Selective responses of three Ginkgo biloba leaf-derived constituents on human intestinal bacteria.

Article abstract of DOI:10.1021/jf011140a

The selective responses of Ginkgo biloba leaf-derived materials against six intestinal bacteria was examined using an impregnated paper disk method and compared with that of bilobalide, ginkgolides A and B, kaempferol, and quercetin. The components of G. biloba leaves were characterized as kaempferol 3-O-alpha-(6' -p-coumaroylglucosyl-beta-1,4-rhamnoside), kaempferol 3-O-(2' '-O-beta-D-glucopyranosyl)-alpha-L-rhamnopyranoside, and quercetin 3-O-alpha-(6' -p-coumaroylglucosyl-beta-1,4-rhamnoside) by spectroscopic analysis. The growth responses varied with each bacterial strain tested. At 2 mg/disk, kaempferol 3-O-alpha-(6' -p-coumaroylglucosyl-beta-1,4-rhamnoside) and quercetin 3-O-alpha-(6' -p-coumaroylglucosyl-beta-1,4-rhamnoside) revealed potent inhibition against Clostridium perfringens, and kaempferol 3-O-(2' '-O-beta-D-glucopyranosyl)-alpha-L-rhamnopyranoside showed a clear inhibitory effect on Escherichia coli. At 0.5 mg/disk, quercetin 3-O-alpha-(6' -p-coumaroylglucosyl-beta-1,4-rhamnoside) showed a strong activity against C. perfringens, but weak activity was exhibited by kaempferol 3-O-alpha-(6' -p-coumaroylglucosyl-beta-1,4-rhamnoside) against C. perfringens and kaempferol 3-O-(2' '-O-beta-D-glucopyranosyl)-alpha-L-rhamnopyranoside against E. coli. No inhibition was observed from treatments conducted with bilobalide, ginkgolides A and B, kaempferol, or quercetin. Furthermore, these isolated compounds did not inhibit Bifidobacterium bifidum, B. longum, B. adolescentis, or Lactobacillus acidophilus.

Full text of DOI:10.1021/jf011140a

1840 J. Agric. Food Chem. 2002, 50, 1840−1844
Selective Responses of Three Ginkgo biloba Leaf-Derived
Constituents on Human Intestinal Bacteria
HOI-SEON LEE* AND MIN-JEONG KIM
Research Center for Industrial Development of Biofood Materials and Institute of Agricultural Science
& Technology, College of Agriculture, Chonbuk National University, Chonju 561-756, Korea
The selective responses of Ginkgo biloba leaf-derived materials against six intestinal bacteria was
examined using an impregnated paper disk method and compared with that of bilobalide, ginkgolides
A and B, kaempferol, and quercetin. The components of G. biloba leaves were characterized as
kaempferol 3-O-R-(6′′′-p-coumaroylglucosyl-â-1,4-rhamnoside), kaempferol 3-O-(2′′-O-â-D-gluco-
pyranosyl)-R-L-rhamnopyranoside, and quercetin 3-O-R-(6′′′-p-coumaroylglucosyl-â-1,4-rhamnoside)
by spectroscopic analysis. The growth responses varied with each bacterial strain tested. At 2 mg/
disk, kaempferol 3-O-R-(6′′′-p-coumaroylglucosyl-â-1,4-rhamnoside) and quercetin 3-O-R-(6′′′-p-
coumaroylglucosyl-â-1,4-rhamnoside) revealed potent inhibition against Clostridium perfringens, and
kaempferol 3-O-(2′′-O-â-D-glucopyranosyl)-R-L-rhamnopyranoside showed a clear inhibitory effect on
Escherichia coli. At 0.5 mg/disk, quercetin 3-O-R-(6′′′-p-coumaroylglucosyl-â-1,4-rhamnoside) showed
a strong activity against C. perfringens, but weak activity was exhibited by kaempferol 3-O-R-(6′′′-
p-coumaroylglucosyl-â-1,4-rhamnoside) against C. perfringens and kaempferol 3-O-(2′′-O-â-D-
glucopyranosyl)-R-L-rhamnopyranoside against E. coli. No inhibition was observed from treatments
conducted with bilobalide, ginkgolides A and B, kaempferol, or quercetin. Furthermore, these isolated
compounds did not inhibit Bifidobacterium bifidum, B. longum, B. adolescentis, or Lactobacillus
acidophilus.
KEYWORDS: Ginkgo biloba; intestinal bacteria; kaempferol 3-O-r-(6′′′-p-coumaroylglucosyl-â-1,4-
rhamnoside); kaempferol 3-O-(2′′-O-â-D-glucopyranosyl)-r-L-rhamnopyranoside; quercetin 3-O-r-(6′′′-p-
coumaroylglucosyl-â-1,4-rhamnoside)
INTRODUCTION
may cause a variety of diseases of abnormal physiological states
(5-7).
Various microorganisms are resident in the human intestinal
tract in a highly complex ecosystem with considerable species
diversity (1, 2). They not only participate in the normal
physiological functions, but also contribute significantly to the
genesis of various disease states, by biotransforming a variety
of ingested or endogenously formed compounds to useful or
harmful derivatives (1, 2). Accordingly, these biotransformations
may influence drug efficacy, toxicity, carcinogenesis, and aging
(3, 4). Differences in the intestinal bacteria between patients
and healthy subjects, and between younger and elderly subjects,
have been observed. A normal gastrointestinal microbiota is
found to be predominantly composed of lactic acid bacteria
which seem to play a large role not only in metabolism, but
also host defense against infection, aging, and immunopoten-
tiation (3, 4). In contrast, the microbiota of cancer patients is
composed of a high concentration of clostridia and eubacteria
with few lactic acid bacteria. It has also been reported that
elderly subjects harbor fewer bifidobacteria and more clostridia
than younger subjects. Thus, any disturbance of the microbiota
Much current concern about human health has been focused
on plant-derived bifidus factors which promote the growth of
bifidobacteria or growth inhibitors against harmful bacteria such
as clostridia, eubacteria, and Escherichia coli because plants
constitute a rich source of bioactive chemicals and many of them
are largely free from harmful adverse effects (8, 9). Earlier
findings have already reported and confirmed that among 78
oriental plant species, the methanol extract of Ginkgo biloba
leaves revealed a potent growth-inhibiting activity toward
Clostridium perfringens (10). This plant species is not only
important as an insecticide, but is considered to possess some
medicinal properties, such as an antioxidative agent and an
inhibitor of arachidonic acid (9, 11, 12). However, relatively
little work has been carried out on the effects of G. biloba leaf-
derived materials on growth of intestinal microorganisms
compared to other areas of intestinal microbiology despite its
excellent pharmacological action.
We assessed the selective responses of three G. biloba leaf-
derived constituents on human intestinal bacteria in order to
develop new and safer types of modulators in human intestine.
In addition, the antibacterial activities of six components derived
* Corresponding author. Tel.: +82-63-270-2544. Fax: 82-63-270-2550.
E-mail: hoiseon@moak.chonbuk.ac.kr.
10.1021/jf011140a CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/26/2002

Intestinal Bacteria and Ginkgo biloba Leaves
J. Agric. Food Chem., Vol. 50, No. 7, 2002 1841
from G. biloba are also discussed in relation to the results
obtained.
Table 1. Growth-Inhibiting Effects on Human Intestinal Bacteria by
Various Fractions Obtained from Me2CO Extracts of Ginkgo biloba
Leaves
MATERIALS AND METHODS
bacterial strainb
Chemicals. Kaempferol and quercetin were obtained from Sigma
Chemical (St. Louis, MO). Bilobalide and ginkgolides A and B isolated
from G. biloba were kindly provided by Prof. Moo-Key Kim from the
Faculty of Biotechnology, College of Agriculture, Chonbuk National
University, Chonju, South Korea. All other chemicals were of reagent
grade.
materiala
B. longum B. bifidum B. adolescentis C. perfringens E. coli
Me2CO extract
hexane fraction
EtOAc fraction
BuOH fraction
++c
-
-
-
-
-
-
++
-
++++
-
++++
-
+
-
+
-
-
-
++++
-
++++
+
++++
-
H O fraction
2
Bacterial Strains and Culture Conditions. The bacterial strains
used in this study were Bifidobacterium bifidum ATCC 29521, B.
longum ATCC 15707, B. adolescentis ATCC 15073, Clostridium
perfringens ATCC 13124, Escherichia coli ATCC 11775, and Lacto-
bacillus acidophilus KCTC 3145 isolated from human feces. Stock
cultures of these strains were routinely stored on Eggerth-Gagnon liver
extract-Fieldes slant at -80 °C, and, when required, were subcultured
on Eggerth-Gagnon (EG) agar (Eiken Chemical Co., Ltd, Tokyo, Japan)
(13). The plates were incubated anaerobically at 37 °C for 2 days in
an atmosphere of 80% N2, 15% CO2, and 5% H2 in an anaerobic
chamber (Coy Lab., Grass Lake, MI). The bacteria were then grown
in EG broth (pH 6.8).
Isolation and Identification. Leaves of G. biloba (10 kg) were dried
in an oven at 60 °C for 2 days, finely powdered, extracted twice with
60% aq Me2CO at room temperature, and filtered (Toyo filter paper
No. 2, Japan). The extract was concentrated via rotary evaporation at
35 °C to yield about 13.4% dry weight of the leaves (14). The extract
(20 g) was sequentially partitioned into hexane, EtOAc, BuOH, and
H2O portions for subsequent bioassay with C. perfringens and B.
fragilis. The organic solvent portions were concentrated to dryness by
rotary evaporation at 35 °C, and the H2O portion was freeze-dried.
For isolation, 5 mg of each G. biloba leaf-derived fraction in methanol
was applied to paper disks (Advantec 8 mm diameter and 1 mm thick,
Toyo Roshi, Japan).
Kaempferol 3-O-R-(6′′′-p-Coumaroylglucosyl-â-1,4-rhamnoside). The
EtOAc extract (51 g) was chromatographed over Amberlite IRN-78
column (Prolabo, 100 g, USA) using a stepwise gradient of H2O/MeOH
(0, 10, 20, 40, and 60%) and then MeOH/0.05 N HCl. The active 40%
fractions (22.5 g) were monitored by TLC. The active 40% fraction
was chromatographed over a silica gel 60 column (Merck, 1.0 kg, USA)
packed with EtOAc and eluted with a stepwise gradient of EtOAc/
MeOH (0, 10, 20, 30, and 40%). The active fraction (3.2 g) was
chromatographed over a Polyclar AT column (Touzart and Matignon,
100 g, USA) packed with CHCl3/MeOH (5:1) and eluted with the same
solvent. The compound was finally purified successively on a Sephadex
LH-20 column (Pharamacia, USA) and cellulose (Merck) eluted with
MeOH. Pure compound (475 mg) was obtained as an amorphous solid.
Kaempferol 3-O-(2′′-O-â-D-Glucopyranosyl)-R-L-rhamnopyranoside.
The concentrated BuOH extract (48 g) was chromatographed over a
Sephadex LH-20 column (Pharmacia, 800 × 49 mm, USA) and a
Sepralyte RP-18 column (Analytichem, prep. grade, 40 µm, 713 × 18.5
mm, USA) with different MeOH/H2O gradients as mobile phases.
HPLC column, Hypersil ODS, 3 µm (Knauer, USA): mobile phase,
MeOH (A) and 0.5% ortho-H3PO4 in H2O (B). A linear gradient was
run from 38 to 48.2% MeOH in 12 min. Pure compound (113 mg)
was obtained. Acetylation of the compound was carried out by stirring
500 mg of the substance with 500 µL of Ac2O in 500 µL of pyridine
overnight at room temperature to afford a nonacetate of the compound.
Complete hydrolysis of the compound was achieved in 2 M trifluoro-
acetic acid by refluxing for 60 min. Identification of the aglycons and
the sugars was completed via TLC on silica gel 60 F254 (Merck, USA).
Diphenylboric acid 2-amino-ethyl ester (1% in MeOH) was used for
the detection of the aglycons.
a Exposed to 5 mg/disk. b Cultured on Eggerth-Gagnon agar at 37 °C for 2
days in an atmosphere of 80% N , 15% CO , and 5% H . c Inhibitory zone diameter
2
2
2
>30 mm, ++++; 21−30 mm, +++; 16−20 mm, ++; 10−15 mm, +; and <10 mm, -.
MeOH (0, 10, 20, 30, 40, and 50%). The active fraction (4.5 g) was
chromatographed over a Sephadex LH-20 column (Pharmacia, 800 ×
49 mm, USA) using a stepwise gradient of H2O/MeOH (20, 25, 30,
35, and 40%). This operation was repeated five times. The active
fraction (250 mg) was chromatographed over a Polyclar AT column
(100 g) packed with CHCl3/MeOH (5:1) and eluted with an increasing
ratio of MeOH (25, 30, 35, 40, 45, and 50%). The compound was finally
purified on a Sephadex column (915 g) eluted with MeOH. Pure
compound (200 mg) was obtained.
Structural determination of the active isolate was based on the
1
spectroscopic analysis. H- and 13C NMR spectra were recorded with
a Bruker AM spectrometer, and chemical shifts are given in ppm. UV
spectra were obtained on a Waters 490 spectrometer. IR spectra were
obtained on a Biorad FT-80 spectrophotometer, and mass spectra were
obtained on a JEOL JMS-DX 30 spectrometer. The method of Park et
al. was used to visualize the isolated compounds (12).
Microbiological Assay. For assaying the inhibitory effect of each
test sample on the microorganisms, one loopful of bacteria was
suspended in 1 mL of sterile physiological saline. An aliquot (0.1 mL)
of the bacterial suspensions was seeded on EG agar. Samples of the
extract dissolved in methanol were applied to paper disks by using a
Drummond glass microcapillary (Advantec, 8 mm, Toyo Roshi, Japan).
After evaporation, the disks were placed on EG agar surface and
incubated at 37 °C for 2 days in an anaerobic chamber. Control disks
were treated with methanol only. All inhibition tests were conducted
in triplicate. The growth responses of test samples were determined
by comparison with those of the control. The inhibitory responses were
classified as previously described: inhibitory zone diameter >30 mm,
++++; 21-30 mm, +++; 16-20 mm, ++; 10-15 mm, +; and
<10 mm, - (15, 16)
RESULTS
Growth-inhibiting activities of human intestinal bacteria to
the fractions obtained from Me2CO extracts of G. biloba leaves
were assayed by using the impregnated paper disk method
(Table 1). The Me2CO extracts showed a very strong activity
(++++) against C. perfringens and E. coli, and moderate
activity (++) against B. longum and B. adolescentis. However,
in tests conducted with B. bifidum, which is predominantly in
the intestines of infants, the Me2CO extracts showed no
inhibitory response. EtOAc and H2O fractions obtained from
Me2CO extracts showed a strong activity against C. perfringens,
and the BuOH fraction exhibited a strong activity against E.
coli. However, the hexane fraction had no inhibitory response
against any intestinal bacteria tested. Purification of the biologi-
cally active compounds from the fractions were chromato-
graphed via repeated silica gel, Amberlite IRN-78, Polyclar AT,
Sephadex LH-20, and Sepralyte RP-18 column chromatography,
and the isolates were bioassayed. Three active isolates showed
a strong and moderate activity. Structural determinatios of the
isolates were made by various spectroscopic analyes, and the
compounds were characterized as kaempferol 3-O-R-(6′′′-p-
Quercetin 3-O-R-(6′′′-p-Coumaroylglucosyl-â-1,4-rhamnoside). The
H2O fraction (189 g) was defatted with C6H6 and chromatographed
over an Amberlite IRN-78 column (100 g) using a stepwise gradient
of H2O/MeOH (0, 10, 20, 30, 40, 50, and 60%) and then MeOH/0.05
M HCl. Active fractions (42 g) were monitored by TLC. The active
30% fraction was chromatographed over a silica gel 60 column (1.0
kg) packed with CHCl3 and eluted with a stepwise gradient of CHCl3/

1842 J. Agric. Food Chem., Vol. 50, No. 7, 2002
Lee and Kim
Figure 1. Structures of three compounds isolated from Ginkgo biloba leaves.
coumaroylglucosyl-â-1,4- rhamnoside), K1, kaempferol 3-O-
(2′′-O-â-D-glucopyranosyl)-R-L-rhamnopyranoside, K2, and quer-
cetin 3-O-R-(6′′′-p-coumaroylglucosyl-â-1,4-rhamnoside), Q3
(Figure 1). The compounds were identified on the basis of the
following evidence.
MHz, DMSO-d6) spectra similar to those published for for
kaempferol 3-O-R-(6′′′-p-coumaroylglucosyl-â-1,4-rhamnoside)
isolated from G. biloba (12).
K2. Mp 308-311 °C. FABMS: m/z 633 [M + K]+, 617 [M
+ Na]+, 595 [M + H]+, 287 [aglycon + H]+. UV λMmaexOH nm:
267, 297sh, 346; NaOMe 273, 324, 388; AlCl3 275, 300, 344,
396; AlCl3-HCl 272, 299, 340, 392; NaOAc 272, 302, 375;
K1. Mp 335-339 °C; [R]2D0 -60° (EtOH; c 1). UV λEtOH
max
nm: 355sh, 310, 263; +NaOAc: 370sh, 304, 267; +NaOAc-
H3BO3: 355sh, 310, 263; +AlCl3: 396sh, 307sh, 302, 275, 225;
1
NaOAc-H3BO3 264, 316sh, 344. H NMR (300 MHz, CD3-
1
+AlCl3-HCl: 397sh, 307sh, 303, 275, 225. H NMR (200
OD) δ: 0.94 (3H, d, J ) 5.9 Hz, H-6′′), 3.19-3.43 (6H, m,
sugar protons), 3.70 (2H, m, H-6′′′), 3.81 (1H, dd, J3′′,2′′ ) 3.5
Hz, J3′′,4′′ ) 9.3 Hz, H-3′′), 4.29 (1H, dd, J2′′,1′′ ) 1.1 Hz, J2′′,3′′
) 3.5 Hz, H-2′′), 4.42 (1H, d, J ) 7.7 Hz, H-1′′′), 5.73 (1H, d,
J ) 1.1 Hz, H-1′′), 6.20 (1H, d, J ) 1.9 Hz, H-6), 6.37 (1H, d,
J ) 1.9 Hz, H-8), 6.94 (2H, d, J ) 8.9 Hz, H-3′ and H-5′),
MHz, DMSO-d6): δ 0.92 (3H, d, J ) 6.0 Hz, Me rhamnose),
3.06-4.24 (m, sugars protons), 4.35 (1H, d, J ) 8 Hz, H-1
GLC), 5.61 (1H, d, J ) 2.0 Hz, H-1 rha), 6.11 (1H, d, J ) 2.5
Hz, H-6), 6.10 (1H, d, J ) 16.0 Hz, H-8 coum), 6.25 (1H, d, J
) 2.5 Hz, H-8), 6.72 (2H, d, J ) 9.0 Hz, H-3 coum and H-5
coum), 6.91 (2H, d, J ) 9.0 Hz, H-3′ and H-6′), 7.37 (2H, d, J
) 9.0 Hz, H-2 coum and H-6 coum), 7.45 (1H, d, J ) 16.0 Hz,
H-7 coum), 7.72 (2H, d, J ) 9.0 Hz, H-2′ and H-6′), 8.49 (s,
OH phenolic). CIMS 70 eV, m/z (rel int.): 741 [M + H]+ (2),
595 (10), 472 (14), 433 (100), 287 (1). K1 gave 13C NMR (50
1
7.76 (2H, d, J ) 8.9 Hz, H-2′ and H-6′). H NMR (300 MHz,
CDCl3) of kaempferol 3-O-(2′′-O-â-D-glucopyranosyl)-R-L-
rhamnopyranoside-nonaacetate (kaempferol 3-O-(2′′-O-â-D-glu-
copyranosyl)-R-L-rhamnopyranoside after acetylation) δ: 0.87
(3H, d, J ) 6.2, H-6′′), 1.96-2.10 (18H, aliphatic OAc units),

Intestinal Bacteria and Ginkgo biloba Leaves
J. Agric. Food Chem., Vol. 50, No. 7, 2002 1843
centrations (Table 2). At 2 mg/disk, K1 and Q3 produced a
very clear inhibitory effect on C. perfringens (++++), and
K2 showed a clear inhibitory effect on E. coli (+++).
Furthermore, any inhibition toward B. longum and L. acidophilus
was not exhibited by these three compounds. At 1 mg/disk,
growth of C. perfringens was significantly inhibited by Q3,
whereas moderate inhibitory activity was obtained by K1.
Furthermore, growth of E. coli was moderately inhibited by K2.
At 0.5 mg/disk, Q3 showed a strong growth-inhibiting activity
(+++) against C. perfringens, but weak activity was exhibited
by K1 against C. perfringens and K2 against E. coli.
Table 2. Growth-Inhibiting Effects of Ginkgo biloba Leaf-Derived
Compounds against Intestinal Bacteria
bacterial straina
dose
compound
K1
(mg/disk) B. bifidum B. longum C. perfringens E. coli
b
5.0
2.0
1.0
0.5
5.0
2.0
1.0
0.5
5.0
2.0
1.0
0.5
5.0
5.0
5.0
5.0
5.0
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
++++
-
-
-
-
++++
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
++
+
-
-
-
-
++++
+++
+++
+++
-
-
-
-
-
K2
Q3
++++
+++
++
+
-
-
-
-
-
-
-
-
-
DISCUSSION
The intestinal microbiota in healthy people remains relatively
constant but is known to be significantly influenced by physical,
biological, chemical, environmental, or host factors (3, 17).
Accordingly, alterations to the microbiota may cause abnormal
physical conditions or diseases. In our study, the growth-
inhibitory activity of G. biloba (Ginkgoaceae) leaf-derived
materials against five intestinal bacteria in vitro was investigated,
and kaempferol 3-O-R-(6′′′-p-coumaroylglucosyl-â-1,4-rham-
noside), kaempferol 3-O-(2′′-O-â-D-glucopyranosyl)-R-L-rham-
nopyranoside, and quercertin 3-O-R-(6′′′-p-coumaroylglucosyl-
â-1,4-rhamnoside) were identified by comparison of their
physical and spectroscopic data with those reported in the
literature (12, 14). The 1H and 13C NMR spectra of K2 showed
signals consistent with the presence of glucose, rhamnose, and
kaempferol. Triplets at δ 62.4 in the 13
bilobalide
ginkgolide A
ginkgolide B
kaempferol
quercetin
a Cultured on Eggerth-Gagnon agar at 37 °C for 2 days in an atmosphere of
80% N , 15% CO , and 5% H . b Inhibitory zone diameter >30 mm, ++++; 21−30
2
2
2
mm, +++; 16−20 mm, ++; 10−15 mm, +; and <10 mm, -.
2.32-2.40 (9H, aromatic OAc units) 3.29 (1H, m, H-5′′), 3.70
(1H, m, H-5′′′), 4.05 (1H, ABX, J6′′′B, 5′′′ ) 2.4 Hz, J6′′′A, 6′′′B
)
12.3 Hz, H-6′′′ B), 4.33 (1H, ABX, J6′′′A, 5′′′ ) 3.8 Hz, J6′′′A, 6′′′B
) 12.3 Hz, H-6′′′ A), 4.41 (1H, m, H-2′′), 4.58 (1H, d, J ) 8.0
Hz, H-1′′′), 4.85 (1H, t, J ) 10.0 Hz, H-4′′), 5.06-5.25 (4H,
m, H-3′′, H-2′′′, H-3′′′, H-4′′′), 5.60 (1H, d, J ) 1.9 Hz, H-1′′),
6.20 (1H, d, J ) 2.2 Hz, H-6), 6.26 (1H, d, J ) 2.2 Hz, H-8),
7.27 (2H, d, J ) 8.8 Hz, H-3′ and H-5′), 7.89 (2H, d, J ) 8.8
Hz, H-2′ and H-6′). The 13C NMR (75 MHz, CD3OD) spectra
of K2 were found to be the same as those for kaempferol 3-O-
(2′′-O-â-D-glucopyranosyl)-R-L-rhamnopyranoside isolated from
C NMR spectra pointed
to a free C(6)H2OH of glucose, and the doublets of the anomeric
1
protons in the H NMR at δ 4.42 (J ) 7.7 Hz) indicated the
terminal position of glucose. Both H-1 signals of the rhamnose
units were doublets at δ 5.73 (J ) 1.1 Hz) indicating, by its
downfield shifted location, the linkage between rhamnose and
C-3 of the aglycon. The â-configuration of glucose (diaxial
coupling) and the R-configuration of rhamnose (diequatorial
coupling) were evidenced by their respective coupling constants
3J1,2. Acetylation of K2 afforded the nonaacetate showing the
signal of H-2′′ at nearly unchanged position, thus confirming
the proposed 2′′f1′′ linkage. The growth-inhibitory effects of
these compounds varied with bacterial strain tested. In this
family, a great number of plant extracts have been investigated
for the biological properties (12, 18, 19). Furthermore, materials
derived from G. biloba have been extensively studied for
pharmacological and pesticidal effects (12, 14, 18).
G. biloba (12).
Q3. Mp 229-233 °C. UV λ
EtOH
max
nm: 360sh, 316, 300sh,
268, 258; + NaOAc: 370sh, 315, 300sh, 269; + NaOAc-H3-
BO3: 373sh, 315, 300sh, 263; + AlCl3: 410sh, 360sh, 315,
300sh, 272; + AlC3-HCl: 400sh, 360sh, 315, 300sh; 277. 1H
NMR (200 MHz, DMSO-d6): δ 0.91 (3H, d, J ) 6.0 Hz, Me
rhamnose), 3.03-4.15 (m, sugars protons), 4.28 (1H, d, J )
8.0 Hz, H-1 GLC), 5.52 (1H, d, J ) 2.0 Hz, H-1 rha), 6.16
(1H, d, J ) 2.0 Hz, H-6), 6.24 (1H, d, J ) 16.0 Hz, H-8 coum),
6.31 (1H, d, J ) 2.0 Hz, H-8), 6.70 (2H, d, J ) 8.6 Hz, H-3
coum and H-5 coum), 6.88 (1H, d, J ) 8.4 Hz, H-5′), 7.25
(1H, dd, J ) 2.0 Hz and 8.4 Hz, H-6′), 7.36 (1H, d, J ) 2.0
Hz, H-2′), 7.41 (2H, d, J ) 8.6 Hz, H-2 coum and H-6 coum),
7.45 (1H, d, J ) 16.0 Hz, H-7 coum). CIMS 70 eV m/z (rel
int.) 757 [M + H]+ (0.3), 611 (1), 595 (0.5), 472 (6.6), 449-
(66.7), 303(100). The 13C NMR spectra of Q3 were found to
be the same as those for quercetin 3-O-R-(6′′′-p-coumaroylglu-
cosyl-â-1,4-rhamnoside) isolated from G. biloba (14).
The growth-inhibiting activities of the isolated compounds
and other components of this plant species toward intestinal
bacteria when treated with 5 mg/disk were determined (Table
2). Responses varied with the chemical and bacterial strain
tested. K1 and Q3 showed a very strong growth-inhibiting
activity (++++) against C. perfringens, but no inhibitory
response against other bacteria tested. K2 exhibited a strong
growth-inhibiting activity against E. coli, but no inhibitory
response against other bacteria. No activity was observed for
bilobalide, ginkgolides A and B, kaempferol, or quercetin
derived from G. biloba.
Infectious diseases caused by clostridia have a broad spectrum
of clinical severity that ranges from mild outpatient illness to
sudden death. Among the clostridia, C. perfringens has been
associated with sudden death, toxicity, and gastrointestinal
disease in man (20, 21). In contrast, bifidobacteria are often
taken as useful indicators of human health under most envi-
ronmental conditions, because they play important roles in
metabolism such as amino acid (22) and vitamin production
(17), aid in the defense against infections (3), association with
longevity (23), antitumor activities (24), pathogen inhibition (25,
26), and immunopotentiation (27, 28). Accordingly, it would
be desirable to both inhibit the growth of potential pathogens
such as clostridia and/or increase the numbers of bifidobacteria
in the human gut. Selective growth promoters for bifidobacteria
or inhibitors for harmful bacteria are especially important for
human health, because intake of these materials may normalize
disturbed physiological functions, resulting in the prevention
and treatment of various diseases caused by pathogens in the
gastrointestinal tract. In recent years, much concern has been
focused on selective plant-derived growth modulators in the
Because of their potent growth-inhibiting activity toward test
bacteria, the isolated compounds were evaluated at low con-

1844 J. Agric. Food Chem., Vol. 50, No. 7, 2002
Lee and Kim
(13) Mitsuoka, T.; Sega, T.; Yamamoto, S. Eine verbesserte Methodik
der qualitativen und quantitativen Analyse der Darmflora von
Menschen und Tieren. Zentralblatt fu¨r Bakteriologie, Parasiten-
kunde, Infektionskrankheiten und Hygiene, I. Originale 1965,
195, 455-469.
(14) Hasler, A.; Gross, G. A.; Meier, B.; Sticher, O. Complex flavonol
glycosides from the leaves of Ginkgo biloba. Phytochemistry
1992, 31, 1391-1394.
(15) Kim, M. K.; Kim, M. J.; Shin, D. H.; Song, C. G.; Lee, H. S.
Growth-inhibiting effects of vegetable extracts on beneficial and
harmful human intestinal bacteria. Agric. Chem. Biotechnol.
2001, 44, 65-70.
(16) Cho, S. H.; Jeon, H. O.; Han, Y. K.; Yeon, S. H.; Ahn, Y. J. In
vitro growth-inhibiting effects of leaf extracts from Pinus species
on human intestinal bacteria. Agric. Chem. Biotechnol. 1999,
42, 202-204.
intestine, based on the fact that many medicinal plant-derived
materials are relatively nontoxic to humans. For example,
extracts from ginseng (Panax ginseng) and green tea (Thea
chinensis L.) have been shown to not only enhance the growth
of bifidobacteria, but selectively inhibit various clostridia (29,
30). In our study, the growth responses of G. biloba leaf-derived
compounds varied according to bacterial strain tested. Growth-
inhibiting activity was more pronounced in C. perfringens and
E. coli, as compared to that of the bifidobacteria. In the test
using 1 mg/disk, Q3 and K1 showed very strong and moderate
activity, respectively, against C. perfringens, and K2 exhibited
moderate activity against E. coli. These three compounds did
not adversely affect the growth of bifidobacteria and lactobacilli
used. These results suggest that the strong and moderate
inhibitory activity of the three isolates confirms their superiority
and usefulness as bacteriocidal agents.
(17) Rasic, J. L.; Kurmann, J. A. Bifidobacteria and Their Role;
Birkhauser Verlag: Boston, MA, 1983.
In conclusion, our results indicate that three components
isolated from G. biloba leaves have growth-inhibiting effects
in vitro against specific bacteria from the human intestine. On
the basis of our limited data and some earlier findings, the
inhibitory action of these compounds against C. perfringens and
E. coli may be an indication of at least one of the pharmacologi-
cal actions of G. biloba leaves. Further work is necessary to
establish whether this activity is exerted in the human large
intestine after consumption of G. biloba leaves by humans.
(18) Kim, M. K.; Lee, S. E.; Lee, H. S. Growth-inhibiting effects of
Brazilian and oriental medicinal plants on human intestinal
bacteria. Agric. Chem. Biotechnol. 2000, 43, 54-58.
(19) Kim, H. K.; Kim, D. G.; Do, J. R.; Lee, Y. C.; Lee, B. Y.
Antioxidative activity and physiological activity of some Korean
medicinal plants. Korean. J. Food. Sci. Technol. 1995, 27, 80-
85.
(20) Bokkenheuser, V. D.; Winter, J. Biotransformation of steroids.
In Human Intestinal Microflora in Health and Disease; Hentges,
D. J., Ed.; Academic Press: New York, 1983; pp 215-239.
(21) Goldman, P. Biochemical pharmacology and toxicology involv-
ing the intestinal flora. In Human Intestinal Microflora in Health
and Disease; Hentges, D. J., Ed.; Academic Press: New York,
1983; pp 241-263.
(22) Matteuzzi, D.; Crociani, F.; Emaldi, O. Amino acids produced
by bifidibacteria and some clostridia. Ann. Microbiol. (Paris)
1978, 129B, 175-181.
(23) Mitsuoka, T.; Hayakawa, K. Die Faekalflora bei Menschen. I.
Mitteilung: Die Zusammensetzung der Faekalflora der ver-
schiedenen altersgruppen. Zentralbl. Bakteriol., Mikrobiol. Hyg.
Abt. I 1973, A223, 333-342.
(24) Tsuyuki, S.; Yamazaki, S.; Akashiba, H.; Kamimura, H.; Sekine,
K.; Toida, T.; Saito, M.; Kawashima, T.; Ueda, K. Tumor-
suppressive effect of a cell wall preparation, WPG, from
Bifidobacterium infantis in germfree and flora-bearing mice.
Bifidobact. Microflora 1991, 10, 43-52.
(25) Bullen, C. L.; Willis, A. T. Resistance of the breast-fed infant
to gastroenteritis. Br. Med. J. 1971, 3, 338-343.
(26) Rasic, J. L. The role of dairy foods containing bifido- and
acidophilus-bacteria in nutrition and health. N. Eur. Dairy J.
1983, 48, 80-88.
(27) Perdigon, G.; Alvarez, S.; Rachid, M.; Aguero, G.; Gobbato, N.
Immune system stimulation by probiotics. J. Dairy Sci. 1995,
78, 1597-1606.
LITERATURE CITED
(1) Hughes, D. B.; Hoover, D. G. Bifidobacteria: their potential
for use in American dairy products. Food Technol. 1991, 45,
74-83.
(2) Modler, H. W.; McKellar, R. C.; Yaguchi, M. Bifidobacteria
and bifidogenic factors. Can. Inst. Food Sci. Technol. J. 1990,
23, 29-41.
(3) Hentges, D. J. Role of the intestinal microflora in host defense
against infection. In Human Intestinal Microflora in Health and
Disease; Hentges, D. J., Ed; Academic: New York, 1983; pp
311-331.
(4) Mitsuoka, T. A Color Atlas of Anaerobic Bacteria; Shobansha:
Tokyo, Japan, 1984.
(5) Finegold, S. M.; Flora, D. J.; Attebery, H. R.; Sutter, V. L. Fecal
bacteriology of colonic polyp patients and control patients.
Cancer Res. 1975, 35, 3407-3417.
(6) Gorbach, S. L.; Nahas, L.; Lerner, P. I.; Weinstein, L. Studies
of intestinal microflora. I. Effects of diet, age, and periodic
sampling on numbers of fecal microorganisms in man. Gastro-
enterology 1967, 53, 845-855.
(7) Mastromarino, A.; Reddy, B. S.; Wynder, E. L. Fecal profiles
of anaerobic microflora of large bowel cancer patients and
patients with nonhereditary large bowel polyps. Cancer Res.
1978, 38, 4485-4462.
(8) Harborne, J. B. Introduction to Ecological Biochemistry, 4th ed.;
Academic Press: London, 1993.
(9) Namba, T. Colored Illustrations of Wakan-Yaku (The Crude
Drugs in Japan, China and the Neighbouring Countries);
Hoikusha Publishing: Osaka, Japan, 1986.
(10) Ahn, Y. J.; Kwon, J. H.; Chae, S. H.; Park, J. H.; Yoo, J. Y.
Growth-inhibitory responses of human intestinal bacteria to
extracts of oriental medicinal plants. Microb. Ecol. Health Dis.
1994, 7, 257-261.
(11) Kang, S. M. Effect of inhibitions of Ginkgo biloba extracts on
induction of reactive oxygen species and release of inflammation
medicator arachidonic acid from U937. Korean J. Food Sci.
Technol. 2000, 32, 1198-1205.
(12) Park, B. S.; Lee, S. E.; Kim, M. K.; Kim, Y. S.; Lee, H. S.
Antioxidative activity of Ginkgo biloba leaves-derived compo-
nents on free radicals and active oxygen species. Food Sci.
Biotechnol. 2000, 9, 317-321.
(28) Pereyra, B. S.; Lemonnier, D. Induction of human cytokines by
bacteria used in dairy foods. Nutr. Res. 1993, 13, 1127-1140.
(29) Ahn, Y. J.; Kim, M.; Kawamura, T.; Yamamoto, T.; Fujisawa,
T.; Mitsuoka, T. Effects of Panax ginseng extract on growth
responses of human intestinal bacteria and bacterial metabolism.
Korean J. Ginseng Sci. 1990, 4, 253-264.
(30) Ahn, Y. J.; Kawamura, T.; Kim, M.; Yamamoto, T.; Mitsuoka,
T. Tea polyphenols: selective growth inhibitors of Clostridium
spp. Agric. Biol. Chem. 1991, 55, 1425-1426.
Received for review August 22, 2001. Revised manuscript received
December 3, 2001. Accepted December 4, 2001. This research was
supported by the Research Center for Industrial Development of
Biofood Materials in Chonbuk National University, Chonju, Korea. The
Research Center for Industrial Development of Biofood Materials is
designated as a Regional Research Center appointed by the Korea
Science and Engineering Foundation (KOSEF), Chollabukdo Provincial
Government and Chonbuk National University.
JF011140A