 18 24 30
Retention Time (Min)

Figure 22. LCMS spectrum for ethyl acetate extract showing three peaks at T1.2, T1.5, T1.9.

Mariner Spec /33:34 (T /1.22:1.26) -26:27 (T -1.22:1.26) ASC=>MC=>NR(3.00)[BP = 619.7, 30]

619.67 30.2
100

90

80
145.64
70

60
% In t e n s it y

620.65
50 329.11

40 625.65
30 147.63
333.12 641.60
20
163.77 330.07 623.68
10 624.65
334.03
114.83 164.80 214.51 486.82
0 0
109.0 244.8 380.6 516.4 652.2 788.0
Mass (m/z)

Figure 23. m/z spectral lines present in peak T1.2 for ethyl acetate extract.

1 1
doublet peak (J-coupling, 7.95 Hz) with one H indicating
1
only one neighboring H and there is likelihood of a
neighboring O-atom at this value. Both the quartet and
doublet having J-coupling of 7.95 Hz are in long range
coupling with the double doublet at 2.48 - 2.53 ppm.
The deuterated methanol (CD3OD) produced its peak at
3.31 ppm and another peak for water produced at 4.67
- 4.95ppm (Plate 1). At 5.86 and 5.94 ppm, there are 2
doublets, each with one H, having meta coupling (J-
coupling 2.45 Hz) with each other in an aromatic ring. At
6.71 - 6.73 ppm, there is a double doublet (J-coupling,
1
8.8 Hz, 1.85 Hz) representing 2 neighboring Hs in meta
and ortho positions in an aromatic ring. The neighboring
1
ortho H-atom produced a doublet at 6.76 - 6.78 ppm (J-
1
coupling, 8.55 Hz) while the neighboring meta H (J-
coupling, 1.85 Hz) produced a doublet at 6.84 ppm (Plate 2).
 Apea-Bah et al. 747

Mariner Spec /39:39 (T /1.45:1.45) -35:35 (T -1.45:1.45) ASC=>MC=>NR(2.00)[BP = 291.2, 127]

291.23 127.0
100

90

80

70

60
% Inte ns ity

50

40

30
292.23
20

10
163.91 214.47 292.98 579.78 629.69
109.82 391.07 467.80 524.93 706.45
0 0
109.0 244.8 380.6 516.4 652.2 788.0
Mass (m/z)

Figure 24. m/z spectral lines present in peak T1.5 for ethyl acetate extract.

Mariner Spec /50:52 (T /1.87:1.95) -26:27 (T -1.87:1.95) ASC=>MC=>NR(2.00)[BP = 291.2, 98]

291.23 98.0
100

90

80

70

60
% Intensity

50

40

30

20
205.49
10 207.48 292.66
113.71 391.19 474.44 579.84 981.17
743.69 864.75
0 0
99.0 319.2 539.4 759.6 979.8 1200.0
Mass (m/z)

Figure 25. m/z spectral lines present in peak T1.9 for ethyl acetate extract

13
Comparing the C and DEPT spectra (Plate 3) reveal 156.91, 157.57 and 157.78 ppm. values above 100
13
CH2 as inverted DEPT peak at 28.56 ppm and quaternary ppm in the C spectra indicated aromatic C-atoms while
13
C-atoms, being present in C spectra but absent in values of about 150 ppm indicated aromatic C-atoms
DEPT spectra, at 100.94, 132.19, 146.25 ppm, 146.28, bonded to O-atom. It is therefore likely that C-atoms at
748 J. Med. Plant. Res.

TIC=>NR(2.00)
T1.5 3.6E+4
100

90
% Intensity

80

70
T7.0 T11.9
T2.8

60
0 4 8 12 16 20
Retention Time (Min)

Figure 26. LCMS spectrum for residue from ethanol extraction showing broad peak at T1.5 and small peak at T2.8

Mariner Spec /31:35 (T /1.15:1.30) -20:25 (T -1.15:1.30) ASC=>NR(2.00)[BP = 328.5, 44]

328.45 44.3
100

90

80

70

60
% Intensity

50 618.76
187.04
40 185.16
624.72
30

20
268.64 620.74
10 147.19 359.32 485.56 908.93
154.35 236.86 332.40 643.62 926.85
505.36 730.17 832.01 1020.06
0 0
99.0 319.2 539.4 759.6 979.8 1200.0
Mass (m/z)

Figure 27. m/z spectral lines present in peak T1.3 for residue from ethanol extraction.

values from 146.25 to 157.78 ppm are bonded to an O- 4.45 - 4.58 ppm was bonded to the C-atom at 82.86 ppm
1
atom. while the H-atoms with 2 doublets at 5.86 and 5.94 ppm,
From the HMQC spectra (Plate 4), it was observed that were bonded to C-atoms at 95.60 and 96.41 ppm. The
1 1
the two H-atoms with double doublet peaks at 2.48 – double doublet H at 6.71 - 6.73 ppm was bonded to the
1
2.53 ppm (Plate 1) were bonded to the C-atom at 28.56 C at 120.16 ppm while the doublet H at 6.76 - 6.78 ppm
1 1
ppm while the quartet H at 3.98 - 4.0ppm was bonded to was bonded to the C at 116.23 ppm. The doublet H at
1
the C-atom at 68.81 ppm. The H with doublet peak at 6.84 ppm was bonded to the C-atom at 115.34 ppm.
 Apea-Bah et al. 749

Mariner Spec /40:42 (T /1.49:1.57) -23:27 (T -1.49:1.57) ASC=>MC=>NR(2.00)[BP = 291.2, 191]

291.2347 190.8
100

90

80

70

60
% Intensity

50

40
329.1065
30

20

10 165.6618 619.6706
292.5783
139.7067 454.4737 577.8330 735.6859 879.6150 1024.0349
0 0
99.0 319.2 539.4 759.6 979.8 1200.0
Mass (m/z)

Figure 28. m/z spectral lines present in peak T1.5 for residue from ethanol extraction.

Mariner Spec /73:74 (T /2.76:2.80) -66:68 (T -2.76:2.80) ASC=>NR(2.00)[BP = 225.1, 72]

225.06 72.4
100

90

80

70

60
% Intensity

50

40

30

20 226.06

10 163.29 448.03
140.22 208.91 246.90 290.80 390.38
348.37 456.25 491.71 572.08
0 0
138 232 326 420 514 608
Mass (m/z)

Figure 29. m/z spectral lines present in peak T2.8 for residue from ethanol extraction.

The HMBC spectra (Plate 5) show the long range
1 1
coupling between the H-atoms and the C-atoms. The H-
atom at 4.57 ppm was coupled to the C-atoms at 28.56,
1
68.81, 115.34, 116.23, 132.20 and 156.91 ppm. The H-
atoms at 5.86 and 5.94 ppm were coupled to C-atoms at
100.94, 95.60, 96.41 and 157 ppm while the double
1
doublet H-atom at 6.72 ppm was coupled to 115.34,
116.23 and 146 ppm. Its ortho-coupled neighbor at 6.77
ppm had similar C-atom coupling in addition to the C-
atom at 132.20 ppm while the meta-coupled neighbor at
6.84 ppm had similar C- atom coupling in addition to the
C-atom at 120.16ppm.
750 J. Med. Plant. Res.

TIC=>NR(2.00)
T1.5 3.1E+4
100

90
% Inte ns ity

80
T6.6 T9.7

70
0 2.2 4.4 6.6 8.8 11.0
Retention Time (Min)

Figure 30. LCMS spectrum for aqueous extract showing broad peak at T1.5.

Mariner Spec /34:36 (T /1.26:1.34) -23:25 (T -1.26:1.34) ASC=>NR(2.00)[BP = 328.5, 130]

328.45 129.6
100

90

80

70

60
% Inte nsity

50

40

30 640.69
145.20
20
624.79
10 147.19 618.76
329.87
529.35 614.76 908.90
109.24 201.09 447.83 741.66 1005.23
0 0
99.0 319.2 539.4 759.6 979.8 1200.0
Mass (m/z)

Figure 31. m/z spectral lines present in peak T1.3 for aqueous extract.

From the NMR analyses, the major bioactive compound in the purified sample was identified as (+)-catechin (Figure 38).
 Apea-Bah et al. 751

Plate 1. 1H NMR of Purified Ethyl acetate extract of Gambier using CD3OD at 500 MHz showing chemical shifts of 2.4ppm to 5.1
ppm.

Mariner Spec /38:40 (T /1.42:1.49) -23:25 (T -1.42:1.49) ASC=>MC=>NR(2.00)[BP = 329.1, 91]

329.11 90.6
100

90

80

70

60
% Intensity

291.24
50

40

30 118.79 330.10
20
145.64
292.23
10 147.63 573.93
331.07 641.57
258.41 406.71 486.38 579.78 723.27
0 0
102.0 250.8 399.6 548.4 697.2 846.0
Mass (m/z)

Figure 32. m/z spectral lines present in peak T1.5 for aqueous extract.
752 J. Med. Plant. Res.

Plate 2. 1H NMR of purified ethyl acetate extract of Gambier using CD3OD at 500 MHz showing chemical shifts of 5.8ppm
to 7.0 ppm.

Plate 3. 13C NMR and dept of purified ethyl acetate extract of gambier using CD3OD at 500 MHz.
 Apea-Bah et al. 753

Plate 4. HMQC NMR of purified ethyl acetate extract of Gambier using CD3OD at 500 MHz.

Mariner Spec /48:50 (T /1.80:1.88) -22:26 (T -1.80:1.88) ASC=>MC=>NR(4.00)[BP = 291.2, 43]

291.23 43.2
100

90

80

70

60
% Intensity

50

40

30
329.11
20

10 118.79 330.10
163.75 207.47 577.82 620.65
116.71 254.51 391.24 474.48 525.94
0 0
105.0 221.4 337.8 454.2 570.6 687.0
Mass (m/z)

Figure 33. m/z spectral lines present in peak T1.9 for aqueous extract.
754 J. Med. Plant. Res.

Plate 5. HMBC NMR of purified ethyl acetate extract of gambier using CD3OD at 500 MHz.

TIC=>NR(2.00)
T1.5 3.5E+4
100
T1.3

90
% Inte ns ity

80

T2.0

70 T5.4

60
0 2 4 6 8 10
Retention Time (Min)

Figure 34. LCMS spectrum for purified extract from ethyl acetate fraction showing three peaks at T1.3, T1.5 and T2.0.
 Apea-Bah et al. 755

Mariner Spec /35:36 (T /1.30:1.34) -26:27 (T -1.30:1.34) ASC=>MC=>NR(2.00)[BP = 187.5, 243]

187.53 243.4
100

90

80

70

60 269.24
% Inte ns ity

50

40
350.88
30
185.66
20 432.48
105.71 514.04
10 270.23 595.59
188.52 351.86 433.45 677.11
121.66 516.26 758.68 840.18
0 0
101.0 261.6 422.2 582.8 743.4 904.0
Mass (m/z)

Figure 35. m/z spectral lines present in peak T1.3 for purified extract from ethyl acetate fraction.

Mariner Spec /39:40 (T /1.45:1.49) -37:37 (T -1.45:1.49) ASC=>MC=>NR(3.00)[BP = 291.3, 69]

291.24 68.6
100

90

80

70

60
% Intensity

50

40

30
145.63 292.22
20
185.65
10 147.64 601.72
148.34 197.68 268.96 350.42 412.84 504.86 565.31 623.58 676.76
0 0
107.0 238.4 369.8 501.2 632.6 764.0
Mass (m/z)

Figure 36. m/z spectral lines present in peak T1.5 for the purified extract.

Conclusion in vitro free radical scavenging using organic extracts of

commercial gambier, 30 g/ml is adequate. The study
From the study it is concluded that ethanolic and ethyl has shown that the ethanolic extract of gambier and
acetate extracts of commercial gambier, as well as their aqueous residue after ethyl acetate extraction, have
residues on extraction have high ability to scavenge moderate in vitro -glucosidase inhibitory activity.
reactive free radicals such as is produced by DPPH and Catechin was identified and confirmed as the major
hence, have high antioxidant activity in vitro. For optimal bioactive compound in gambier.
756 J. Med. Plant. Res.

Mariner Spec /51:53 (T /1.91:1.99) -44:47 (T -1.91:1.99) ASC=>MC=>NR(3.00)[BP = 291.2, 34]

291.21 34.1
100

90

80

70

60
% Intensity

50
205.47
40 163.73
30 207.47
20 292.17
209.46
10 390.92
140.65 214.54 293.19 601.60
454.96 498.39
0 0
115.0 232.8 350.6 468.4 586.2 704.0
Mass (m/z)

Figure 37. m/z spectral lines present in peak T2.0 for the purified extract.

OH

OH
146.28
(6.83,d)
115.34 146.25

(5.86, d) 132.20 116.23

HO O 120.16
(6.77,d)
95.60
157.78 156.91 82.86 (4.57) (6.72,d)

96.41 100.94 68.81

(5.94, d)
157.57 28.56 (3.98, q)
(2.52) (2.83) OH

OH
Figure 38. Structure of catechin [2-(3,4-dihydroxyphenyl)chroma-3,5,7-triol] showing 1H and
13
C chemical shifts as elucidated using NMR, 500MHz.
Exact mass: 290.08, Molecular weight: 290.27. m/z: 290.08(100.0%). 3291.08(16.5%),
292.09(1.3%), 292.08(1.2%)

ACKNOWLEDGEMENTS Agriculture Research Institute of the Ghana Atomic

The research team is grateful to WAITRO, LIPI and
ISESCO for funding the research through organization of
the WAITRO research fellowship programme 2008. The
Research Centre for Chemistry of the Indonesian Institute
of Sciences (RCChem/LIPI) is gratefully acknowledged
for hosting the study. The Biotechnology and Nuclear
Energy Commission (BNARI/GAEC) is acknowledged for
its support towards the success of the study.
REFERENCES
Ahmed FR, Ng AS, Fallis AG (1978). 7a-Acetoxydihydronomilin: isola-
tion, spectra, and crystal structure. Can. J. Chem. 56: 1020-1025.
 Apea-Bah et al. 757

Aquino R, De Feo V, De Simone F, Pizza C, Cirino G (1991). Plant Schmidit DD, Frommer W, Junge B, Muller L, Wingender W, Truscheit
metabolites. New compounds and anti-inflammatory activity of E, Schafer D (1977). Alpha-Glucosidase inhibitors. New complex
Uncaria tomentosa. J. Nat. Prod. 54: 453–459. oligosaccharides of microbial origin. Naturwissenschaften; 64: 535-
Cerri R, Aquino R, De Simone F, Pizza C (1988). New quinovic acid 536.
glycosides from Uncaria tomentosa. J. Nat. Prod. 51: 257–261. Standl E, Baumgartl HJ, Fuchtenbusch M, Stemplinger J (1999). Effect
Chang P, Koh YK, Geh SL, Soepadmo E, Goh SH, Wong AK (1989). of acarbose on additional insulin therapy in type 2 diabetes patients
Cardiovascular effects in the rat of dihydrocorynantheine isolated with late failure of sulphonylurea therapy. Diabet. Obes. Metab. 1:
from Uncaria callophylla. J. Ethnopharmacol. 25: 213–215. 215-220.
Corry DB, Tuck ML (2000). Protection from vascular risk in diabetic Statgraphics (2005). Statgraphics Centurion XV windows-based
hypertension. Curr. Hypertens. Rep. 2: 154-159. statistical software.
Das NP, Griffiths LA (1967). Studies on flavonoid metabolism biosyn- Sutedja L (2005). Bioassay of antidiabetes based on -glucosidase inhi-
thesis of (+)-[14C]catechin by the plant Uncaria gambir Roxb. bitory activity. In Functional foods: Trends and challenges. Funct.
Biochem. J. 105: 73-77 Food Prod. Technol. pp.309-315.
a
Desmarchelier C, Mongelli E, Coussio J, Ciccia G (1997). Evaluation of Taniguchi S, Kuroda K, Doi K, Inada K, Yoshikado N, Yoneda Y,
the in vitro antioxidant activity in extracts of Uncaria tomentosa Tanabe M, Shibata T, Yoshida T, Hatano T (2007). Evaluation of
(Willd.) DC. Phytother. Res. 11:254–256. gambir quality based on quantitative analysis of polyphenolic
Goke B, Herrmann-Rinke C (1998). The evolving role of alpha- components. Yakugaku Zasshi. 127(8):1291-1300.
b
glucosidase inhibitors. Diabet. Metab. 14: 1531-1538. Taniguchi S, Kuroda K, Doi K, Tanabe M, Shibata T, Yoshida T,
Hansawasdic C, Kawabata J, Kasai T (2000). -amylase inhibitors from Hatano T (2007). Revised structures of gambiriins A1, A2, B1, and
Rosell (Hibiscus sabdariffa Linn.) Tea. Biosci. Biotechnol. Biochem. B2, chalcane-flavan dimers from Gambir (Uncaria gambir Extract).
64:1041-1043. Chem. Pharm. Bull. 55(2): 268-272.
Hatano T, Kagawa H, Yasuhara T, Okuda T (1988). Two new flavonoids Toeller M. 1994. -Glucosidase inhibitors in diabetes: efficacy in NIDDM
and other constituents in licorice root: their relative astringency and subjects. Eur. J. Clin. Invest. 24: 31-35.
radical scavenging effects. Chem. Pharm. Bull. 36:2090-2097. Wirth C, Wagner H (1997). Pharmacologically active procyanidines from
Heitzman ME, Neto CC, Winiarz E, Vaisberg AJ, Hammond GB (2005). the bark of Uncaria tomentosa. Phytomed. 4: 265–266.
Ethnobotany, phytochemistry and pharmacology of Uncaria Wurm M, Kacani L, Laus G, Keplinger K, Dierich MP (1998). Penta-
(Rubiaceae). Phytochem. 66: 5–29. cyclic oxindole alkaloids from Uncaria tomentosa induce human
Holman RR, Cull CA, Turner RC (1999). A randomized double-blind trial endothelial cells to release a lymphocyte-proliferation regulating
of acarbose in type 2 diabetes shows improved glycaemic control factor. Planta Med. 64: 701–704.
over 3 years (UK prospective Diabetes Study 44). Diabet. Care 22:
960-964.
Kim YM, Jeong YK, Wang MH, Lee WY, Rhee HI (2005). Inhibitory
effect of pine extract on -glucosidase activity and postprandial
hyperglycaemia. Nutrition 21: 756-761.
King H, Aubert RE, Herman WH (1998). Global burden of diabetes,
1995-2025: prevalence, numerical estimates and projections. Diabet.
Care; 21:1414-1431.
Lemaire I, Assinewe V, Cano P, Awang DVC, Arnason JT (1999). Sti-
mulation of interleukinh-1 and -6 production in alveolar macrophages
by the neotropical liana, Uncaria tomentosa (Una de Gato). J.
Ethnopharmacol. 64:109–115.
Montgomery DC (1991). Design and Analysis of Experiments, 3rd
Edition. John Wiley and Sons, New York pp. 134-140.
Packer L, Rimback G, Virgili F. Antioxidant activity and biologic proper-
ties of a procyanidin-rich extract from pine (Pinus maritima) bark,
Pycnogenol. Free Radic. Biol. Med. 27: 704-24.
Phillipson JD, Hemingway SR, Ridsdale CE (1978). Alkaloids of Uncaria.
Part V. Their occurrence and chemotaxonomy. Lloydia 41: 503–570.
Rizzi R, Re F, Bianchi A, De Feo V, De Simone F, Bianchi L, Stivala LA
(1993). Mutagenic and antimutagenic activities of Uncaria tomentosa
and its extracts. J. Ethnopharmacol. 38: 63–77.
Saito N, Sakai H, Sekihara H, Yajima Y (1998). Effect of an -gluco-
sidase inhibitor (voglibose), in combination with sulphonilureas on
glycaemic control in type 2 diabetes patients. J. Int. Med. Res. 26:
219-232.
Sang S, Cheng X, Stark RE, Rosen RT, Yang CS, Ho CT (2002).
Chemical studies on antioxidant mechanism of tea catechins:
analysis of radical reaction products of catechin and epicatechin with
2,2-diphenyl-1-picrylhydrazyl. Bioorg. med. chem. 10:2233-2237.