Available online at www.sciencedirect.

com

Chemico-Biological Interactions 173 (2008) 32–42

Protective effects of Ginkgo biloba against rat liver carcinogenesis

Marcos C. Dias a , Maria A.M. Rodrigues b , Maria C.H. Reimberg c , Luı́s F. Barbisan a,∗
a UNESP São Paulo State University, Institute of Biosciences, Department of Morphology, Botucatu, SP 18618-000, Brazil
b UNESP São Paulo State University, Faculty of Medicine, Department of Pathology, Botucatu, SP 18618-000, Brazil
c CentroFlora Group, Botucatu, SP 18603-970, Brazil

Received 7 November 2007; received in revised form 11 January 2008; accepted 21 January 2008
Available online 12 February 2008

Abstract
Ginkgo biloba (EGb) has been proposed as a promising candidate for cancer chemoprevention and has shown protective effects
on the liver against chemically induced oxidative injury and fibrosis. The potential beneficial effects of EGb were investigated in
two rat liver carcinogenesis bioassays induced by diethylnitrosamine (DEN). In a short-term study for anti-initiating screening,
male Wistar rats were fed a basal diet or supplemented diet with 500 or 1000 ppm EGb and initiated 14 days later with a single dose
of DEN (100 mg/kg i.p.). The respective groups were killed 24 h or 2 weeks after DEN-initiation. Liver samples were collected for
the analysis of proliferating cell nuclear antigen (PCNA), transforming growth factor alpha (TGF-␣), p53, apoptosis and induction
of single hepatocytes and minifoci positive for the enzyme glutathione S-transferase P-form (GST-P). In a medium-term study for
anti-promoting screening, the animals received a single dose of DEN (200 mg/kg i.p.) and, 2 weeks later, were fed a basal diet or
supplemented diet with 500 or 1000 ppm EGb for 6 weeks. All animals underwent 70% partial hepatectomy (PH) at week 3 and
killed at week 8. Liver samples were colleted to analyze development of preneoplastic foci of altered hepatocytes (FAH) expressing
GST-P. In the short-term study, pretreatment of rats with 1000 ppm EGb significantly reduced the rates of cell proliferation, apoptosis
and p53, TGF-␣ immunoreactivity and the number of GST-P-positive hepatocytes. In the medium-term study, EGb treatment during
the post-initiation stage failed to reduce the development of DEN-induced GST-P-positive foci. Thus, EGb presented inhibitory
actions during initiation but not promotion of rat liver carcinogenesis induced by DEN.
© 2008 Elsevier Ireland Ltd. All rights reserved.

Keywords: Diethylnitrosamine; Putative initiated hepatocytes; GST-P-positive preneoplastic foci; Liver carcinogenesis; Ginkgo biloba; Chemopre-
vention

1. Introduction that has presented various pharmacological activities

Ginkgo biloba extract (EGb) is a commercial medic-
inal herb which comes from the green leaves of the
ginkgo tree, one of the oldest living plant species [1].
EGb is a mixture, composed mainly of flavone gly-
cosides and terpenoids (ginkgolides and bilobalide),
∗ Corresponding author at: Departamento de Morfologia, Instituto
de Biociências, Universidade Estadual Paulista (UNESP), Botucatu,
18618-000 SP, Brazil. Tel.: +55 14 68116264; fax: +55 14 68116264.
E-mail address: barbisan@ibb.unesp.br (L.F. Barbisan).
[2–4]. This extract has shown several in vivo effects,
including augmentation of blood flow and inhibi-
tion of platelet activating factor. It protects the cell
membrane against damage induced by free radicals
and presents protective effects against myocardial and
brain ischemia/reperfusion injury [2–5]. EGb has been
widely used in the clinical treatment of cardiovascular
and neurological diseases like Alzheimer’s, dementia,
labyrintopathies and other cognitive dysfunctions [2,6].
EGb has shown protective effects on the liver against
chemically induced oxidative injury and fibrosis [7–10].

0009-2797/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.cbi.2008.01.012
 M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42
Also, EGb has been shown to modulate the hepatic
phase I enzymes, including the induction or inhibi-
tion of specific cytochrome P450 (CYP) isozymes, and
phase II enzymes, specifically the induction of glutathion
S-transferases, DT-diaphorase and quinine reductase
[11–14]. Thus, changes in metabolizing phase I and
phase II enzymes by EGb can alter the balance between
the activation of procarcinogens or detoxification of
potential carcinogens.
There are few in vivo and in vitro studies on the
potential anti-carcinogenic effects of EGb. Crude EGb
or their specific compounds induced cellular death or
anti-proliferative activities against the following cancer
cell lines: HepG2, Hep 3B, and SMMC-7721 human
hepatocellular [15,16], SCC 1483 human oral [17] and
OVCA429, 433 and 420 human ovarian [18]. EGb, when
fed orally to female Swiss mice, reduced forestomach
tumor multiplicity induced by benzo(a)pyrene [19]. EGb
and bilobalide, given orally to male Fischer 344 rats,
inhibited the development of colonic aberrant crypt foci
induced by azoxymethane [20].
Foci of altered hepatocytes (FAH) have been
described as putative preneoplastic lesions detected in
various rodent models of chemical liver carcinogenesis
[21]. More recently, different types of FAH with similar
morphological and biochemical alterations in the hepato-
cellular phenotype were identified in chronic human liver
diseases associated with, or predisposing to hepatocel-
lular carcinoma [22]. Glutathione S-transferase P-form
(GST-P) expression is a useful marker for preneoplastic
and neoplastic rat liver lesions [23]. Single hepatocytes
and minifoci highly positive for GST-P develop very
early in carcinogen-treated rat liver, and are considered
precursors of large FAH and nodules [24,25]. The detec-
tion of GST-P-positive single cells and minifoci or large
GST-P-positive FAH is an important tool for analyz-
ing relevant carcinogenic or anti-carcinogenic responses
during the initiation and promotion stages of rat liver
carcinogenesis [26–28].
Using a short-term (anti-initiating screening) and a
medium-term (anti-promoting screening) liver bioassay,
the present study investigated the modifying influence
of a G. biloba extract (EGb) on the initiation and pro-
moting phases of rat liver carcinogenesis induced by
diethylnitrosamine (DEN).
2. Material and methods
2.1. Animals and treatments
Four-week-old male Wistar rats were obtained from
CEMIB (UNICAMP Campinas, SP, Brazil). The animals
33
were kept in polypropylene cages (five animals/cage)
covered with metallic grids in a room maintained at
22 ± 2 ◦ C, 55 ± 10% humidity under a 12-h light:12-h
dark cycle. They were fed commercial NUVILAB-CR-
1 chow (NUVITAL, Curitiba, PR, Brazil) and water ad
libitum for a 2-week acclimation period before begin-
ning the experiment. Samples of lyophilized extract of
G. biloba leaves (EGb), generously supplied by the Cen-
troFlora Group (Botucatu, SP, Brazil), were obtained
from hydroalcoholic extraction by a spray dryer system.
The EGb (cod. 500821) used in this study contained
known amounts of approximately 24% flavone glyco-
sides (i.e., quercetin, kaempferol and isorhamnetin) and
6% terpene trilactones (i.e., ginkgolide A–C, bilobalide)
as determined by the HPLC method. EGb was supple-
mented to basal diet at 500 and 1000 ppm, levels which
correspond to 1 and 2 times the doses that were used in
an in vivo feeding study for beneficial effects against rat
colon carcinogenesis [20].
2.2. Experimental design
The protocols used were consistent with Ethical Prin-
ciples for Animal Research adopted by the Brazilian
College of Animal Experimentation (COBEA). The
animals were randomly allocated to two experimental
protocols: a short- and a medium-term liver bioassay,
respectively (Fig. 1).
2.2.1. Experiment 1 (short-term bioassay)
This study was performed to investigate the modify-
ing effects of EGb intake on the first stage of rat liver
carcinogenesis in an initiation bioassay model that uses
the detection of initiated hepatocytes and minifoci pos-
itive for GST-P as the endpoint [24,25]. The animals
were randomly allocated into four groups: groups G1A
and G4A were fed basal diet and groups G2A and G3A
were fed a diet supplemented with 500 or 1000 ppm EGb
during 2 weeks. Then, groups G1A to G3A were given
a single i.p. injection of 100 mg/kg b.w. of diethylni-
trosamine and group 4A (control) received 0.9% NaCl
(DEN vehicle). Twenty-four hours after DEN admin-
istration, groups G1A to G3A were fed basal diet ad
libitum. The animals were sacrificed 24 h or 2 weeks
after DEN treatment (end of week 4).
2.2.2. Experiment 2 (medium-term bioassay)
This study was performed to investigate the modify-
ing effects of EGb intake on the second stage of rat liver
carcinogenesis in a post-initiation bioassay model that
uses preneoplasia detection (GST-P-positive foci) as the
endpoint [26–29]. The animals were randomly allocated
34 M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42

Fig. 1. Experimental design of the short-term and medium-term liver bioassays. () 100 or ( ) 200 mg/kg body weight of diethylnitrosamine (DEN);
() 0.9% NaCl; ( ) 70% partial hepatectomy; () diet containing 500 ppm EGb; () diet containing 1000 ppm EGb; n = number of animals/group;
s1–3 : sacrifice 24 h, 2 and 8 weeks after DEN-initiation, respectively.

into five groups. At the beginning of the experiment,
groups G1B to G3B were given a single i.p. injection of
200 mg/kg b.w. of DEN and groups 4B and 5B received
0.9% NaCl. From the 2nd week, groups G2B, G3B and
G5B were fed for 6 weeks a diet supplemented with 500,
1000 or 1000 ppm EGb, respectively. Groups G1B and
G4B were maintained on the basal diet until the end of
the experiment at week 8. All animals were submitted to
70% partial hepatectomy (PH) at week 3 and sacrificed
at week 8. PH was performed in order to induce vigor-
ous regenerative cell replication to increase the initiated
hepatocyte population [29].
In experimental protocols, clinical examinations were
performed daily and detailed physical examinations
were accomplished weekly. Food and water consump-
tions were recorded twice a week (Tuesday/Wednesday
and Thursday/Friday, 18:00 p.m. to 8:00 a.m.) and the
animals were weighed individually once a week through-
out the experimental periods. The diets containing 500
and 1000 ppm of EGb were prepared weekly, maintained
under −4 ◦ C and offered to the rats daily ad libitum
to preserve the stability of specific components (mainly
flavonoids) in EGb-supplemented diet. The animals were
fasted overnight and sacrificed by exsanguination under
sodium pentobarbital anesthesia (45 mg/kg p.c.).
2.3. Biochemical, tissue processing, histology and
immunohistochemical procedures
Immediately before necropsy, whole blood was col-
lected and serum enzyme analyses for alanine amino
transferase (ALT) and aspartate amino transferase (AST)
were carried out spectrophotometrically (Ortho-Clinical
Diagnostics, Johnson & Johnson Co., SP, Brazil) to mon-
itor hepatocellular injury.
At sacrifice, the liver was removed and weighed
after which samples were collected and fixed in
10% phosphate-buffered formalin solution, embedded
in paraffin, cut (5 ␮m thickness) and stained with
hematoxylin and eosin (H&E). Immunohistochemical
investigation of glutathione S-transferase P-form (GST-
P), transforming growth factor alpha (TGF-␣), p53 and
proliferating cell nuclear antigen (PCNA) expression
was performed as follows: Briefly, deparaffinized 5-
␮m-thick serial liver sections on poly-l-lysine coated
slides were treated with 3% H2 O2 in phosphate-buffered
saline for 15 min, nonfat milk for 60 min, polyclonal
anti-rabbit GST-P (Medical and Biological Laboratories
Co., Tokyo, Japan, clone 311, 1:1000 dilution), mono-
clonal anti-mouse TGF-␣ (Oncogene Science Inc., New
York, USA, clone Ab-2, 1:200 dilution), monoclonal
 M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42
anti-mouse PCNA (Dako Corporation, Carpinterie, CA,
USA, clone PC10, 1:200 dilution), polyclonal anti-sheep
p53 (Roche, Mannheim, Germany, clone BGM-1B1,
1:50 dilution) antibodies for 12 h at 4 ◦ C, biotinylated
anti-rabbit, anti-mouse or anti-sheep IgG antibodies
(Vector Laboratories, Inc., Burlingame, CA, USA, 1:200
dilution) for 60 min, and streptavidin–biotin–peroxidase
solution (TissuGnost Kit was from Merck, Darmstadt,
Germany, 1:1:50 dilution) for 45 min. Antigen retrieval
was performed for sections submitted to TGF-␣ and p53
immunostaining as follows: for TGF-␣, 0.25% trypsin
(Sigma–Aldrich Co., St. Louis, MO, USA) for 15 min
and 0.05% saponin (Calbiochem, La Jolla, CA, USA)
solution for 30 min. For p53, 0.01M citrate buffer (pH
6.0) heated twice, 5 min each time, in a microwave
oven. Chromogen color development was accomplished
with 3,3 -diaminobenzidine tetrahydrochloride (DAB,
Sigma-Aldrich Co.) as the substrate to demonstrate
the sites of peroxidase binding. The slides were coun-
terstained with Harris’s hematoxylin or H&E. Liver
sections were processed by omitting incubation with the
primary antibodies (GST-P, TGF-␣, p53 and PCNA) as
negative controls for all the immunoreactions.
2.4. Immunoreactivity analysis of GST-P, TGF-α,
p53 and PCNA markers
Single hepatocytes and minifoci (2–15 hepatocytes)
positive for GST-P (short-term liver bioassay) and
GST-P-positive FAH larger than 0.15 mm in diame-
ter (medium-term liver bioassay) were analyzed using
a Nikon photomicroscope (Microphot-FXA) connected
to a KS-300 apparatus (Kontron Elektronic, Germany).
The liver areas were measured in a special Macro-
Stand device (supported by Canon TV zoom lens
V6×16/16–100 mm plus a Canon 58 mm close-up 240
lens connected to a CCD black-and-white video camera
module with a Sony DC-777 camera unit) connected to
the KS-300. Data were expressed as number/cm2 (sin-
gle hepatocytes, minifoci or FAH per liver unit area)
and aggregated area (mm2 /cm2 ) for GST-P-positive
FAH.
For short-term bioassay, immunoreactivity for TGF-
␣ was analyzed in the hepatocytes from zone 3 and/or
zone 2 of the liver acinus. The PCNA S-phase and p53
labeling indices (LI%) were calculated as the percentage
of labeled hepatocyte nuclei divided by the total num-
ber of cells scored (∼2000 hepatocytes). The apoptosis
index (AI%) was defined as the number of hepato-
cytes in apoptosis divided by the total number of cells
scored (∼2000 hepatocytes). Criteria for identification
and quantification of hepatocytes in apoptosis and of
35
apoptotic bodies in liver sections stained with H&E
were taken from the literature [30]. For medium-term
bioassay, the number of clearly discernable TGF-␣-
positive FAH per liver unit area was analyzed with
the aid of KS 300 system as indicated above. PCNA
labeling and apoptotic indices were determined by eval-
uating all S-phase labeled hepatocytes and hepatocytes
in apoptosis, involving at least 2000 hepatocytes/rat in
randomly selected FAH identified by H&E staining. The
immunoreactivity for p53 (wild-type and mutated forms)
was also analyzed in the FAH identified by H&E stain-
ing.
2.5. Statistical analysis
The statistical analysis was performed using the Jan-
del Sigma Stat software (Jandel Corporation, San Rafael,
CA, USA). Body weight and body-weight gain, absolute
and relative liver weights, food consumption, number
of putative hepatocytes and FAH positive for GST-
P and TGF-␣, the mean size and aggregated area of
FAH and serum biochemistry data were analyzed by
the ANOVA or Kruskal–Wallis tests. PCNA labeling
and apoptosis data were analyzed by the Mann–Whitney
or Kruskal–Wallis tests. Significant differences were
assumed when P < 0.05.
3. Results
3.1. Short-term liver bioassay (anti-initiating
screening)
No significant alterations in body weight, body-
weight gain or food consumption associated with EGb
ingestion was observed in weeks 1 and 2, before DEN-
treatment (Table 1). In contrast, body weight and food
consumption were significantly reduced in the days
immediately after DEN initiation (100 mg/kg b.w.) when
compared to the non-initiated group (G1A to G3A
vs. G4A, P < 0.001) (data not shown). At the end of
week 4, body weight and food consumption of DEN-
initiated groups had returned to normal levels (data not
shown).
In animals killed 24 h after DEN administration,
all DEN-initiated groups had significantly higher lev-
els of serum transaminases (AST and ALT) and rates
of cell proliferation, apoptosis and TGF-␣ expression
than the non-initiated group (G1A to G3A vs. G4A,
P < 0.001) (Table 1, Figs. 2 and 3). Hepatocytes exhibit-
ing immunoreactivity to p53 and TGF-␣ were observed
in zones 2 and 3 of liver acinus in DEN-initiated
control group (Fig. 2) while a few centrilobular TGF-
36 M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42

Table 1
General, liver, biochemical and single/minifoci positive for GST-P data in the short-term bioassay
Parameters Groupsa

DEN 500 EGb + DEN 1000 EGb + DEN Control

(G1A, n = 10) (G2A, n = 10) (G3A, n = 10) (G4A, n = 05)

General datab
Final body weight (g) 255.8 ± 23.91 265.0 ± 30.13 270.00 ± 22.44 280.40 ± 22.44
Body-weight gain (g) 88.60 ± 18.89 97.60 ± 10.74 104.20 ± 12.73 113.60 ± 7.70
Food consumption (g/rat/day) 24.3 ± 2.05 23.70 ± 2.32 23.91 ± 2.18 24.30 ± 2,65
EGb consumption (mg/rat/day) 0 11.85 ± 1.16 23.91 ± 2.18 0
Liver and biochemical datab
Liver absolute weight (g) 10.05 ± 1.36 10.57 ± 1.78 10.79 ± 0.79 12.62 ± 0.93
Relative liver weight (%) 3.93 ± 0.24 3.98 ± 0.39 4.00 ± 0.40 4.50 ± 0.40
AST (U/L) 215.60 ± 43.26* 196.40 ± 31.66* 155.0 ± 38.39*,§ 91.80 ± 20.91
ALT (U/L) 123.80 ± 8.87* 124.60 ± 14.55* 106.4 ± 9.02*,§ 81.60 ± 8.20
GST-P-positive hepatocyte datac
Single (cm2 ) 36.06 ± 6.92 29.16 ± 9.62 18.86 ± 4.04* ND
Minifoci (cm2 ) 4.61 ± 0.53 3.17 ± 0.77 2.38 ± 1.08* ND
Total (cm2 ) 46.84 ± 7.38 41.24 ± 8.36 24.76 ± 5.24* ND

Values are means ± S.D.

* Different from group G4A, 0.05 < P < 0.001.
§ Trend from G1A, 0.05 < P < 0.07; ND: not determined.
a DEN: Diethylnitrosamine (100 mg/kg b.w., i.p.), EGb: Ginkgo biloba extract at 500 and 1000 ppm in basal diet 2 weeks before and 24 h after

DEN-initiation.
b ,c Sacrifice 24 h and 2 weeks after DEN administration, respectively.

␣-positive hepatocytes and no staining for wild-type
p53 protein were observed in non-initiated group liver
(data not shown). A reduction of TGF-␣ expression,
which was detected only in zone 3 hepatocytes, was
observed in 1000 ppm EGb DEN-treated animals. This
group also presented a significant decrease in the mean
number of PCNA S phase-positive and p53-positive
hepatocytes and of hepatocytes in apoptosis when com-
pared to the liver of the respective control group (G3A
vs. G1A, 0.05 < P < 0.03 and P = 0.049, respectively)
(Figs. 2 and 3). Serum ALT and AST levels were lower
in the 1000 ppm EGb DEN-treated group than in the
respective control group (G3A vs. G1A, P = 0.055 and
0.065, respectively) (Table 1).
In the animals killed 2 weeks after DEN-initiation,
single hepatocytes and minifoci positive for GST-P
were mainly located in zones 2 and 3 and, to a
lesser extent, in zone 1 of liver acinus in all DEN-
initiated groups (Fig. 4A and B). At the end of week
4, the 1000 ppm EGb treatment, administered before
and during DEN-initiation, had significantly decreased
the mean number of GST-P-positive single hepatocytes
and minifoci per liver unit area (P < 0.001) when com-
pared to the respective control group (G3A vs. G1A)
(Table 1).
3.2. Medium-term liver bioassay (anti-promoting
screening)
The 6-week post-initiation treatment with EGb did
not cause any significant alterations in body weight,
body-weight gain or food consumption in both non-
initiated and DEN-initiated groups (Table 2). At the end
of the experiment, neither liver weights (absolute and rel-
ative) nor serum ALT and AST levels were modified by
EGb treatment (Table 2). The possible modifying effect
of EGb treatment per se on cell proliferation and apop-
tosis was evaluated in the non-initiated groups (G4B and
G5B). Dietary 1000 ppm EGb dose level did not change
PCNA labeling or apoptotic indices in the liver (G4B vs.
G5B) (Table 2).
Post-initiation treatment with EGb did not sup-
press the development of putative preneoplastic GST-P
or TGF-␣-positive FHA induced by DEN treatment
(Fig. 4C and D, Table 2). PCNA S-phase label-
ing and apoptotic indices were not altered into FAH
by 6-week EGb treatment. No immunoreactivity for
p53 was detected in FAH in this early stage of rat
liver carcinogenesis. Non-initiated groups (G4B and
G5B) did not develop any GST-P or TGF-␣-positive
FHA.
 M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42 37

Fig. 2. Liver from DEN-initiated group (A, C, E, G) and 1000 ppm EGb plus DEN-initiated group (B, D, F, H), showing apoptosis (A vs. B), PCNA-
positive hepatocytes (C vs. D), p53 nuclear expression (E vs. F) and TGF-␣ expression (G vs. H) in the short-term bioassay. Arrows: Apoptotic
hepatocytes, TV: terminal venule.
38 M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42
Fig. 3. Effects of EGb treatment on liver PCNA labeling (PCNA
LI%), p53 labeling (p53 LI%) and apoptotic (AI%) indices in short-
term bioassay. G1A = DEN, G2A = 500 EGb + DEN, G3A = 1000
EGb + DEN, G4A = control. DEN: Diethylnitrosamine (100 mg/kg
b.w., i.p.), EGb = Ginkgo biloba extract at 500 and 1000 ppm in
basal diet for 2 weeks before and 24 h after DEN-initiation. Val-
ues are mean ± S.D., * different from group G4A, 0.05 < P < 0.001;
**,*** different from G1A, 0.05 < P < 0.03 and P = 0.049, respectively.
4. Discussion
The results described herein indicate that EGb intake
inhibited the initiation but not the promotion stage of
rat liver carcinogenesis induced by DEN. Therefore,
EGb acted as an anti-initiating agent but did not present
any suppressing effect on the promotion step of chemi-
cally induced rat liver carcinogenesis under the present
experimental conditions. Importantly, rats fed 1000 ppm
EGb did not present any adverse effects after 6-week
exposure period, as indicated by unaltered body and
liver weights, transaminase levels, liver morphology and
kinetics parameters such as cell proliferation and apop-
tosis rates. This absence of toxicity should be taken into
account if safety measures for public health are to be
implemented in response to increased ingestion of this
herbal drug by human populations, given that EGb has
been clinically prescribed for the treatment of various
diseases [3–6].
The hepatoprotective effect of EGb administra-
tion in the early phase of rat liver carcinogenesis
could be due to a modifying influence on the bio-
transformation/detoxification of DEN, thus reducing
its liver toxicity, mutagenicity, and carcinogenicity.
CYP2E1 is known to be associated primarily with the
biotransformation of a range of compounds, includ-
ing DEN in the liver, but other P450 isozymes
have been found to bioactivate as well [31]. It has
been reported that their biotransformation produces
the promutagenic adducts O6 -ethyldeoxyguanosine, O4
and O6 -ethyldeoxythymidine and 8-hydroxyguanine (8-
OHG) that play a role in the initiation step of rat liver
carcinogenesis [31–34]. Therefore, it could be suggested
that EGb intake before and during initiation with DEN
may provide a protective effect against carcinogenesis,
since this hepatocarcinogen requires metabolic activa-
tion.
After DEN exposure, one hepatic response to
DNA damage and centrilobular cytotoxicity/necrosis is
characterized by regenerative cell proliferation [34].
The response to the damage in both non-initiated
and initiated hepatocytes is associated with the over-
expression of oncogenes/suppressor tumor genes and
growth factors, including p53 and TGF-␣, respectively
[35,36].
Tumor suppressor protein p53 can protect cells from
growth and division by mediating cell-cycle arrest, DNA
repair and apoptosis [37]. The positivity for p53 in the
liver of rats treated with DEN may be attributed to
the DNA damage induced by this hepatocarcinogenic
dosage, resulting in cell death and delayed entrance into
critical phases of the cell cycle such as early S or M [35-
38]. PCNA, a 36kDa molecule that acts as a co-factor for
delta-DNA-polymerase leading to DNA replication, has
been considered a feasible marker for liver cell prolifer-
ation [39]. Thus, PCNA and p53 analyses are believed to
be important indicators of S-phase DNA synthesis and
cell-cycle progression [37,39]. TGF-␣ is a growth fac-
tor (mitogen stimulus) that activates the protein tyrosine
kinase activity of epidermal growth factor receptor and
M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42 39

plays an important role in liver regeneration and progres-

sion of carcinogenesis [40,41]. The increase of TGF-␣
expression in DEN-initiated animals could be associ-
ated with the increase in cell proliferation as seen after
tissue deficit created by partial hepatectomy [40]. In the
present study, the beneficial influence of EGb on cell
loss induced by DEN was characterized by a down reg-
ulation of PCNA and TGF-␣ expression, both regulated
by p53 function [39,42]. Therefore, the inhibitory effect
of EGb may be due to its modulation of DEN-induced
DNA damage.
As initiation is a rare event affecting only a few hep-
atocytes, the number of initiated hepatocytes is a major
determinant of the risk for development of liver cancer
and could be used as an endpoint for chemopreven-
tive approaches [26]. Single hepatocytes and minifoci
positive for GST-P develop very early in carcinogen-
treated rats and are mainly located in zones 2 and 3
of liver acinus after a single DEN treatment [24,25].
Although numerical reductions of both single hepato-
cytes and minifoci positive for GST-P were observed in
rats fed 1000 ppm EGb, not all GST-P-positive single
hepatocytes are thought to give rise to putative preneo-
plastic foci and liver tumors [25,43]. Thus, an effective
influence of EGb on the initiation stage of liver carcino-
genesis should continue to be investigated, mainly in
long-term bioassays.
No reports are available on the protective effects of
EGb treatment during the promotion step of chemical
carcinogenesis in rodents. Our data showed that EGb
intake did not modify the development of GST-P and
TGF-␣-positive foci in early phases of rat liver carcino-
genesis. Immunohistochemical expression of TGF-␣ has
been demonstrated in rat liver preneoplastic and neo-
plastic lesions, although a few GST-P-positive FAH
exhibit immunoreactivity for this growth factor [41]. A
selective reduction of TGF-␣-positive foci could rep-
resent a potential chemopreventive property of EGb
since this growth factor may act as a potent hep-
atocyte mitogen during liver carcinogenesis process
[41]. Probably, higher doses of EGb could be more
effective in reducing the clonal expansion of initiated
hepatocytes after DEN initiation or perhaps the anti-
promoting potential of EGb may be expressed in target

Fig. 4. Immunohistochemistry for glutathione S-transferase placen-

tal form (GST-P) (A–C). Putative single hepatocyte (A) and minifoci
(B) positive for GST-P were analyzed in the short-term bioassay. In
addition, foci of altered hepatocytes (FAH) (C) positive for GST-P
were analyzed in the medium-term bioassay. This picture represents
a progressive evolution of the biomarkers during early-phase rat liver
carcinogenesis.
40 M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42

Table 2
General, liver, biochemical and FAH data in the medium-term bioassay
Parameters Groupsa

DEN DEN + 500 EGb DEN + 1000 EGb Control 1000 EGb
(G1B, n = 10) (G2B, n = 10) (G3B, n = 10) (G4B, n = 05) (G5B, n = 05)

General data
Final body weight (g) 358.98 ± 30.82 357.89 ± 33.12 367.22 ± 29.71 371.88 ± 18.42 380.68 ± 17.00
Body-weight gain (g) 105.06 ± 21.75 102.48 ± 25.29 110.30 ± 22.29 122.10± 14.39 126.68 ± 21.58
Food consumption (g/rat/day) 27.01 ± 2.58 26.70 ± 2.84 27.27 ± 2.52 27.46 ± 3.10 27.20 ± 3.30
EGb consumption (mg/rat/day) 0 13.35 ± 1.42 27.27 ± 2.52 0 27.20 ± 3.30

Liver and biochemical data

Liver absolute weight (g) 9.33 ± 1.33 9.55 ± 1.07 9.37 ± 0.83 9.08 ± 1.09 10.12 ± 0.90
Relative liver weight (%) 2.59 ± 0.22 2.66 ± 0.11 2.55 ± 0.11 2.48 ± 0.25 2.66 ± 0.18
AST (U/L) 85.33 ± 14.6 91.92 ± 22.2 83.58 ± 16.30 82.57 ± 9.54 83.00 ± 7.71
ALT (U/L) 66.75 ± 16.1 69.00 ± 14.1 61.17 ± 8.89 58.43 ± 11.84 55.80 ± 10.18
PCNA LI (%)b 1.75 ± 0.55 1.42 ± 0.65 1.62 ± 0.45 0.42 ± 0.15 0.33 ± 0.05
AI%b 0.15 ± 0.07 0.13 ± 0.05 0.14 ± 0.08 0.08 ± 0.04 0.07 ± 0.03

FHA data
GST-P-positive foci
Number (foci/cm2 ) 9.21 ± 4.44 12.1 ± 4.53 11.65 ± 2.96 0 0
Area (mm2 /cm2 ) 0.23 ± 0.14 0.27 ± 0.15 0.33 ± 0.17 0 0
TGF-␣-positive foci
Number (foci/cm2 ) 1.32 ± 0.35 1.75 ± 0.46 1.65 ± 0.51 0 0

Values are means ± S.D.

a DEN: Diethylnitrosamine (200 mg/kg b.w., i.p.), EGb: Ginkgo biloba extract at 500 and 1000 ppm in basal diet for six weeks; n = number of

rats/group.
b PCNA S-phase-positive hepatocytes and hepatocytes in apoptosis were determined in foci of altered hepatocytes (FHA) (groups G1B to G3B)

and in normal liver (groups G4B and G5B).

organs other than the liver. Given that oxidative stress
is potentially deleterious to cells and associated with
the progression of many diseases, including cancer
[44], the antioxidant effects derived from crude EGb
or their specific components [7-11] would be a poten-
tial mechanism for cancer prevention in a late phase
of rat liver carcinogenesis. Since EGb has been shown
to induce cell-cycle arrest and apoptosis in different
human cancer cell lines [15–18], this herbal drug may
be useful for the prevention of liver tumor progres-
sion.
In conclusion, the results of the present study
indicate that EGb presented inhibitory actions during
initiation but not post-initiation stage of rat liver car-
cinogenesis induced by DEN. Thus, this herbal drug
presents blocking actions but not suppressing actions
during chemically induced liver carcinogenesis [45].
Our findings are interesting for their possible appli-
cation to human cancer chemoprevention among the
population continuously exposed to certain carcino-
gens. The underlying mechanism(s) of chemoprevention
of DEN-induced hepatocarcinogenesis still must be
investigated.
Acknowledgements
This study was supported by CAPES and TOXICAM.
Dias, M.C. was recipient of a fellowship from CAPES.
References
[1] B.P. Jacobs, W.S. Browner, Ginkgo biloba: a living fossil, Am. J.
Med. 108 (2000) 341–342.
[2] B. Ahlemeyer, J. Krieglstein, Neuroprotective effects of Ginkgo
biloba extract, Cell Mol. Life Sci. 60 (2003) 1779–1792.
[3] F.V. DeFeudis, V. Papadopoulos, K. Drieu, Ginkgo biloba extracts
and cancer: a research area in its infancy, Fundam. Clin. Pharma-
col. 17 (2003) 405–417.
[4] T. Boonkaew, N.D. Camper, Biological activities of Ginkgo
extracts, Phytomedicine 12 (2004) 318–323.
[5] W.R. Zhang, T. Hayashi, H. Kitagawa, Protective effect of ginkgo
extract on rat brain with transient middle cerebral artery occlusion,
Neurol. Res. 22 (2000) 517–521.
[6] Y. Luo, Ginkgo biloba neuroprotection: therapeutic implications
in Alzheimer’s disease, J. Alzheimers Dis. 3 (2001) 401–407.
[7] J. Ding, J. Yu, C. Wang, Ginkgo biloba extract alleviates liver
fibrosis induced by CCl4 in rats, Liver Int. 25 (2005) 1224–1232.
[8] S.X. He, J.Y. Luo, Y.P. Wang, Effects of extract from Ginkgo
biloba on carbon tetrachloride-induced liver injury in rats, World
J. Gastroenterol. 12 (2006) 3924–3928.
 M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42
[9] G. Sener, G.Z. Omurtag, O. Sehirli, Protective effects of Ginkgo
biloba against acetaminophen-induced toxicity in mice, Mol. Cell
Biochem. 283 (2006) 39–45.
[10] S.R. Naik, V.S. Panda, Antioxidant and hepatoprotective effects of
Ginkgo biloba phytosomes in carbon tetrachloride-induced liver
injury in rodents, Liver Int. 27 (2007) 393–399.
[11] K. Sasaki, S. Hatta, K. Wada, Effects of extract of Ginkgo biloba
leaves and its constituents on carcinogen-metabolizing enzyme
activities and glutathione levels in mouse liver, Life Sci. 70 (2002)
1657–1667.
[12] S.D. Ryu, W.G. Chung, Induction of the procarcinogen-activating
CYP1A2 by a herbal dietary supplement in rats and humans, Food
Chem. Toxicol. 41 (2003) 861–866.
[13] C. Gaudineau, R. Beckerman, S. Welbourn, K. Auclair, Inhibition
of human P450 enzymes by multiple constituents of the Ginkgo
biloba extract, Biochem. Biophys. Res. Commun. 318 (2004)
1072–1078.
[14] T. Sugiyama, Y. Kubota, K. Shinozuka, S. Yamada, K. Yamada,
K. Umegaki, Induction and recovery of hepatic drug metabolizing
enzymes in rats treated with Ginkgo biloba extract, Food Chem.
Toxicol. 42 (2004) 953–957.
[15] J.C. Chao, C.C. Chu, Effects of Ginkgo biloba extract on cell
proliferation and cytotoxicity in human hepatocellular carcinoma
cells, World J. Gastroenterol. 10 (2004) 37–41.
[16] Q. Chen, G.W. Yang, L.G. An, Apoptosis of hepatoma cells
SMMC-7721 induced by Ginkgo biloba seed polysaccharide,
World J. Gastroenterol. 8 (2002) 832–836.
[17] B. Ye, M. Aponte, Y. Dai, L. Li, M.C. Ho, A. Vitonis, D. Edwards,
T.N. Huang, D.W. Cramer, Ginkgo biloba and ovarian cancer
prevention: epidemiological and biological evidence, Cancer Lett.
251 (2007) 43–52.
[18] K.S. Kim, K.H. Rhee, J.H. Yoon, J.G. Lee, J.H. Lee, J.B. Yoo,
Ginkgo biloba extract (EGb 761) induces apoptosis by the acti-
vation of caspase-3 in oral cavity cancer cells, Oral Oncol. 41
(2005) 383–389.
[19] A.M. Agha, A.A. El-Fattah, H.H. Al-Zuhair, A.C. Al-Rikabi,
Chemopreventive effect of Ginkgo biloba extract against
benzo(a)pyrene-induced forestomach carcinogenesis in mice:
amelioration of doxorubicin cardiotoxicity, J. Exp. Clin. Cancer
Res. 20 (2001) 39–50.
[20] R. Suzuki, H. Kohno, S. Sugie, Preventive effects of extract of
leaves of ginkgo (Ginkgo biloba) and its component bilobalide on
azoxymethane-induced colonic aberrant crypt foci in rats, Cancer
Lett. 210 (2004) 159–169.
[21] P. Bannasch, H. Zerban, Predictive value of hepatic preneoplastic
lesions as indicators of carcinogenic response, in: H. Vainio, P.
Magee, D. McGregor, A.J. McMichael (Eds.), Mechanism of Car-
cinogenesis in Risk Identification, IARC Scientific Publication
116, International Agency of Research on Cancer, Lyon, 1992,
pp. 389–427.
[22] Q. Su, P. Bannasch, Relevance of hepatic preneoplasia for human
hepatocarcinogenesis, Toxicol. Pathol. 31 (2003) 126–133.
[23] M. Tatematsu, T. Aoki, M. Kagawa, Y. Mera, N. Ito, Recipro-
cal relationship between development of glutathione S-transferase
positive liver foci and proliferation of surrounding hepatocytes in
rats, Carcinogenesis 9 (1988) 221–225.
[24] M.A. Moore, K. Nakagawa, K. Satoh, T. Ishikawa, K. Sato, Single
GST-P positive liver cells-putative initiated hepatocytes, Carcino-
genesis 8 (1987) 483–486.
[25] K. Satoh, I. Hatayama, N. Tateoka, Transient induction of single
GST-P positive hepatocytes by DEN, Carcinogenesis 10 (1989)
2107–2111.
41
[26] M.A. Moore, H. Tsuda, S. Tamano, Marriage of a medium-term
liver model to surrogate markers: a practical approach for risk and
benefit assessment, Toxicol. Pathol. 27 (1999) 237–242.
[27] N. Ito, S. Tamano, T. Shirai, A medium-term rat liver bioassay
for rapid in vivo detection of carcinogenic potential of chemicals,
Cancer Sci. 94 (2003) 3–8.
[28] F. Pinheiro, R.R. Faria, J.L.V. de Camargo, A.L. Spinardi-
Barbisan, E.F. da Eira, L.F. Barbisan, Chemoprevention of
preneoplastic liver foci development by dietary mushroom Agar-
icus blazei Murrill in the rat, Food Chem. Toxicol. 41 (2003)
1543–1550.
[29] N. Ito, H. Tsuda, M. Tatematsu, T. Inoue, Y. Tagawa, T. Aoki,
S. Uwagawa, M. Kagawa, T. Ogiso, T. Masui, Enhancing effect
of various hepatocarcinogens on induction of preneoplastic glu-
tathione S-transferase placental form positive foci in rats—an
approach for a new medium-term bioassay system, Carcinogen-
esis 9 (1988) 387–394.
[30] S. Levin, T.J. Bucci, S.M. Cohen, The nomenclature of cell death:
recommendations of an ad hoc Committee of the Society of Tox-
icologic Pathologists, Toxicol. Pathol. 27 (1999) 484–490.
[31] L. Verna, J. Whysner, G.M. Williams, N-Nitrosodiethylamine
mechanistic data and risk assessment: bioactivation, DNA-adduct
formation, mutagenicity, and tumor initiation, Pharmacol. Ther.
71 (1996) 57–81.
[32] J.A. Swenberg, D.G. Hoel, P.N. Magee, Mechanistic and
statistical insight into the large carcinogenesis bioassays on N-
nitrosodiethylamine and N-nitrosodimethylamine, Cancer Res.
51 (1991) 6409–6414.
[33] D. Nakae, Y. Kobayashi, H. Akai, N. Andoh, H. Satoh, K. Ohashi,
M. Tsutsumi, Y. Konishi, Involvement of 8-hydroxyguanine
formation in the initiation of rat liver carcinogenesis by low
dose levels of N-nitrosodiethylamine, Cancer Res. 57 (1997)
1281–1287.
[34] G.M. Williams, M.J. Iatropoulos, C.X. Wang, Diethylni-
trosamine exposure-responses for DNA damage, centrilobular
cytotoxicity, cell proliferation and carcinogenesis in rat liver
exhibits some non-linearities, Carcinogenesis 17 (1996) 2253–
2258.
[35] P. Lennartsson, J. Hogberg, U. Stenius, Wild-type p53 expression
in liver tissue and in enzyme-altered foci: an in vivo investiga-
tion on diethylnitrosamine-treated rats, Carcinogenesis 19 (1998)
1231–1237.
[36] P. Lennartsson, U. Stenius, J. Hogberg, P53 expression and
TGF-alpha-induced replication of hepatocytes isolated from rats
exposed to the carcinogen diethylnitrosamine, Cell Biol. Toxicol.
15 (1999) 31–39.
[37] A.J. Levine, P53: the cellular gatekeeper for growth and division,
Cell 88 (1997) 323–331.
[38] Y.P. Dragan, J.R. Hully, J. Nakamura, M.J. Mass, J.A. Swenberg,
H.C. Pitot, Biochemical events during initiation of rat hepatocar-
cinogenesis, Carcinogenesis 15 (1994) 1451–1458.
[39] J. Foley, T. Ton, R. Maronpot, B. Butterworth, T.L. Goldsworthy,
Comparison of proliferating cell nuclear antigen to tritiated thymi-
dine as a marker of proliferating hepatocytes in rats, Environ.
Health Perspect. 5 (1993) 199–205.
[40] J.E. Mead, N. Fausto, Transforming growth factor alpha may be
a physiological regulator of liver regeneration by means of an
autocrine mechanism, Proc. Natl. Acad. Sci. U.S.A. 86 (1989)
1558–1562.
[41] M. Gitano, J. Wada, Y. Ariki, M. Kato, H. Wanibuchi, K.
Morimura, T. Hidaka, K. Hosoe, S. Fukushima, Possible tumour
development from double positive foci for TGF-alpha and GST-P
42 M.C. Dias et al. / Chemico-Biological Interactions 173 (2008) 32–42

observed in early stages on rat hepatocarcinogenesis, Cancer Sci.
97 (2006) 478–483.
[42] Y. Inoue, T. Tomiya, M. Yanase, P53 may positively regulate
hepatocyte proliferation in rats, Hepatology 36 (2002) 336–344.
[43] K. Satoh, I. Hatayama, Anomalous elevation of glutathione
S-transferase P-form (GST-P) in the elementary process of
epigenetic initiation of chemical hepatocarcinogenesis in rats,
Carcinogenesis 23 (2002) 1193–1198.
[44] E.S. Hwang, G.H. Kim, Biomakers for oxidative stress status of
DNA, lipids, and proteins in vitro and in vivo cancer research,
Toxicology 229 (2007) 1–10.
[45] L.W. Wattenberg, Inhibition of tumorigenesis in animals, in:
B.W. Stewart, D. McGregor, P. Kleihwes (Eds.), Principles of
Chemoprevention, IARC Scientific Publication 139, International
Agency of Research on Cancer, Lyon, 1996, pp. 151–158.