Oxidative stress in carcinogenesis. Correlation between lipid peroxidation and induction of preneoplastic lesions in rat hepatocarcinogenesis.

Abstract

Oxidative stress during carcinogen metabolism seems to participate in liver tumor production in the rat. N-diethylnitrosamine is an important carcinogen used in liver cancer animal models. This indirect alkylating agent produces DNA-ethyl adducts and oxidative stress. In contrast, N-ethyl-N-nitrosourea, a direct mutagen, which generates DNA-ethyl adducts, does not produce liver tumors in rat unless it is given under oxidative stress conditions such as partial hepatectomy or phenobarbital treatment. To gain insight into the relation between oxidative stress and hepatocarcinogenicity, the induction of preneoplastic liver lesions was compared among three different initiation protocols related to the initiation-promotion-resistant hepatocyte model. In addition, liver lipid peroxidation levels, determined as thiobarituric acid reactive substances were studied early during the initiation stage. Rats initiated with N-ethyl-N-nitrosourea, 25 days after treatment developed fewer and smaller gamma-glutamyl transpeptidase positive preneoplastic lesions than rats initiated with N-diethylnitrosamine. A pre-treatment with the antioxidant quercetin 1 h before N-diethylnitrosamine initiation, significantly prevented development of gamma-glutamyl transpeptidase-positive lesions. Increased lipid peroxidation levels were induced with N-diethylnitrosamine from 3 to 24 h after initiation, while N-ethyl-N-nitrosourea did not induce increments, and importantly, pre-treatment with quercetin decreased lipid peroxidation induced by N-diethylnitrosamine. These results show correlation between lipid peroxidation and hepatocarcinogenicity and support the important role of oxidative stress on liver carcinogenesis.

Full text

Page 1

Oxidative stress in carcinogenesis. Correlation between lipid
peroxidation and induction of preneoplastic lesions in rat
hepatocarcinogenesis
Yesennia Sa ´nchez-Pe ´reza, Claudia Carrasco-Legleua, Claudia Garcı ´a-Cuellarb,
Julio Pe ´rez-Carreo ´na, Sergio Herna ´ndez-Garcı ´aa, Martha Salcido-Neyoya,
Leticia Alema ´n-Lazarinia, Sau ´l Villa-Trevin ˜oa,*
aDepartamento de Biologı ´a Celular, Centro de Investigacio ´n y de Estudios Avanzados del IPN (CINVESTAV),
Av. IPN No. 2508 Col. San Pedro Zacatenco, Me ´xico 14, DF, CP 07360, Mexico
bInstituto Nacional de Cancerologı ´a, San Fernando No. 22, Tlalpan, Me ´xico, DF 14080, Mexico
Received 1 June 2004; received in revised form 1 July 2004; accepted 6 July 2004
Abstract
Oxidative stress during carcinogen metabolism seems to participate in liver tumor production in the rat. N-diethylnitrosamine
is an important carcinogen used in liver cancer animal models. This indirect alkylating agent produces DNA–ethyl adducts and
oxidative stress. In contrast, N-ethyl-N-nitrosourea, a direct mutagen, which generates DNA–ethyl adducts, does not produce
liver tumors in rat unless it is given under oxidative stress conditions such as partial hepatectomy or phenobarbital treatment. To
gain insight into the relation between oxidative stress and hepatocarcinogenicity, the induction of preneoplastic liver lesions
was compared among three different initiation protocols related to the initiation–promotion-resistant hepatocyte model. In
addition, liver lipid peroxidation levels, determined as thiobarituric acid reactive sustances were studied early during the
initiation stage. Rats initiated with N-ethyl-N-nitrosourea, 25 days after treatment developed fewer and smaller g-glutamyl
transpeptidase positive preneoplastic lesions than rats initiated with N-diethylnitrosamine. A pre-treatment with the antioxidant
quercetin 1 h before N-diethylnitrosamine initiation, significantly prevented development of g-glutamyl transpeptidase-positive
lesions. Increased lipid peroxidation levels were induced with N-diethylnitrosamine from 3 to 24 h after initiation, while
N-ethyl-N-nitrosourea did not induce increments, and importantly, pre-treatment with quercetin decreased lipid peroxidation
induced by N-diethylnitrosamine. These results show correlation between lipid peroxidation and hepatocarcinogenicity and
support the important role of oxidative stress on liver carcinogenesis.
q 2004 Elsevier Ireland Ltd. All rights reserved.
Keywords: Oxidative stress; Initiation; Lipid peroxidation; Preneoplastic lesions; Hepatocarcinogenesis
0304-3835/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.canlet.2004.07.019
Cancer Letters 217 (2005) 25–32
www.elsevier.com/locate/canlet
* Corresponding author. Tel.: C52-55-5061-3993; fax: C52-55-
5747-7081.
E-mail address: svilla@cell.cinvestav.mx (S. Villa-Trevin ˜o).

Page 2

1. Introduction
There are clear examples of the participation of
reactiveoxygenspecies(ROS)inhepatocarcinogenesis
in the rat. Nakae et al. [1] showed that initiation with
lowdosesofN-diethylnitrosamine(DEN)inducedliver
DNA-8-hydroxydeoxy-guanosine
suggested that oxidative stress participates in hepato-
carcinogenesis. Therefore, one can assume that
initiation with high doses of DEN produces oxidative
stress,asinthecaseofhepatocarcinogenicmodelssuch
as that by Solt and Farber, initiated with DEN and
promoted with 2-acetylaminoflourene (2-AAF) and
partial hepatectomy (PH) as proliferative stimulus [2],
or modifications of this, as in the Semple-Robert’s [3]
model. During a choline-deficient diet, the representa-
tive marker of oxidative DNA damage 8-hydroxy-
deoxy-guanosinisinduced[4]andthesameistrueafter
administration of cipofibrate, one of the more efficient
peroxisomeproliferatorsthatinduceslivercancerinthe
rat [5]. Trimethylarsine oxide, an organic metabolite of
inorganic arsenics, produced liver tumors in male
Fischer 344 rats and authors implicate a possible
mechanistic role of oxidative DNA damage and
enhanced cell proliferation [6]. Also, in male Fischer
344 rats, it was demonstrated that a,a-bis(p-chloro-
phenyl)-b,b,b trichloroethane (DDT) induces eosino-
philic foci and hepatocellular carcinoma (HCC) as a
result of oxidative DNA damage [7].
It is not clear yet if participation of oxidative stress
depends on the DNA oxygen adducts or if there is a
concomitant alteration of signalization by the abrupt
induction of ROS during initiation, or if both, DNA
damage and a new intracellular reduced steady state
are necessary for carcinogenesis to take place. Evi-
dence with U937 cells treated with two well-known
nitrosamines, 4-(methylnitrosamino)-1-(3-pyridyl)-1-
butanone (NNK) and DEN showed that ROS are
produced; these activate nuclear factor kB (NF-kB);
subsequently, cyclooxygenase-1 (COX-1) activity is
induced, and this pathway increases prostaglandin E2
(PGE2) synthesis [8]. These results, and evidence that
acetylsalicylic acid, a non-steroidal anti-inflammatory
drug, and NNK lung carcinogenesis inhibitors block
activation of NF-kB, induction of COX-1 and PGE2
synthesis, together lend support to the proposition that
ROS participate in carcinogenesis. When NNK or
DEN were substituted by their respective O-acetate
adductsand
derivatives, which do not need to be metabolized, they
did not activate NF-kB or induce PGE2synthesis,
even though it is known that these nitrosamines as
well as their acetates produce DNA adducts [8,9].
There is evidence that oxidative stress is an obli-
gatory component of carcinogenesis. It was recently
communicated that COX-1 or COX-2-deficient mice
hadalteredepidermaldifferentiationand,whentreated
with 7,12-dimethylbenz(a)anthracene (DMBA) and
12-O-tetradecanoylphorbol-12,13-acetate(TPA),they
presented reduced skin tumorigenesis, although
DMBA stable DNA adducts were increased twice
[10]. This study clearly shows that there is no
correlation between adduct levels and tumorigenesis.
Even though the authors do not demonstrate that
oxidative stress is produced, a direct relation between
COX-1 and COX-2 deficiency is hypothetically
associated to a lesser degree of oxidative stress [10].
We propose that, in the rat hepatocarcinogenesis
model, both direct alkylating DNA damage and
alterations produced by ROS induction during carci-
nogen metabolism are necessary processes for liver
cancer induction. The dilemma is how to differentiate
the participation of DNA alteration by ethyl adducts
[9] from the participation of cell modifications
induced by ROS or, even more, from the obliged
hypothetical participation of both to induce initiation.
To gain insight into this problem, experiments were
carried out to compare (a) induction of gamma-
glutamyl transpeptidase-positive (GGTC) preneo-
plastic lesions on the 25th day, as an early end point
of hepatocarcinogenesis in a DEN-2AAF-PH model
or (b) in the same model, substituting DEN by
N-ethyl-N-nitrosourea (ENU), a direct carcinogen,
that is only carcinogenic in the rat liver under very
special circumstances [11–13] or (c) by administra-
tion of quercetin, an antioxidant, administered
previous to the DEN, 2AAF treatment. And as a key
comparative feature, in these three different groups,
LPX was measured during the initiation period.
2. Methods
2.1. Reagents and animals
All reagents were purchased from SIGMA
(St Louis, MO, USA). Male Fischer-344 rats
Y. Sa ´nchez-Pe ´rez et al. / Cancer Letters 217 (2005) 25–3226

Page 3

(180–200 g) were obtained from the CINVESTAV
animal house, they had access to food (PMI Feeds,
Inc., Laboratories Diet) and water at all times; food
cups were replenished three times weekly.All animals
received humane care and the study protocols were in
compliance with the institutional guidelines for use of
laboratory animals.
2.2. Experimental protocols
Two groups of rats were initiated with a dose of
200 mg/kg of DEN and a third group with a dose of
240 mg/kg of intraperitoneal ENU. All groups
received 2-AAF orally, at doses of 20 mg/kg per
day, during 3 consecutive days before PH. PH was
performed 10 days after initiation. A third group was
administered 50 mg/kg weight intraperitoneal quer-
cetin dissolved in PBS–ethanol 10%, 1 h before DEN
administration (Fig. 1). Animals were killed 3, 6, 12,
24 h and 25 days after initiation. Livers were excised,
quickly frozen in liquid nitrogen and stored at K80 8C
until analysis.
2.3. Quantitative analysis of GGTCpreneoplastic
lesions
Representative 20 mm thick sections from liver
slices were obtained with a cryostat (Slee cryostat
MTC, Germany) and stained for GGT activity
according to Rutenburg et al. [14]. GGTCnodules
were quantitatively analyzed using images of histo-
logical sections taken with a digital camera (Color-
View 12, Soft Imaging System GmbH, Germany)
coupled to stereological microscopy, and processed
by an image software analysis kit (AnalySIS, Soft
Imaging System GmbH, Germany). Focal GGTC
areas greater than 0.05 mm2were registered to avoid
small bile-duct cell detection (arbitrary intensity
values) and were also classified according to GGT
stain intensity. The number of preneoplastic lesions
and their dimensions were scored in three sections/rat
in 14 rats from group 1, 7 rats from group 2 and 7 rats
from group 3. The data were analyzed by a two-tailed
Student’s t-test and P values of 0.05 or lower were
considered statistically significant.
2.4. Thibarbituric acid reactive substances (TBARS)
Samplesof0.8 g/mlfrozenliverwerehomogenized
in a buffer containing 10 mM Tris, 10 mM PMSF and
150 mM NaCl, pH 7.4; protein was determined
according to Smith et al. [15]. LPX was measured as
TBARS, a widely used assay, according to the method
by Buege and Aust [16]. Briefly, 600 mg of protein of
liver homogenates plus 300 ml of 0.8% of thiobarbi-
turic acid (TBA) in 20% acetic acid, pH 3.0, were
mixed and heated at 100 8C for 45 min. The samples
werecooled,addedwith200 mlof1.2%KCl,0.5 mlof
1:15 pyridine/butanol and centrifuged at 7500 rpm for
10 min. The absorbance of the resulting solution was
determined at 532 nm, and TBARS were expressed
withrespecttomalonyldialdehyde(MDA)byusingthe
MDA extinction coefficient (EZ1.56!105).
3. Results
3.1. Preneoplastic lesions
Liver preneoplastic nodules were induced with a
modified Semple-Roberts model [3], under three
Fig. 1. Experimental design. Three groups of Fischer-344 rats were
treated under three different conditions according to a modified
Semple-Roberts hepatocarcinogenesis model. Group 1 (nZ30) and
2 (nZ23) received 200 mg/kg of intraperitoneal (IP) DEN as
initiator.Group 2 receivedIP quercetin(Q) 50 mg/kg,1 h beforethe
200 mg/kg of DEN. Group 3 (nZ23) received 240 mg/kg of IP
ENU. Sixteen rats of each group were sacrificed 3, 6, 12, and 24 h
after initiation. Livers were homogenized and LPX determined.
Finally, 14 rats from group 1, 7 from group 2 and 7 from group 3
received 20 mg/kg of oral 2-AAF from day 7 to day 10, in three
daily consecutives doses, before partial hepatectomy (PH) at day10.
These rats were sacrificed at day 25.
Y. Sa ´nchez-Pe ´rez et al. / Cancer Letters 217 (2005) 25–32 27

Page 4

different conditions as shown in Fig. 1. The GGTC
lesions were measured at day 25 and our reference
group initiated by DEN clearly presented abundant
and high-intensity GGTCstained preneoplastic
lesions while the other two groups of either ENU as
initiator or quercetin plus DEN presented a few high-
intensity GGTCstained preneoplastic lesions. How-
ever, all groups showed diffuse low-intensity GGTC
stained areas in the liver (Fig. 2). A graphic
representation of the area and amount of GGTC
lesions is shown in Fig. 3. In group 1, high-
intensity GGTCstained nodules occupied a total
area of 5G3.3%, and the number of nodules was
31G6.4/cm2, while the group initiated with ENU
presented a lesion-occupied area of only 0.6% and
2.6G1.4 nodules/cm2and in the group with quercetin
plus DEN, these values were 1.1G0.66% in area and
16G0.44 for nodules/cm2.
Fig. 2. Histochemical determination of GGTClesions. Thick histological sections (20 mm) obtained from liver slices on the 25th day of
treatment initiation were stained for GGT activity. Representative sections of each initiation protocol are shown. (A) Rat liver from group 1
initiated with DEN, (B) rat liver from group 2 which received 50 mg/kg IP quercetin 1 h before administration of DEN and (C) rat liver from
group 3 initiated with ENU. Amplified images from C show high-intensity GGTClesions (D) and low intensity GGTClesions (E).
Y. Sa ´nchez-Pe ´rez et al. / Cancer Letters 217 (2005) 25–3228

Page 5

3.2. Lipid peroxidation
Since it has been shown that LPX is an expression
of oxidative stress in cells [17], we determined
TBARS as a global approach and compared the levels
obtained in a 24-h period after initiation, with the
levels obtained during formation of preneoplastic
lesions quantified 25 days after initiation. Three hours
after initiation with DEN, an increment of LPX of 3.8-
fold above control level was observed. The highest
level, 4.62-fold, was obtained 12 h after DEN
initiation, but all LPX determinations in this group
at 3, 6, 12 and 24 h after DEN initiation showed
statistically significant increments with respect to the
vehicle control group (Fig. 4). Importantly, adminis-
tration of quercetin, 1 h before DEN, prevented full
expression of LPX at both, 3 and 6 h; represented only
15% and at 12 and 24 h, 67% and 66%, respectively,
these values with respect to a similar group that did
not receive quercetin. LPX levels at 3, 6, 12 and 24 h
after initiation with ENU, instead of DEN, were very
low compared with those obtained in non-treated
animals and differences with respect to the DEN
group were highly significant. The lowest of LPX
levels induced with ENU, correlated with the
induction of few preneoplastic lesions.
4. Discussion
To gain insight intooxidative stress participation in
the induction of preneoplastic lesions, we analyzed
Fig. 3. Quantitative analysis of GGTClesions in the initiation protocols. Focal GGTCareas larger than 0.05 mm2were registered and classified
according to their low or high intensity by image analysis. The number of lesions and their dimensions were scored in three sections/rat. Group
1, liver section of rats initiated with DEN (nZ14); group 2, liver of rats which received 50 mg/kg quercetin 1 h after being initiated with DEN
(nZ7); group 3, liver sections of rats initiated with ENU (nZ7). Mean valuesGSD. Statistical differences between groups are indicated by *.
Fig. 4. Determination of lipoperoxidation after initiation with three
different protocols. Malonyldialdehyde concentration as a result of
LPX was determined in liver homogenates at 3, 6, 12 and 24 h and
25 days, after administration of initiator carcinogen (Fig. 1).
Additionally, liver homogenates of rats treated with corresponding
vehicles were used as controls. Water as DEN vehicle (group 1);
10% ethanol in PBS as quercetin vehicle (group 2); sodium citrate
as ENU vehicle (group 3). Mean valuesGSD. J P values %0.03
compared with their control vehicle. *P values %0.03 and
**pv%0.001 compared with group 1 at their respective times.
Y. Sa ´nchez-Pe ´rez et al. / Cancer Letters 217 (2005) 25–3229

Page 6

the effects during cancer initiation with ENU, a direct
and seldom rat liver carcinogen [18] and with DEN an
indirect and very efficient rat liver carcinogen,
followed by 2AAF administration as a promoter and
PH as proliferative stimulus. We chose to evaluate
GGTCnodule formation at day 25 after initiation and
LPX during 24 h after initiation. We show that DEN, a
ROS-generating carcinogen [1] induces many pre-
neoplastic lesions while ENU, a non-ROS-generating
carcinogen, as it can be extrapolated from our result
presented in Fig. 4, induces less preneoplastic lesions
F. Also, administration of quercetin, an antioxidant
[19], counteracted the oxidative stress induced by
DEN and, as corollary of our results, the size and
number of preneoplastic lesions closely correlated
with the appearance of LPX.
Even though ENU has been shown to be a potent
mono functional-ethylating and mutagenic agent in a
variety of systems [20], its hepatocarcinogenic effect
in rats seems to require previous PH [11,12] or to be
followed with phenobarbital treatment [13]. In this
respect, even intraportal administration to rats of
30 mgofENU didnotproduce liver tumors [18].ENU
is used as a progressor in the initiation–promotion–
progression model, late PH is performed and rats are
sacrificed 6 months later, after which 12% incidence
of HCC occurs in the PH rats and this incidence rises
to 89% when PH is followed by PB treatment [13]. It
should be noted that in both, PH [21,22] and PB [23]
treatment, ROS are generated and PH strongly
increases the mutagenicity or carcinogenicity induced
by ENU [11,12,24]. These results agree with our
proposal of the participative role of oxidative stress
during hepatocarcinogenesis. In another context,
Sakai et al. [25] investigated 26 chemicals in a
model for detection of initiation activity. Among them
they assayed N-methyl-N-nitrosourea (MNU), a
carcinogen closely related to ENU, and showed that
nitrosourea has activity as initiator. But in the used
model, MNU was administered 24 h after hepatect-
omy, the period in which LPX increases to a
maximum [22]. This evidence is in agreement with
the hypothesis that the oxidative stress that occurs
immediately during liver regeneration may contribute
to initiator activity of MNU and by extension of other
direct mutagens such as ENU [11–13].
In rats initiated with ENU, we can assume that liver
DNA alkylation takes place, as has been shown by
Frei et al. [26]. Nevertheless, in the hepatocarcino-
genic model used herein, PH is performed 10 days
after initiation, thus, ENU is not under the effect of the
oxidative stress of HP. It may then be suggested that
ENU alkyl-DNA lesions alone are not sufficient to
induce preneoplastic lesions in rat liver. A similar
conclusion can be reached in a multistage mouse skin
model; in COX-1 or COX-2-deficient mice, which are
hypothetically under a lesser oxidative stress, treat-
ment with DMBA produced only 25% skin tumors,
compared with wild-type mice, while stable DMBA–
DNA adducts in COX-deficient mice increased
two-fold, indicating again that the amount of
DMBA–DNA adducts did not correlate with tumor
number [10].
It has been accepted that the correlation between
lipid peroxidation and appearance of preneoplastic
lesions is indicative of ROS participation in carcino-
genesis. It is tempting to speculate that alterations in
Redox cell signaling produce a change that, synergi-
cally with DNA alterations, induces deregulation of
cell control and hence cell proliferation. Such
alterations have been shown in U937 cells treated
with the carcinogens NNK and NDMA, which are
metabolized by P450 cytochrome enzymes and
produce oxidative stress, subsequently activate
cyclooxygenases 1 and 2, leading to increase of
PGE2 and activation of NF-kB. In contrast, the
O-acetates of the nitrosamines that do not require
metabolism do not induce oxidative stress, or activate
NF-kB or induce PGE2[8]. A possible mechanism of
quercetin to reduce LPX levels and prevent preneo-
plastic lesion induction could be an inhibition of DEN
metabolism capacity. Inhibition of carcinogen meta-
bolism by quercetin has been previously reported
[27]. Additionally, quercetin exhibits an antioxidant
and free radical scavenging activities against
H2O2[28].
In summary, we presently show that liver LPX
levels in rats initiated with DEN versus those initiated
with ENU were several fold higher and differences
were highly significant. This finding correlates with
the virtual absence of liver preneoplastic lesions after
initiation with ENU. Quercetin in our system blocks
LPX and prevents preneoplastic lesion induction
suggesting that the antioxidant effect participated in
these phenomena. Our results support the proposal of
ROS participation in the initiation of carcinogenesis.
Y. Sa ´nchez-Pe ´rez et al. / Cancer Letters 217 (2005) 25–3230

Page 7

The close correlation between the induction of liver
LPX and the presence of preneoplastic lesions
induced in the three different conditions studied with
this hepatocarcinogenic model adds to the existing
evidence of ROS participation in tumor induction and
opens the possibility offurther studies with this model
to shed light on the mechanism of ROS participation
in rat hepatocarcinogenesis.
Acknowledgements
We would like to thank Samia Fattel and Evelia
Arce Popoca for technical assistance during this
project. We also wish to acknowledge the excellent
technical support of Jorge Fernandez, head of the
Animal House and Manuel Flores, Rafael Leyva and
Ricardo Gaxiola. We thank Isabel Pe ´rez Montfort for
revising the English version of the manuscript. This
work was supported by grants 31665-N and 34547-M
from CONACyT, Me ´xico, DF, Mexico. Fellowship
from CONACyT: YSP;144244, CECL;112857,
JIPC;144549, MESN;119303.
References
[1] D. Nakae, Y. Kobayashi, H. Akai, N. Andoh, H. Satoh,
K. Ohashi, et al., 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.
[2] D.B. Solt, E. Farber, New principle for the analysis of
chemical carcinogenesis, Nature (London) 26 (1976)
702–703.
[3] E. Semple-Roberts, M.A. Hayes, D. Armstrong, R.A. Becker,
W.J. Racz, E. Farber, Alternative methods of selecting rat
hepatocellular nodules resistant to 2-acetylaminofluorene, Int.
J. Cancer 40 (1987) 643–645.
[4] R.A. Floyd, Y. Kotake, K. Hensley, Y. Konishi, Reactive
oxygen species in choline deficiency induced carcinogenesis
and nitrone inhibition, Mol. Cell. Biochem. 234/235 (2002)
195–203.
[5] C-Y. Huang, M.W. Wilson, L.T. Lay, C.K. Chow,
L.W. Robertson, H.P. Glauert, Increased 8-hydroxydeoxy-
guanosine in hepatic DNA of rats treated with the peroxisome
proliferators ciprofibrate and perfluorodecanoic acid, Cancer
Lett. 87 (1994) 223–228.
[6] J. Shen, H. Wanibuchi, E.I. Salim, M. Wei, A. Kinoshita,
K. Yoshida, et al., Liver tumorigenicity of trimethylarine
oxide in male Fischer 344 rats-association with oxidative
DNA damage and enhanced cell proliferation, Carcinogenesis
24 (2003) 1827–1835.
[7] T. Harada, S. Yamaguchi, R. Ohtsuka, M. Takeda,
H. Fujisawa, K. Yoshida, et al., Mechanisms of promotion
and progression of preneoplasic lesions in hepatocarcinogen-
esis by DDT in F344 rats, Toxicol. Pathol. 31 (2003) 87–98.
[8] N. Rioux, A. Castonguay, The induction of cyclooxygenase-1
by tobacco carcinogen in U937 human macrophages is
correlated to the activation of NF-kappaB, Carcinogenesis
21 (2000) 1745–1751.
[9] L. Verna, J. Whysner, G.M. Williams, N-Nitrosodiethylamine
mechanistic data and risk assessment: bioactivation, DNA-
adduct formation, mutagenicity, and tumor initiation, Phar-
macol. Ther. 71 (1996) 57–81.
[10] H.F. Tiano, C.D. Loftin, J. Akunda, C.A. Lee, J. Spalding,
A.Sessoms,etal.,Deficiencyofeithercyclooxygenase(COX)-
1 orCOX-2alters epidermaldifferentiation andreducesmouse
skin tumorigenesis, Cancer Res. 62 (2002) 3395–3401.
[11] V.M. Craddock, J.V. Frei, Induction of liver cell adenomata in
the rat by a single treatment with N-ethyl-N-nitrosourea given
at various times after partial hepatectomy, Br. J. Cancer 30
(1974) 503–511.
[12] D.B. Solt, E. Cayama, D.S. Sarma, E. Farber, Persistent of
resistant putative preneoplastic hepatocytes induced by N-
nitrosodiethylamine or N-ethyl-N-nitrosourea, Cancer Res. 40
(1980) 1112–1118.
[13] Y.P. Dragan, L. Sargent, Y.-D. Xu, Y.-H. Xu, H.C. Pitot, The
initiation–promotion–progression model of rat hepatocarcino-
genesis, Proc. Soc. Exp. Biol. Med. 202 (1993) 16–24.
[14] A.M. Rutenburg, H. Kim, J.W. Fishbein, J.S. Hanker,
H.L. Wasserkrug, A.M. Seligman, Histochemical and ultra-
structural demonstration of g-glutamyl transpeptidase
activity, J. Histochem. Cytochem. 17 (1969) 517–526.
[15] P.K. Smith, R.I. Krohn, G.T. Hermanson, A.K. Mallia,
F.H. Gartner,M.D.Provenzano,etal.,Measurementofprotein
using bicinchoninic acid, Anal. Biochem. 150 (1985) 76–85.
[16] J.A. Buege, S.D. Aust, Microsomal lipid peroxidation,
Methods Enzymol. 52 (1978) 302–310.
[17] R. Kohen, A. Nyska, Oxidation of biological systems:
oxidative stress phenomena, antioxidants, redox reactions,
and methods for their quantification, Toxicol. Pathol. 30
(2002) 620–650.
[18] R. Hasegawa, M. Futakuchi, Y. Mizoguchi, T. Yamaguchi,
T. Shirai, N. Ito, W. Lijinsky, Studies of initiation and
promotion of carcinogenesis by N-nitroso compounds, Cancer
Lett. 123 (1998) 185–191.
[19] A. Kahraman, N. Erkasap, T. Ko ¨ken, M. Serteser, F. Aktepe,
S. Erkasap, The antioxidative and antihistaminic properties of
quercetin in ethanol-induced gastric lesions, Toxicology 183
(2003) 133–142.
[20] T. Shibuya, K. Morimoto, A review of the genotoxicity of 1-
ethyl-1-nitrosourea, Mutat. Res. 297 (1993) 3–38.
[21] F.M. Guerrieri, G. Vendemiale, I. Grattagliano, T. Cocco,
G. Pellecchia, E. Altamore, Mitochondrial oxidative altera-
tions followingpartial hepatectomy,Free Radic. Biol. Med. 26
(1999) 34–41.
[22] L. Lambotte, Y. d’Udekem, F. Lambert, P. Gressens,
M. Amrani, J. Dubois, Possible role of free radicals in liver
regeneration, Adv. Biosci. 79 (1989) 177–185.
Y. Sa ´nchez-Pe ´rez et al. / Cancer Letters 217 (2005) 25–32 31

Page 8

[23] A. Kinoshita, H. Wanibuchi, S. Imaoka, M. Ogawa,
C. Masuda, K. Morimura, et al., Formation of 8-hydroxy-
deoxyguanosine and cell-cycle arrest in the rat liver via
generation of oxidative stress by Phenobarbital: association
with expression profiles of p21WAF/Cip1, cyclin D1 and Ogg1,
Carcinogenesis 23 (2002) 341–349.
[24] T. Hara, H. Sui, K. Kawakami, Y. Shimada, T. Shibuya,
Partial hepatectomy strongly increased the mutagenicity of
N-ethyl-N-nitrosourea in MutaMouse liver, Environ. Mol.
Mutagen. 34 (1999) 121–123.
[25] H. Sakai, T. Tsukamoto, M. Yamamoto, K. Kobayashi,
H. Yuasa, T. Imai, et al., Distinction of carcinogens from
mutagens by induction of liver cell foci in a model for
detection of initiation activity, Cancer Lett. 188 (2002) 33–38.
[26] J.V. Frei, D.H. Swenson, W. Warren, P.D. Lawley, Alkylation
of deoxyribonucleic acid in vivo in various organs of C57BL
mice by the carcinogens N-methyl-N-nitrosourea, N-ethyl-N-
nitrosourea and ethyl methanesulphonate in relation to induc-
tion of thymic lymphoma. Some applications of high-pressure
liquid chromatography, Biochem. J. 174 (1978) 1031–1044.
[27] C.A. Musonda, H. Helsby, J.K. Chipman, Effects of quercetin
on drug metabolizing enzymes and oxidation of 2,7-dichloro-
fluorescein in HepG2cells, Hum. Exp. Toxicol. 16 (1997)
700–708.
[28] S.A. Aherne, N.M. O’Brien, Protection by the flavonoids
myricetin, quercetin, and rutin agaist hydrogen peroxide-
induced DNA damage in Caco and HepG2cells, Nutr. Cancer
34 (1999) 160–166.
Y. Sa ´nchez-Pe ´rez et al. / Cancer Letters 217 (2005) 25–32 32