ABERRANT REPAIR OF ETHENO DNA ADDUCTS IN
LEUKOCYTES AND COLON TISSUE OF COLON CANCER PATIENTS
Tomasz Obtułowicz a, Alicja Winczura a, Elżbieta Speina a, Maja Swoboda a, Justyna Janik a,
Beata Janowska a, Jarosław M. Cieśla a, Paweł Kowalczyk b, Arkadiusz Jawien c, Daniel
Gackowski d, Zbigniew Banaszkiewicz c, Ireneusz Krasnodebski e, Andrzej Chaber e, Ryszard
Olinski d, Jagadesaan Nair f, Helmut Bartsch f, Thierry Douki g, Jean Cadet g, Barbara Tudek
a,h,*
a
b
c
d
e
f
g
h
Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warsaw, Poland
Interdisciplinary Centre for Mathematical and Computational Modeling, Warsaw University,
Warsaw, Poland
Department of Surgery, Collegium Medicum, Nicolaus Copernicus University, Bydgoszcz,
Poland
Department of Clinical Biochemistry, Collegium Medicum, Nicolaus Copernicus
University, Bydgoszcz, Poland,
Public Central Teaching Hospital, Medical University of Warsaw, Warsaw, Poland
German Cancer Research Center hi, Heidelberg, Germany
Laboratoire Lésions des Acides Nucléiques, Service de Chimie Inorganique et Biologique
UMR-E 3 CEA-UJF, CNRS FRE 3200, CEA/DSM/INAC, CEA-Grenoble, France.
Institute of Genetics and Biotechnology, Warsaw University, Warsaw, Poland
JN has deceased
*
Corresponding author: Barbara Tudek,
Institute of Biochemistry and Biophysics,
Polish Academy of Sciences,
Pawinskiego 5a,
02-106 Warsaw, Poland
tel: (+4822) 592-3334; fax: (+4822) 592-2190
e-mail: tudek@ibb.waw.pl
Running title: Repair of etheno DNA adducts in colorectal cancer patients
1
ABSTRACT
To assess the role of lipid peroxidation (LPO)-induced DNA damage and repair in colon
carcinogenesis, excision rate and the level of 1,N6-etheno-2’-deoxyadenosine (εdA), 3,N4etheno-2’-deoxycytidine (εdC) and 1,N2-etheno-2’-deoxyguanosine (1,N2-εdG) were analyzed
in polymorphic blood leukocytes (PBL) and resected colon tissues of 54 colorectal carcinoma
(CRC) patients and PBL of 56 healthy individuals. In PBL the excision rate of 1,N6ethenoadenine (εAde) and 3,N4-ethenocytosine (εCyt), measured by the nicking of
oligodeoxynucleotide duplexes with single lesions, and unexpectedly also the level of εdA and
1,N2-εdG, measured by LC/MS/MS, were lower in CRC patients than in controls. In contrast
mRNA levels of repair enzymes, alkylpurine- (ANPG) and thymine-DNA glycosylases (TDG)
and abasic site endonuclease (APE1) were higher in PBL of CRC patients than of controls, as
measured by QPCR. In the target colon tissues εAde and εCyt excision rate was higher, while
εdA and εdC level in DNA, measured by 32P-postlabeling method, was lower in tumor than in
adjacent colon tissue, although the higher mRNA level was observed only for APE1. This
suggests that during onset of carcinogenesis, etheno adducts repair in the colon seems to be
under complex transcriptional and post-transcriptional control, whereby deregulation may act
as a driving force to malignancy.
Keywords: Oxidative Stress; Colon Cancer; Base Excision Repair; 1,N6-Ethenoadenine; 3,N4Ethenocytosine
2
Colorectal cancer is a frequent cause of death in developed countries [1, 2]. Important
etiological risk factors of sporadic colorectal tumors are inflammatory bowel diseases, excess
consumption of animal fat and meat, alcohol drinking, and tobacco smoking [3]. Chronic
inflammation is associated with the large release of reactive oxygen and nitrogen species [4]
leading to oxidation of nucleic acids, proteins and lipids. Oxidation of polyunsaturated fatty
acids generates enals with 3 to 10 carbon atoms and an unsaturated double bond in α, βposition, such as 4-hydroxynonenal (HNE), acrolein and croton aldehyde [5]. These either
react directly with DNA [6] or, as in the case of HNE, undergo further oxidation to more
reactive 2,3-epoxy-4-hydroxynonanal [7]. Reaction of LPO products in vivo with DNA yields
inter alia the exocyclic etheno DNA adducts, 1,N6-etheno-2’-deoxyadenosine (dA),
3,N4-etheno-2’-deoxycytidine
(dC),
1,N2-etheno-2’-deoxyguanosine
and
(1,N2-dG).
Epoxidation of enals occurs in the presence of hydrogen peroxide, but is also catalyzed by
enzymes producing inflammatory mediators, such as cyclooxygenase-2 and lipooxygenases
[8], which are increased during colon cancer development [9]. Lipid peroxidation (LPO)derived DNA damage, manifested by increased levels of dA and dC, was found in nontumorous (premalignant) colon of CRC-prone subjects, affected by familial adenomatous
polyposis and inflammatory bowel diseases [10]. LPO-derived DNA adducts in the
mammalian cells induce point mutations, chromosomal aberrations and recombination [11],
implicating LPO-derived DNA damage in the initiation and progression of colon tumor
development.
The ability to eliminate pro-mutagenic modified nucleotides or mismatched bases from
DNA or polymorphisms in repair genes affect individual cancer susceptibility [12]. During
sporadic colon carcinogenesis, malfunctioning of mismatch repair system (MMR) was
3
observed in about 15% of cases, among which 5% were due to new mutations, and 10% to
silencing of the MLH1 gene [13]. Also polymorphism of the MYH gene, engaged in excision
of adenine mismatched with 8-oxo-7,8-dihydroguanine (8-oxoGua), increases the risk of
colorectal tumors [12]. Several mechanisms appear to be engaged in the removal of specific
exocyclic DNA adducts. These include base excision repair, nucleotide excision repair,
mismatch repair, abasic sites (AP) endonuclease-mediated nucleotide incision repair as well as
oxidative demethylation, catalyzed by AlkB-family proteins [14, 15]. The major pathway of
etheno adducts elimination from DNA seems to be base excision repair (BER) with DNA
glycosylases as the key enzymes of this pathway. 1,N6-Ethenoadenine (Ade) is excised by
alkylpurine DNA N-glycosylase (ANPG) [16], whereas 3,N4-ethenocytosine (Cyt) by
thymine DNA glycosylase (TDG); the latter one also excises thymine from dG:dT pairs,
resulting from deamination of 5-methylcytosine [17]. Both enzymes are monofunctional DNA
glycosylases, and require AP endonuclease to incise DNA at the site of the removed base to
continue the BER process. 3,N4-Ethenocytosine is also excised by MBD4 glycosylase
although its activity is strongly limited to CpG islands [18, 19].
ANPG glycosylase has a wide substrate specificity and besides Ade excises from
DNA
also
1,N2-ethenoguanine
[20],
hypoxanthine
and
the
alkylated
bases,
3-methyladenine and 7-methylguanine [16]. TDG excises Cyt, thymine (Thy) and uracil
(Ura) from pairs with guanine, and its activity is strongly stimulated by the next enzyme in the
BER pathway, AP endonuclease, APE1 [21]. TDG and APE1 also play the role of
transcription regulators [22-25].
The aims of our case-control study were to measure and compare: (i) the repair
capacity of enzymes involved in the removal of etheno DNA adducts in polymorphic blood
4
leukocytes (PBL) of colon cancer patients vs. healthy controls and in colon tissues of CRC
patients; (ii) the levels of LPO-derived etheno DNA adducts in PBL of patients vs. healthy
controls and in tumorous vs. surrounding normal colon specimens of CRC patients; (iii) to
compare whether these disease-related variables in surrogate (PBL) vs. target tissue (colon)
are correlated.
Materials and methods
Study subjects
This case-control study was conducted in two groups; first group consisted of 54 CRC
patients and the second one comprised 56 healthy individuals (see Table 1 for characteristics
of the studied groups). All participants were Caucasians and there were no relatives among
them. All individuals participating in the study were recruited through the hospital (Collegium
Medicum, Nicolaus Copernicus University, Bydgoszcz, Poland) and were examined by
colonoscopy. The control group was recruited from individuals undergoing routine colon
cancer screening, in whom colonoscopy revealed no cancer or polyps. All the subjects, when
recruited to the study, filled in a questionnaire concerning demographic data, smoking, diet
and medical history. Interviewees were asked to estimate the average frequency of
consumption of various dietary items in the year preceding the interview. The majority of
them reportedly consumed 3 servings of fruits and vegetables and about 250 g of meat and fat
per day. To make the group even more homogenous, the subjects who reported the extreme
consumption, as well as those who reported supplementation within the last month were
excluded from the study. The control group was chosen to maximally match the patient group
by age, sex, diet (consumption of fat, carbohydrates and vitamin intake), body weight, and
smoking status. Between healthy volunteers and CRC patients, two groups were distinguished
5
in relation to their smoking status, namely those who had never smoked and smokers who
consumed 20 or more cigarettes per day.
Histological categorization of tumor was based on the WHO classification. The
samples were coded before the analysis and information concerning age, sex, smoking habit
and type of tumor became available only before statistical analysis. The patients were not
treated with any anticancer drugs or vitamins during the time from the diagnoses until surgery
(up to 4 weeks).
The study was conducted in accordance with the Helsinki Declaration and was
approved by the medical ethics committee of Collegium Medicum, Nicolaus Copernicus
University, Bydgoszcz, Poland (in accordance with Good Clinical Practice, Warsaw 1998).
All participants signed informed consent.
Isolation of leukocytes from venous blood
Blood samples were withdrawn from patients and controls in the morning before
breakfast in Clinical Units of Collegium Medicum, Nicolaus Copernicus University in
Bydgoszcz. Blood samples (18 ml) were carefully applied on top of Histopaque 1119 solution
(Sigma-Aldrich Inc.; St.Louis, MO, USA) and leukocytes were isolated by centrifugation
according to the manufacturer’s procedure.
Preparation of tissue extracts
Blood leukocytes and colon tissues (tumor and histologically unchanged colon; for
simplicity the latter one will be called normal colon) were homogenized with 4 volumes of 50
mM Tris-HCl, pH 7.5 buffer containing 1 mM EDTA and proteases inhibitor cocktail (Sigma).
Cells were disrupted by sonication (three 15 s pulses and 30 s intervals). The cell debris was
6
removed by centrifugation (7000xg, 4°C, 15 min), and the supernatant was collected. Protein
concentration was determined by the Bradford method [26] using the protein assay reagent
(Sigma). Supernatants were stored in aliquots at -80°C for further analysis.
Ade and Cyt excision activity assay
Etheno adducts excision activity was measured by the nicking assay using an
oligodeoxynucleotide duplex containing a single Ade or Cyt residue [27]. The assay
monitors
excision
of
Ade
or
Cyt
from
a
5’-radiolabeled
synthetic
DNA
oligodeoxynucleotide by DNA glycosylases contained in tissue extract, and incision of abasic
site by AP endonuclease. Briefly, the oligodeoxynucleotides (40 nt) containing a single
modified base at position 20 in the sequence 5’-d(GCT ACC TAC CTA GCG ACC TXC
GAC TGT CCC ACT GCT CGA A)-3’ (where X indicates either Ade or Cyt; Eurogentec
Herstal, Belgium) were
P-labeled at the 5’-end with T4 polynucleotide kinase and [-
32
32
P]ATP (GE Healthcare). The labeled oligomers were annealed to complementary
oligonucleotides in a double molar excess. They contained T opposite dA and dG opposite
dC.
The reaction mixtures (20 µl) contained 1 pmol of
32
P-labeled duplex and increasing
concentrations of tissue extract (1-100 µg) in 25 mM Tris-HCl (pH 7.8), 50 mM NaCl, 5 mM
-mercaptoethanol and 1 mM EDTA. Samples were incubated at 37°C for 1 h and
subsequently digested with proteinase K (20 µg, 1 h, 37°C). Reactions were terminated with
the addition of 10 μl of formamide stop solution (90% formamide, 20 mM EDTA, 0.1%
bromophenol blue and 0.1% xylene cyanol), and heated to 95°C for 5 min. Products were
resolved on 20% polyacrylamide, 7 M urea denaturing gels. A Storm phosphorImager was
7
used for detection and ImageQuant software (GE Health Sciences) was used for quantification
of the reaction products. From each data set, a curve reflecting excision rate as a function of
the amount of extract protein was plotted and the enzyme activities were calculated from the
linear part of the curve. In each experiment, two control samples were used: a) a negative
control in which untreated oligonucleotide was subjected to denaturing PAGE to show
possible degradation that could appear during the procedure; and (b) a positive control in
which labeled oligonucleotide was digested with an excess of the appropriate pure DNA
glycosylase (a kind gift from Dr. Murat Saparbaev and Dr. Jacques Laval, Institut Gustave
Roussy, Villejuif, France).
DNA isolation
For genotyping, DNA was extracted from frozen colon tissues (both normal colon and
tumor) and blood leukocyte samples using Genomic Mini Kit (A&A Biotechnology, Gdansk,
Poland).
For etheno DNA adducts determination, the colon tissue or leukocyte pellet was
homogenized or dispersed by vortexing, respectively, in ice-cold buffer B (10 mM Tris, 5 mM
Na2EDTA, 0.15 mM deferoxamine mesylate, pH 8.0). A solution of SDS was added (to the
final concentration of 0.5%), and vortexing was repeated; RNase in 10 mM Tris pH 8.0 was
added, and the mixture was gently vortexed. After incubation for 30 min at 37°C, the protease
was added; the mixture was gently vortexed and incubated at 37°C for 1 h. The mixture was
cooled to 4°C and transferred to a centrifuge tube containing chloroform/3-methyl-1-butanol
and vortexed vigorously. After centrifugation, supernatant containing DNA was treated with 2
volumes of cold absolute ethanol in order to precipitate high molecular weight DNA. The
8
precipitate was removed with a plastic spatula, washed with 70% ethanol and after
centrifugation dissolved in nuclease P1 buffer (40 mM sodium acetate, 0.1 mM ZnCl2, pH
5.1).
The quality and quantity of the DNA was measured spectrophotometrically with
A260/A280 ratio ranging from 1.8 to 2.0. For DNA samples used to measure etheno DNA
adducts, the whole spectrum between 200 and 350 nm was used to measure the DNA
concentration. The DNA samples were kept at -20°C until tested.
TDG and ANPG genotyping by the Multitemperature Single Strand Conformation (MSSCP)
method
Aliquots of 4 l of template DNA (~100 ng) were added to a PCR mix containing 1 U
Taq DNA Polymerase, 10 pmols of each primer, 5 l 10 x Taq buffer, 10 mM dNTPs, 1.5 mM
MgCl2 and water to a final volume of 50 l. All exons of ANPG and TDG genes were
amplified using intron based primers and cycling conditions as described in ref. [28] for
ANPG and in ref. [29] for TDG. Cycling conditions for TDG were: 94C for 3 min; 35 cycles
of 94C for 20 s, 55C (for exons: 1, 2, 6, 11) or 50C (for exons: 3, 4, 5, 7, 8, 9, 10) for 30 s,
72C for 30 s; extension step at 72C for 5 min. Water was used as a negative template control
for each PCR reaction. PCR products were visualized on agarose gels, followed by ethidium
bromide staining. Subsequently, samples were analyzed using MSSCP method [30]. PCR
products (4 l) were mixed with 4 l of denaturing buffer A (BioVectis, Poland), heated at
55C for 10 min, quenched on ice and then separated by non-denaturating 10-12%
polyacrylamide gel electrophoresis. Electrophoresis was performed in 0.5xTBE buffer in DNA
Pointer System (BioVectis) at three temperatures: 35-15-5C for 30 min each at 40 W. Band
9
patterns were visualized with a Silver Staining Kit (BioVectis) according to the
manufacturer’s instructions. Samples which showed altered migration were sequenced in the
DNA Sequencing and Oligonucleotides Synthesis Laboratory, IBB PAS, Warsaw, Poland.
Etheno DNA adducts determination in DNA isolates
Colon tumor tissue and histologically normal parts of the colon from the same resected
material were analyzed for -adducts, dA (n = 14) and dC (n = 12), by immuno-enriched
32
P-postlabeling method [31]. It has a detection limit of ca 5 adducts/1010 parent nucleotides
with a coefficient of variation of 15-20%. In PBL-DNA samples of CRC patients (n = 18) and
of healthy individuals (n = 23) dA and 1,N2-dG were measured by LC/MS/MS [32]. Results
are expressed as adducts per 108 parent nucleotides.
RNA extraction and cDNA synthesis
Total RNA was isolated from frozen blood leukocytes and colon tissues using the
TRIzol reagent [33] from Invitrogen (Carlsbad, CA, USA), according to the manufacturer’s
recommendations. The concentration of RNA samples was ascertained by measuring optical
density at 260 nm.
Total RNA (1 g as starting material) from each sample was used to generate cDNA
using the Advantage RT-for-PCR cDNA synthesis kit (Clontech; Mountain View, CA, USA)
with oligo (dT) primers. cDNA purity was verified by amplifying the GAPDH region between
exon 3 and 4 (Table 2). For further analysis, samples in which only one PCR product was
detected were used. 18S rRNA was used as a reference gene. As 18S rRNA is polyadenylated
during degradation [34, 35], it was also amplified using oligo (dT) primer. Since the level of
10
18S rRNA is a quite stable marker, the rates of its synthesis and degradation should also be
relatively stable.
Real-time PCR using SYBR-Green chemistry
Real-time PCR assays were carried out on the Applied Biosystems 7500 apparatus
(Foster City, CA, USA). Each measurement was carried out in 25 μl reaction mixture
containing: 1 x concentrated commercial buffer (without MgCl2) supplied with Taq
polymerase (Invitrogen), 3 mM MgCl2, 0.01% Tween 20, 0.8% glycerol, 5% DMSO, 0.5 ng/μl
acetylated BSA, dATP, dCTP, dGTP and dTTP - 400 μM each, 1 x concentrated reference dye
ROX, 1:40000 diluted SYBR Green dye, 0.625 U of Taq polymerase, forward and reverse
primers – 400 μM each, and an appropriate amount of template cDNA. Time-temperature
program was as follows: 95ºC for 3 min as the initial denaturation step, followed by 45 cycles
consisting of denaturation step at 95ºC for 30 s, primer annealing at 60°C for 30 s and
extension step at 72°C for 1 min. Fluorescence was read during the extension step of each
cycle. Melting-point temperature analysis was performed in the range of 60-92°C with
temperature increments of 0.33°C. Background range and threshold for Ct evaluation in each
experiment were adjusted manually.
All primers were designed using Primer Express program (Applied Biosystems). The
product from each pair was 131-132 bp long. Table 2 contains primer sequences and annealing
temperatures. Before use, the primers were verified for equal efficiency of the PCR reaction.
To ensure that, calculation the 2-ΔΔCt method was applied [36], each experiment involved
measurement of ct values for four or five amounts of the template, each in duplicate. The
template amounts per sample were as follows: for 18S rRNA – 10, 20, 40, 80 and 160 ng, for
APE1 – 40, 80, 160, 320 and 640 ng, for ANPG and TDG – 80, 160, 320, 640 and 1280 ng.
11
During measurement of mRNA expression, four reactions were carried out for each cDNA
sample, using two template amounts, 10 and 40 ng, each in duplicate. The quality of results
was evaluated on the basis of expected ct differences between two cDNA amounts as well as
product melting curves. Rare outlying results were omitted in the calculations. For each gene
the amounts of cDNA used were chosen individually (if possible the same for all genes) to
obtain ct values in range between 14 and 34 cycles.
All ct values were normalized to higher amount of cDNA used for 18S rRNA reference
gene (40 ng), and then standard ΔΔct method was applied. The results were calculated with
normalization of Ct values to mean Ct value for the 18S rRNA reference gene as described
[37]. This procedure ensured that the same value for 18S rRNA gene was used in the
calculation of each target gene expression.
Statistical analysis
All variables were examined for normality and homogeneity of variance. Data are
presented as means and standard deviations. Differences in the enzyme activities, the levels of
mRNA, and DNA adducts between leukocytes of CRC patients and healthy controls or
between normal lung and lung tumor tissues of CRC patients were tested for statistical
significance using either Mann-Whitney U-test or Wilcoxon rank sum test.
To determine whether tobacco smoking, sex, age, and cancer stage was associated with
enzyme activities and mRNA levels in leukocytes of patients and controls, or in normal lung
and in tumor, data were analyzed with Kruskal-Wallis or Friedman ANOVA. Associations
between different variables within the whole patient and control population were calculated
using Spearman's correlation analysis. All statistical analyses were performed using
12
STATISTICA 6.0 (StatSoft, Inc., Tulsa, OK). All statistical tests were two-sided, and p values
less than 0.05 were considered statistically significant.
Results
Ade and Cyt excision activity in PBL of CRC patients and healthy controls
We have previously found that PBL of lung cancer patients have lower repair activity
for Ade than healthy controls, and the development of specific inflammation-related lung
adenocarcinoma is also associated with the lower Cyt repair [27]. Since one of colon cancer
risk factors is inflammation-related oxidative stress, we wanted to evaluate a role of LPOderived DNA-damage repair in the pathology of colon cancer. We determined the Ade and
Cyt excision activity in PBL (as a surrogate for target tissue) and compared patients with
healthy controls, using oligodeoxynucleotides with a single modified Ade or Cyt base (Fig.
1). CRC patients had significantly lower Ade and Cyt excision activities than controls (mean
values ± SD, pmols/h/mg protein, are for Ade 17.00 ± 5.35 in patients vs. 33.50 ± 12.50 in
controls, p = 0.00004; for Cyt 2.53 ± 0.75 in patients vs. 4.04 ± 1.40 in controls, p = 0.02).
The excision activity for Ade in PBL was higher than that for Cyt.
Analysis for ANPG and TDG genes mutations and polymorphism
To explain the differences in Ade and Cyt excision activities in PBL between CRC
patients and controls, as well as between tumor and non-affected colon tissues, we searched
for mutations and polymorphisms in the ANPG and TDG genes. PBL of 56 healthy controls
and colon tissues (tumor and normal surrounding) of 43 CRC patients were analyzed. The
ANPG and TDG genes were screened in all exons using the MSSCP method. No
13
polymorphisms were found in the ANPG gene. TDG analysis revealed a G/A substitution in
exon 5 in both PBL and in normal and tumor tissues of only three patients. This change of
Gly199 to Ser in the protein sequence does not alter TDG enzymatic activity of the
polymorphic variant in comparison to the wild type protein [38]. In our study, mean Cyt
excision activity (pmols/h/mg protein SD) in normal colon tissue (2.57 2.04) and colon
tumor (5.45 1.57) of patients with TDG Gly199Ser variant was similar to mean activities for
the whole group (2.57 1.88 in normal colon and 4.07 2.4 in colon tumor). Thus, mutations
and polymorphisms within the coding sequences of ANPG and TDG DNA glycosylases were
not responsible for the lower Ade and Cyt excision in PBL of our CRC patients.
mRNA level of ANPG, TDG and APE1 in PBL of CRC patients and controls
In order to further investigate the underlying differences in repair capacity between
CRC patients and healthy controls we measured the mRNA level of TDG and ANPG
glycosylases and of APE1 in PBL. APE1 can activate Cyt excision from damaged DNA in
vitro up to 400-fold [21]. mRNA levels of all three enzymes were significantly higher in PBL
of CRC patients in comparison to healthy controls (Table 3).
A strong positive correlation was observed between mRNA level of ANPG and APE1
as well as TDG and APE1 in leukocytes of both analyzed groups (ρ = 0.81, p = 0.0000, n = 44,
and ρ = 0.94, p = 0.0000, n = 39 for ANPG and APE1 of CRC patients and healthy controls,
respectively; and ρ = 0.52, p = 0.0000, n = 44, and ρ = 0.65, p = 0.0000, n = 39 for TDG and
APE1 of patients and controls, respectively). Moreover, a positive association was found
between mRNA level of ANPG and TDG in leukocytes of healthy individuals (ρ = 0.58, p =
0.0001, n = 39). In leukocytes of CRC patients there was only a week positive correlation
14
between mRNA level of ANPG and TDG, however of borderline significance (ρ = 0.3, p =
0.052, n = 44). The correlations were based on ANPG, TDG and APE1 gene expressions
where the reference gene was the same for all three genes.
Etheno adducts levels in PBL of CRC patients and controls
We also measured the level of εdA and 1,N2-εdG in DNA of PBL of CRC patients and
controls by LC/MS/MS. Contrary to our expectation, the mean DNA adduct level of both
lesions was found to be lower in patients than in controls (Table 4). The values expressed per
108 parent nucleotide were for εdA 0.85 ± 0.46 in CRC patients vs. 2.61 ± 1.82 in controls (p =
0.00004); and for 1,N2-εdG 3.71 ± 1.45 in patients vs. 5.19 ± 2.03 in controls (p = 0.013).
Repair capacity and adducts level for Ade and Cyt in tumor and asymptomatic colon tissues
of CRC patients
Ade and Cyt excision activities were measured in tumorous and normal colon tissues
(free of histological changes) of CRC patients. Both excision activities were significantly
higher in tumor vs. normal colon tissues (Fig. 2A), being 6- and 2-fold higher for Ade and
Cyt, respectively. We found a weak positive correlation between Ade excision activity in
asymptomatic colon tissue and in leukocytes of colon cancer patients (ρ = 0.32, p = 0.035, n =
44). There was no such association for Cyt excision activity in colon tissues and leukocytes.
dA and dC levels were measured by immunoaffinity-32P-postlabeling [31] in DNA
of tumor and normal colon tissues of 12-14 patients (insufficient biological material was
available for etheno adducts analyses of all subjects). The mean levels of etheno adducts (per
108 parent nucleotides) were: 5.39 6.56 for dA in tumor and 9.93 16.24 in normal colon
15
(p = 0.37), and 5.76 6.09 for dC in tumor and 16.66 30.35 in normal colon (p = 0.23; Fig.
2B). Even with this limited data set a negative correlation was found between dC level in
DNA of normal colon tissue and its excision rate (ρ = -0.64, p = 0.013, n = 14). No such
association was, however, found for deexcision rate and dA level in DNA.
mRNA levels of ANPG and TDG were similar in tumor and normal colon tissues
(Table 5). However, mRNA level of APE1 was significantly higher in tumor (61.07 ± 6.88 x
10-4 arbitrary units) than in normal colon (35.27 ± 4.75 x 10-4, p = 0.034; Table 5). Strong
positive correlation was observed between mRNA level of ANPG and APE1 (ρ = 0.97, p =
0.0000, n = 52 in normal colon; and ρ = 0.94, p = 0.0000, n = 52 in tumor) as well as between
TDG and APE1 (ρ = 0.95, p = 0.000, n = 52 in normal colon; and ρ = 0.92, p = 0.0000, n = 52
in tumor). Moreover, there was a positive association between mRNA level of ANPG and
TDG in normal colon (ρ = 0.96, p = 0.0000, n = 52) and in tumor (ρ = 0.53, p = 0.0000, n =
52).
Effect of tobacco smoking, sex, age and cancer stage
No apparent effects of tobacco smoking, sex and age on Ade and Cyt excision
activities and mRNA level of repair enzymes were observed in PBL of CRC patients and
controls, and in colon tissues of CRC patients. Also no repair differences were found in
relation to the cancer stage according to Duke’s classification (data not shown).
Discussion
The etiology of sporadic colorectal tumors involves inflammatory processes, excess
consumption of animal fat and meat, obesity, lack of physical activity, alcohol drinking, and
16
tobacco smoking [3]. Most of these factors may increase oxidatively generated DNA damage,
and alcohol consumption decreases the activity of some DNA repair mechanisms, namely O6methylguanine methyltransferase [39]. Generation of DNA damage via lipid peroxidation
occurs both during inflammation [10] and in response to dietary imbalance, like choline- or
amino acid deficiency or alcohol drinking [40, 41]. Moreover, LPO products may affect the
activity of DNA repair enzymes [42-44].
Since etiopathogenesis of colon cancer often has an inflammatory and dietary
component [3], we searched for possible differences in repair capacity of LPO-derived etheno
DNA adducts in PBL of CRC patients vs. healthy control subjects.
First, we measured Ade and Cyt excision activity in PBL and found it to be
significantly lower in CRC patients when compared to controls matched for age, sex and
smoking habit (Figure 1B and Table 1). However, in colon tissues, excision activities for Ade
and Cyt were significantly higher in tumor than in adjacent, asymptomatic colon tissue. Then
we analyzed the levels of etheno adducts by ultrasensitive and specific methods in PBL of
CRC patients and controls as well as in colon tissues. The level of dA was lower than that of
dC and 1,N2-dG (Table 4 and Figure 2B). In PBL of CRC patients dA and 1,N2-dG level
was unexpectedly lower than in controls (Table 4), and the level of dA and dC was also
lower in tumor than in normal colon tissue (Figure 2B). This suggests that the regulation of
etheno adduct level and repair may be multifactorial and may differ in target (colon) and
surrogate (PBL) tissues.
We are currently unable to pinpoint the mechanisms responsible for the lower Ade
and Cyt excision activity in PBL of CRC patients. We hypothesized that the observed lower
etheno adduct excision in patients could be due to (i) inactivating mutations, and
17
polymorphisms in DNA repair genes or (ii) because of repair enzyme inhibition by the
products/signals of oxidative stress/inflammation, that are increasingly released during
premalignant and malignant disease stages, or (iii) different subtypes in the population of
leukocytes in patients than in controls, due to oxidative stress-driven selection. As shown by
our results, lower excision was not linked to mutations in exons of ANPG and TDG genes nor
to downregulation of mRNA expression of ANPG and TDG glycosylases or its BER activator,
APE1 endonuclease; in the contrary, mRNA levels of these repair enzymes were significantly
higher in PBL of CRC patients (Table 3). Thus direct inactivation/inhibition of repair enzymes
activities by inflammatory mediators released in the blood stream is therefore a more likely
explanation. Indeed repair inhibition by hydrogen peroxide, NO [45-47] and LPO-products,
such as HNE has been described [42-44] in vitro and in cultured cells. HNE-protein adducts
were also found to be good oxidative stress and predictive markers for disease-free survival in
brain cancer and hepatocellular carcinoma patients [48, 49]. Also HNE-adducted proteins may
be more rapidly degraded by the proteasome [50], thus lowering the excision rate of Ade and
Cyt. Indeed as shown for several human cancer sites and animal tumors, persistent oxidative
stress and LPO-induced DNA damage increased already during premalignant stages in a timedependent manner [51]. This would explain why in CRC patients excision capacity was not
different according to Duke’s stage of the tumor. However, one can not exclude the possibility
that the differences in excision of Ade and Cyt may just reflect different leukocyte types
between CRC patients and controls.
Increased oxidative stress may induce repair processes in cancer patients in relation to
controls, what can be suggested by our results from mRNA analysis of repair genes (Table 3).
A number of studies demonstrated increase of mRNA level of several repair enzymes upon
18
oxidative stress [12, 52, 53]. The observed lower Ade and Cyt excision and simultaneous
lower level of etheno adducts in PBL of cancer patients may thus suggest that other
mechanisms are also up-regulated in CRC patients and control etheno adducts level in DNA.
Other repair systems engaged in etheno adduct elimination from DNA include e.g. oxidative
demethylation by AlkB family proteins or nucleotide excision repair [14, 54].
Of interest are our findings in tumorous colon tissue. The Ade and Cyt excising
activities were much higher than in the surrounding normal colon and agreed with a 2-3 times
(statistically non significant) lower dA and dC adduct levels (Fig. 2). In normal tissue a
negative correlation was found between dC level in DNA of normal colon tissue and its
excision rate (ρ = -0.64, p = 0.013, n = 14), which supports a role of BER in Cyt-elimination,
however, does not exclude other etheno adduct control systems. In several types of (non
tumorous) human tissues previously analyzed, the dA level in DNA was found to be lower
than that of dC [51], consistent with our observation that excision activity was found to be
higher for Ade than for Cyt (Fig. 1 and 2A).
Concerning the lower ratio of etheno DNA adduct levels in the malignant vs. noncancerous (premalignant) colon, similar observations have been made in previous studies: in
colon cancer-prone patients, suffering from ulcerative colitis, Crohn's disease or familial
adenomatous polyposisdA and dC in DNA of premalignant epithelium were significantly
higher [55], while in tumor tissues of sporadic colon carcinomas, the levels of dA and dC as
in our study were lower in comparison to the surrounding normal tissue [56]. Higher Ade and
Cyt excision activity in tumor tissue in relation to adjacent, normal colon may be due to
increased transcription caused by oxidative stress and mutations in genes encoding regulatory
proteins. We observed higher mRNA level of APE1 (Table 5), which stimulates Cyt excision
19
[21]. Another possibility are frequent mutations in APC gene in colon tumors [57]. APC tumor
suppressor protein inhibits long patch BER, engaged in repair of Ade and Cyt via direct
interaction with DNA polymerase β and FEN1 endonuclease [58]. BER activity was inversely
associated with APC expression in several breast cancer cell lines [59].
Of special interest for a better understanding of carcinogenesis mechanisms are recent
data to show that overstimulation and unbalancing of BER pathway may favor genomic
instability and cancer progression: in inflamed non-cancerous colon tissues of ulcerative
colitis patients (at high risk for colon cancer, that accumulate high etheno DNA adducts [55],
ANPG and APE1 were significantly increased, and microsatellite instability (MSI) was
positively correlated with their imbalanced activities of repair enzymes [60]. Similarly in yeast
and human cells over-expression of alkylpurine DNA glycosylase (human ANPG) and/or
Ape1 was associated with frameshift mutations and MSI [60].
In the present study, we observed in tumorous colon tissue of CRC patients which
contained etheno DNA adducts, higher (ANPG related) Ade excision activity and mRNA
level of APE1. In analogy we infer from our (limited) data that deregulation of repair
processes may act as a driving force for colon carcinogenesis, whereby LPO-derived DNA
damage and induction of DNA repair enzymes [61] may occur years before clinical
manifestation of malignancy.
We are aware that our case-control studies have limitations. (i) The experiments are
done on a surrogate tissue, PBL. Although PBL may not be the direct target of stresstriggering compounds, which can promote colon cancer development and/or progression, cells
from peripheral blood that migrate and circulate through the colon may be exposed to these
compounds in this tissue, as shown for tobacco carcinogens [62]. Moreover, leukocytes are
20
often used as surrogate cells, which are supposed to inform about oxidative stress in other
tissues [63]. It is also noteworthy that leukocytes reflect the genetic profile of an individual
and are the only available cells that may be analyzed in all subject groups. (ii) This casecontrol study design does not allow to state whether lower repair activities and higher mRNA
level of ANPG, TDG and APE1 observed in patients are a cause or a result of the disease.
However, Paz-Elizur et al. [64] showed that OGG1 activities in PBL of patients with
squamous cell carcinoma of the head and neck were very similar at diagnosis and then 3-4
years after recovery.
In conclusion, we have shown, for the first time, that excision activity of Ade and
Cyt is lower in PBL of CRC patients when compared to healthy individuals. However, our
unexpected findings, that the level of dA and 1,N2-dG was lower in DNA of PBL from CRC
patients than in healthy controls, despite the reduced adduct excising activity of the former,
currently remains mechanistically difficult to explain. If such etheno adducts, particularly Cyt
excision deficiency also occurs during premalignant stages in the target organ, still may favor
acquisition of mutations in critical genes leading to neoplastic transformation. TDG
glycosylase removes Thy and Ura paired with guanine, and may prevent induction of
mutations caused by cytosine and 5-methylcytosine deamination [65]. Investigating
correlation between Ade and Cyt excision activities in PBL and colon tissues of CRC
patients, we found only a weak association between Ade excision rate in PBL and normal
colon tissue (ρ = 0.32, p = 0.035, n = 44). In future studies, it would be of great interest to
compare etheno adduct repair capacities in histologically unchanged colon from colon cancer
patients, and in colon tissues resected for non-neoplastic diseases, e.g., rectal prolapse,
diverticular disease or volvulus.
21
Acknowledgments
This work was supported by Ministry of Science and Higher Education, grant No 3
P05A 019 25. We thank all patients and volunteers who participated in this study.
Abbreviations
BER – Base Excision Repair; CRC – colorectal cancer; Ade – 1,N6-ethenoadenine; dA –
1,N6-etheno-2’-deoxyadenosine; Cyt – 3,N4-ethenocytosine;
dC
– 3,N4-etheno-2’-
deoxycytidine; 1,N2- εdG – 1,N2-etheno-2’-deoxyguanosine; LPO – lipid peroxidation; MMR
– mismatch repair; MSSCP – Multitemperature Single Strand Conformation Polymorphism;
ROS – reactive oxygen species; RNS – reactive nitrogen species
References
[1]
Greenlee, R. T.; Murray, T.; Bolden, S.; Wingo, P. A. Cancer statistics, 2000. CA
Cancer J. Clin. 50:7-33; 2000.
[2]
Didkowska, J.; Wojciechowska, W.; Tarkowski, W.; Zatonski, W. A. Cancer in
Poland. Warsaw: Centre of Oncology Scientific Publication 2002.
[3]
Levi, F.; Pasche, C.; La Vecchia, C.; Lucchini, F.; Franceschi, S. Food groups and
colorectal cancer risk. Br. J. Cancer 79:1283-1287; 1999.
[4]
Dizdaroglu, M.; Olinski, R.; Doroshow, J. H.; Akman, S. A. Modification of DNA
bases in chromatin of intact target human cells by activated human polymorphonuclear
leukocytes. Cancer Res. 53:1269-1272; 1993.
[5]
Bartsch, H.; Nair, J. Oxidative stress and lipid peroxidation-derived DNA-lesions in
inflammation driven carcinogenesis. Cancer Detect. Prev. 28:385-391; 2004.
[6]
Chung, F. L.; Nath, R. G.; Nagao, M.; Nishikawa, A.; Zhou, G. D.; Randerath, K.
Endogenous formation and significance of 1,N2-propanodeoxyguanosine adducts. Mutat. Res.
424:71-81; 1999.
[7]
Chung, F. L.; Pan, J.; Choudhury, S.; Roy, R.; Hu, W.; Tang, M. S. Formation of trans4-hydroxy-2-nonenal- and other enal-derived cyclic DNA adducts from omega-3 and omega-6
polyunsaturated fatty acids and their roles in DNA repair and human p53 gene mutation.
Mutat. Res. 531:25-36; 2003.
[8]
Chen, H. J.; Chung, F. L. Epoxidation of trans-4-hydroxy-2-nonenal by fatty acid
hydroperoxides and hydrogen peroxide. Chem. Res. Toxicol. 9:306-312; 1996.
22
[9]
Reddy, B. S.; Rao, C. V. Novel approaches for colon cancer prevention by
cyclooxygenase-2 inhibitors. J. Environ. Pathol. Toxicol. Oncol. 21:155-164; 2002.
[10] Bartsch, H.; Nair, J. Chronic inflammation and oxidative stress in the genesis and
perpetuation of cancer: role of lipid peroxidation, DNA damage, and repair. Langenbecks
Arch. Surg. 391:499-510; 2006.
[11] Bartsch, H.; Barbin, A.; Marion, M. J.; Nair, J.; Guichard, Y. Formation, detection, and
role in carcinogenesis of ethenobases in DNA. Drug Metab. Rev. 26:349-371; 1994.
[12] Tudek, B. Base excision repair modulation as a risk factor for human cancers. Mol.
Aspects Med. 28:258-275; 2007.
[13] Ghimenti, C.; Tannergard, P.; Wahlberg, S.; Liu, T.; Giulianotti, P. G.; Mosca, F.;
Fornaciari, G.; Bevilacqua, G.; Lindblom, A.; Caligo, M. A. Microsatellite instability and
mismatch repair gene inactivation in sporadic pancreatic and colon tumours. Br. J. Cancer
80:11-16; 1999.
[14] Hang, B. Repair of exocyclic DNA adducts: rings of complexity. Bioessays 26:11951208; 2004.
[15] Delaney, J. C.; Smeester, L.; Wong, C.; Frick, L. E.; Taghizadeh, K.; Wishnok, J. S.;
Drennan, C. L.; Samson, L. D.; Essigmann, J. M. AlkB reverses etheno DNA lesions caused
by lipid oxidation in vitro and in vivo. Nat. Struct. Mol. Biol. 12:855-860; 2005.
[16] Saparbaev, M.; Kleibl, K.; Laval, J. Escherichia coli, Saccharomyces cerevisiae, rat
and human 3-methyladenine DNA glycosylases repair 1,N6-ethenoadenine when present in
DNA. Nucleic Acids Res. 23:3750-3755; 1995.
[17] Saparbaev, M.; Laval, J. 3,N4-ethenocytosine, a highly mutagenic adduct, is a primary
substrate for Escherichia coli double-stranded uracil-DNA glycosylase and human mismatchspecific thymine-DNA glycosylase. Proc. Natl. Acad. Sci. U. S. A. 95:8508-8513; 1998.
[18] Bellacosa, A. Role of MED1 (MBD4) Gene in DNA repair and human cancer. J. Cell
Physiol. 187:137-144; 2001.
[19] Jiricny, J.; Menigatti, M. DNA Cytosine demethylation: are we getting close? Cell
135:1167-1169; 2008.
[20] Saparbaev, M.; Langouet, S.; Privezentzev, C. V.; Guengerich, F. P.; Cai, H.; Elder, R.
H.; Laval, J. 1,N(2)-ethenoguanine, a mutagenic DNA adduct, is a primary substrate of
Escherichia coli mismatch-specific uracil-DNA glycosylase and human alkylpurine-DNA-Nglycosylase. J. Biol. Chem. 277:26987-26993; 2002.
[21] Privezentzev, C. V.; Saparbaev, M.; Laval, J. The HAP1 protein stimulates the
turnover of human mismatch-specific thymine-DNA-glycosylase to process 3,N(4)ethenocytosine residues. Mutat. Res. 480-481:277-284; 2001.
[22] Demple, B.; Sung, J. S. Molecular and biological roles of Ape1 protein in mammalian
base excision repair. DNA Repair (Amst) 4:1442-1449; 2005.
[23] Um, S.; Harbers, M.; Benecke, A.; Pierrat, B.; Losson, R.; Chambon, P. Retinoic acid
receptors interact physically and functionally with the T:G mismatch-specific thymine-DNA
glycosylase. J. Biol. Chem. 273:20728-20736; 1998.
[24] Tini, M.; Benecke, A.; Um, S. J.; Torchia, J.; Evans, R. M.; Chambon, P. Association
of CBP/p300 acetylase and thymine DNA glycosylase links DNA repair and transcription.
Mol. Cell 9:265-277; 2002.
[25] Missero, C.; Pirro, M. T.; Simeone, S.; Pischetola, M.; Di Lauro, R. The DNA
glycosylase T:G mismatch-specific thymine DNA glycosylase represses thyroid transcription
factor-1-activated transcription. J. Biol. Chem. 276:33569-33575; 2001.
23
[26] Bradford, M. M. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254;
1976.
[27] Speina, E.; Zielinska, M.; Barbin, A.; Gackowski, D.; Kowalewski, J.; Graziewicz, M.
A.; Siedlecki, J. A.; Olinski, R.; Tudek, B. Decreased repair activities of 1,N(6)-ethenoadenine
and 3,N(4)-ethenocytosine in lung adenocarcinoma patients. Cancer Res. 63:4351-4357; 2003.
[28] Rusin, M.; Samojedny, A.; Harris, C. C.; Chorazy, M. Novel genetic polymorphisms in
DNA repair genes: O(6)-methylguanine-DNA methyltransferase (MGMT) and Nmethylpurine-DNA glycosylase (MPG) in lung cancer patients from Poland. Hum. Mutat.
14:269-270; 1999.
[29] Schmutte, C.; Baffa, R.; Veronese, L. M.; Murakumo, Y.; Fishel, R. Human thymineDNA glycosylase maps at chromosome 12q22-q24.1: a region of high loss of heterozygosity
in gastric cancer. Cancer Res. 57:3010-3015; 1997.
[30] Kaczanowski, R.; Trzeciak, L.; Kucharczyk, K. Multitemperature single-strand
conformation polymorphism. Electrophoresis 22:3539-3545; 2001.
[31] Nair, J.; Barbin, A.; Guichard, Y.; Bartsch, H. 1,N6-ethenodeoxyadenosine and 3,N4ethenodeoxycytine in liver DNA from humans and untreated rodents detected by
immunoaffinity/32P-postlabeling. Carcinogenesis 16:613-617; 1995.
[32] Douki, T.; Odin, F.; Caillat, S.; Favier, A.; Cadet, J. Predominance of the 1,N2propano 2'-deoxyguanosine adduct among 4-hydroxy-2-nonenal-induced DNA lesions. Free
Radic. Biol. Med. 37:62-70; 2004.
[33] Chomczynski, P. A reagent for the single-step simultaneous isolation of RNA, DNA
and proteins from cell and tissue samples. Biotechniques 15:532-534, 536-537; 1993.
[34] Lange, H.; Holec, S.; Cognat, V.; Pieuchot, L.; Le Ret, M.; Canaday, J.; Gagliardi, D.
Degradation of a polyadenylated rRNA maturation by-product involves one of the three
RRP6-like proteins in Arabidopsis thaliana. Mol. Cell Biol. 28:3038-3044; 2008.
[35] Wery, M.; Ruidant, S.; Schillewaert, S.; Lepore, N.; Lafontaine, D. L. The nuclear
poly(A) polymerase and Exosome cofactor Trf5 is recruited cotranscriptionally to nucleolar
surveillance. Rna 15:406-419; 2009.
[36] Livak, K. J.; Schmittgen, T. D. Analysis of relative gene expression data using realtime quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25:402-408; 2001.
[37] Luu-The, V.; Paquet, N.; Calvo, E.; Cumps, J. Improved real-time RT-PCR method for
high-throughput measurements using second derivative calculation and double correction.
Biotechniques 38:287-293; 2005.
[38] Sard, L.; Tornielli, S.; Gallinari, P.; Minoletti, F.; Jiricny, J.; Lettieri, T.; Pierotti, M.
A.; Sozzi, G.; Radice, P. Chromosomal localizations and molecular analysis of TDG generelated sequences. Genomics 44:222-226; 1997.
[39] Sutandyo, N. Nutritional carcinogenesis. Acta Med Indones 42:36-42.
[40] Kawai, Y.; Kato, Y.; Nakae, D.; Kusuoka, O.; Konishi, Y.; Uchida, K.; Osawa, T.
Immunohistochemical detection of a substituted 1,N(2)-ethenodeoxyguanosine adduct by
omega-6 polyunsaturated fatty acid hydroperoxides in the liver of rats fed a choline-deficient,
L-amino acid-defined diet. Carcinogenesis 23:485-489; 2002.
[41] Wang, Y.; Millonig, G.; Nair, J.; Patsenker, E.; Stickel, F.; Mueller, S.; Bartsch, H.;
Seitz, H. K. Ethanol-induced cytochrome P4502E1 causes carcinogenic etheno-DNA lesions
in alcoholic liver disease. Hepatology (Baltimore, Md) 50:453-461; 2009.
24
[42] Feng, Z.; Hu, W.; Tang, M. S. Trans-4-hydroxy-2-nonenal inhibits nucleotide excision
repair in human cells: a possible mechanism for lipid peroxidation-induced carcinogenesis.
Proc. Natl. Acad. Sci. U. S. A. 101:8598-8602; 2004.
[43] Feng, Z.; Hu, W.; Marnett, L. J.; Tang, M. S. Malondialdehyde, a major endogenous
lipid peroxidation product, sensitizes human cells to UV- and BPDE-induced killing and
mutagenesis through inhibition of nucleotide excision repair. Mutat. Res. 601:125-136; 2006.
[44] Feng, Z.; Hu, W.; Hu, Y.; Tang, M. S. Acrolein is a major cigarette-related lung cancer
agent: Preferential binding at p53 mutational hotspots and inhibition of DNA repair. Proc.
Natl. Acad. Sci. U. S. A. 103:15404-15409; 2006.
[45] Wink, D. A.; Laval, J. The Fpg protein, a DNA repair enzyme, is inhibited by the
biomediator nitric oxide in vitro and in vivo. Carcinogenesis 15:2125-2129; 1994.
[46] Jaiswal, M.; LaRusso, N. F.; Nishioka, N.; Nakabeppu, Y.; Gores, G. J. Human Ogg1,
a protein involved in the repair of 8-oxoguanine, is inhibited by nitric oxide. Cancer Res.
61:6388-6393; 2001.
[47] Kelley, M. R.; Parsons, S. H. Redox regulation of the DNA repair function of the
human AP endonuclease Ape1/ref-1. Antioxid. Redox Signal. 3:671-683; 2001.
[48] Juric-Sekhar, G.; Zarkovic, K.; Waeg, G.; Cipak, A.; Zarkovic, N. Distribution of 4hydroxynonenal-protein conjugates as a marker of lipid peroxidation and parameter of
malignancy in astrocytic and ependymal tumors of the brain. Tumori 95:762-768; 2009.
[49] Maki, A.; Kono, H.; Gupta, M.; Asakawa, M.; Suzuki, T.; Matsuda, M.; Fujii, H.;
Rusyn, I. Predictive power of biomarkers of oxidative stress and inflammation in patients with
hepatitis C virus-associated hepatocellular carcinoma. Ann Surg Oncol 14:1182-1190; 2007.
[50] Grune, T.; Davies, K. J. The proteasomal system and HNE-modified proteins. Mol.
Aspects Med. 24:195-204; 2003.
[51] Bartsch, H.; Nair, J. Accumulation of lipid peroxidation-derived DNA lesions:
potential lead markers for chemoprevention of inflammation-driven malignancies. Mutat. Res.
591:34-44; 2005.
[52] Cherng, S. H.; Huang, K. H.; Yang, S. C.; Wu, T. C.; Yang, J. L.; Lee, H. Human 8oxoguanine DNA glycosylase 1 mRNA expression as an oxidative stress exposure biomarker
of cooking oil fumes. J. Toxicol. Environ. Health A 65:265-278; 2002.
[53] Das, A.; Hazra, T. K.; Boldogh, I.; Mitra, S.; Bhakat, K. K. Induction of the human
oxidized base-specific DNA glycosylase NEIL1 by reactive oxygen species. J. Biol. Chem.
280:35272-35280; 2005.
[54] Ringvoll, J.; Moen, M. N.; Nordstrand, L. M.; Meira, L. B.; Pang, B.; Bekkelund, A.;
Dedon, P. C.; Bjelland, S.; Samson, L. D.; Falnes, P. O.; Klungland, A. AlkB homologue 2mediated repair of ethenoadenine lesions in mammalian DNA. Cancer Res 68:4142-4149;
2008.
[55] Nair, J.; Gansauge, F.; Beger, H.; Dolara, P.; Winde, G.; Bartsch, H. Increased ethenoDNA adducts in affected tissues of patients suffering from Crohn's disease, ulcerative colitis,
and chronic pancreatitis. Antioxid. Redox Signal. 8:1003-1010; 2006.
[56] Schmid, K.; Nair, J.; Winde, G.; Velic, I.; Bartsch, H. Increased levels of promutagenic
etheno-DNA adducts in colonic polyps of FAP patients. Int. J. Cancer 87:1-4; 2000.
[57] Luchtenborg, M.; Weijenberg, M. P.; Wark, P. A.; Saritas, A. M.; Roemen, G. M.; van
Muijen, G. N.; de Bruine, A. P.; van den Brandt, P. A.; de Goeij, A. F. Mutations in APC,
CTNNB1 and K-ras genes and expression of hMLH1 in sporadic colorectal carcinomas from
the Netherlands Cohort Study. BMC Cancer 5:160; 2005.
25
[58] Jaiswal, A. S.; Balusu, R.; Armas, M. L.; Kundu, C. N.; Narayan, S. Mechanism of
adenomatous polyposis coli (APC)-mediated blockage of long-patch base excision repair.
Biochemistry 45:15903-15914; 2006.
[59] Narayan, S.; Jaiswal, A. S.; Balusu, R. Tumor suppressor APC blocks DNA
polymerase beta-dependent strand displacement synthesis during long patch but not short
patch base excision repair and increases sensitivity to methylmethane sulfonate. J. Biol. Chem.
280:6942-6949; 2005.
[60] Hofseth, L. J. The adaptive imbalance to genotoxic stress: genome guardians rear their
ugly heads. Carcinogenesis 25:1787-1793; 2004.
[61] Winnepenninckx, V.; Lazar, V.; Michiels, S.; Dessen, P.; Stas, M.; Alonso, S. R.;
Avril, M. F.; Ortiz Romero, P. L.; Robert, T.; Balacescu, O.; Eggermont, A. M.; Lenoir, G.;
Sarasin, A.; Tursz, T.; van den Oord, J. J.; Spatz, A. Gene expression profiling of primary
cutaneous melanoma and clinical outcome. J. Natl. Cancer Inst. 98:472-482; 2006.
[62] Wiencke, J. K. DNA adduct burden and tobacco carcinogenesis. Oncogene 21:73767391; 2002.
[63] Lenton, K. J.; Therriault, H.; Fulop, T.; Payette, H.; Wagner, J. R. Glutathione and
ascorbate are negatively correlated with oxidative DNA damage in human lymphocytes.
Carcinogenesis 20:607-613; 1999.
[64] Paz-Elizur, T.; Ben-Yosef, R.; Elinger, D.; Vexler, A.; Krupsky, M.; Berrebi, A.;
Shani, A.; Schechtman, E.; Freedman, L.; Livneh, Z. Reduced repair of the oxidative 8oxoguanine DNA damage and risk of head and neck cancer. Cancer Res 66:11683-11689;
2006.
[65] Cortazar, D.; Kunz, C.; Saito, Y.; Steinacher, R.; Schar, P. The enigmatic thymine
DNA glycosylase. DNA Repair (Amst) 6:489-504; 2007.
Figure legend
Fig. 1. Excision repair of Ade and Cyt in PBL of CRC patients. (A) Typical gel of Ade
excision by increasing amounts of extracts from PBL of a patient and a healthy control. The
amount of cleaved oligodeoxynucleotide was estimated by PhosphorImaging, and PBL repair
capacity was calculated from the linear part of the curve. (B) The mean and standard deviation
of Ade and Cyt excision rate in PBL of patients and control individuals. *, p<0.05; ***,
p<0.001 (Mann-Whitney U-test).
26
Fig. 2. (A) The mean and standard deviation of Ade and Cyt excision rate in colon tumor
and normal surrounding tissues of CRC patients. (B) The mean (± standard deviation) level of
dA/108 dA and dC/108 dC in colon tumor and adjacent, asymptomatic tissue of CRC
patients. ***, p<0.001; n.s., non significant (Wilcoxon test).
27
Table 1
Characteristics of the studied subjects
Characteristic
CRC cases
n = 54 (%)
Controls
n = 56 (%)
Mean age (range)
62 (27-90)
55 (42-65)
Male
28 (52)
23 (41)
Female
26 (48)
33 (59)
Non-smokers
20 (37)
18 (32)
Light smokers
21 (39)
23 (41)
Heavy smokers
12 (22)
13 (23)
A
3 (5)
-
B1
16 (30)
-
B2
8 (15)
-
C1
2 (4)
-
C2
18 (33)
-
D
7 (13)
-
Sex
Smoking status
Cancer stage
(Duke's scalea)
a
with Astler-Coller modification
28
Table 2
Primers for Real-Time PCR
Locus
Primer
Sequence (5’ to 3’)
Annealing
temperature
NC_000012.10
GAPDHf
GAPDHr
CCATGGAGAAGGCTGGGG
CAAAGTFGTCATGGATGACC
55ºC
55ºC
18S
rRNA
X03205.1
18SpfF
18SpfR
ATTCGAACGTCTGCCCTATCA
TGCCTTCCTTGGATGTGGTAG
60ºC
59ºC
ANPG
BC014991
ANPGpeF
ANPGpeR
CGCAGCATCTATTTCTCAAGC
GTGCCATTAGGAAGTCGCC
57ºC
57ºC
TDG
BC037557
TDGpeF
TDGpeR
TAATGGGCAGTGGATGACCC
TGCAGCATTTAAGCAGAGCTGA
59ºC
60ºC
APE1
NM_080648.1
NM_001641.2
APE1peF
APE1peR
AGCCTTTCGCAAGTTCCTGA
GCGTGAAGCCAGCATTCTTT
59ºC
59ºC
Gene
GAPDH
29
Table 3
mRNA level of ANPG and TDG DNA glycosylases as well as AP endonuclease, APE1,
in relation to 18S rRNA (x 10-5) in PBL of CRC patients and healthy controls
ANPG
mRNA level (arbitrary units)
Patientsa
Controlsa
n = 44
n = 39
3.64 ± 1.60
0.32 ± 0.01
0.000000
TDG
7.25 ± 2.3
0.85 ± 0.06
0.000000
APE1
9.42 ± 6.82
1.82 ± 1.4
0.000002
DNA
glycosylase
a
p
Comparison of mRNA levels, expressed as means and standard deviations, between CRC
patients and healthy controls (Mann-Whitney U-test).
30
Table 4
The level of 1,N6-etheno-2’-deoxyadenosine (εdA) and 1,N2-etheno-2’-deoxyguanosine (1,N2εdG) in DNA of PBL of CRC patients and healthy controls
Patientsb
n = 18
Controlsb
n = 24
εdA
0.85 ± 0.46
2.61 ± 1.82
0.00004
1,N2-εdG
3.71 ± 1.45
5.19 ± 2.03
0.013
DNA adducta
p
, εdA/108 dA, 1,N2-εdG/108 dG
b
Comparison of DNA adduct, expressed as means and standard deviations, between CRC
patients and healthy controls (Mann-Whitney U-test).
a
31
Table 5
mRNA level of ANPG and TDG DNA glycosylases as well as AP endonuclease, APE1, in
relation to 18S rRNA (x 10-4) in tumor and normal adjacent colon tissues of CRC patients
ANPG
mRNA level (arbitrary units)
Tumora
Normal colona
n = 52
n = 52
31.83 ± 3.9
27.63 ± 4.8
0.47
TDG
7.85 ± 0.47
8.84 ± 0.27
0.17
APE1
61.07 ± 6.88
35.27 ± 4.75
0.034
DNA
glycosylase
a
p
Comparison of mRNA levels, expressed as means and standard deviations, between tumor
and normal colon tissue of CRC patients (Wilcoxon rank sum test).
32