Mutation Research 611 (2006) 18

Metformin does not prevent DNA damage in lymphocytes despite its antioxidant properties against cumene hydroperoxide-induced oxidative stress
Ilhan Onaran a, , Gulgun S. Guven a , Sule Beyhan Ozdas c, Gonul Kanigur a , Suphi Vehid b
a b

Department of Medical Biology, Cerrahpasa Medical Faculty, Istanbul University, Istanbul, Turkey Department of Public Health, Cerrahpasa Medical Faculty, Istanbul University, Istanbul, Turkey c Department of Chemical and Biological Engineering, Koc University, Istanbul, Turkey Received 27 October 2005; received in revised form 29 March 2006; accepted 25 June 2006 Available online 26 September 2006

Abstract Metformin (1-(diaminomethylidene)-3,3-dimethyl-guanidine), which is the most commonly prescribed oral antihyperglycaemic drug in the world, was reported to have several antioxidant properties such as the inhibition of advanced glycation end-products. In addition to its use in the treatment of diabetes, it has been suggested that metformin may be a promising anti-aging agent. The present work was aimed at assessing the possible protective effects of metformin against DNA-damage induction by oxidative stress in vitro. The effects of metformin were compared with those of N-acetylcysteine (NAC). For this purpose, peripheral blood lymphocytes from aged (n = 10) and young (n = 10) individuals were pre-incubated with various concentrations of metformin (1050 M), followed by incubation with 15 M cumene hydroperoxide (CumOOH) for 48 h, under conditions of low oxidant level, which do not induce cell death. Protection against oxidative DNA damage was evaluated by use of the Comet assay and the cytokinesis-block micronucleus technique. Changes in the levels of malondialdehyde + 4-hydroxy-alkenals, an index of oxidative stress, were also measured in lymphocytes. At concentrations ranging from 10 M to 50 M, metformin did not protect the lymphocytes from DNA damage, while 50 M NAC possessed an effective protective effect against CumOOH-induced DNA damage. Furthermore, NAC, but not metformin, inhibited DNA fragmentation induced by CumOOH. In contrast to the lack of protection against oxidative damage in lymphocyte cultures, metformin signicantly protected the cells from lipid peroxidation in both age groups, although not as effective as NAC in preventing the peroxidative damage at the highest doses. Within the limitations of this study, the results indicate that pharmacological concentrations of metformin are unable to protect against DNA damage induced by a pro-oxidant stimulus in cultured human lymphocytes, despite its antioxidant properties. 2006 Elsevier B.V. All rights reserved.
Keywords: Metformin; Oxidative stress; DNA damage; Human lymphocytes; Comet assay; Micronucleus assay

1. Introduction Oxidative stress is thought to play a major role in the etiology of a wide variety of diseases including diabetes and cancer, as well as in the aging process [1,2]. It is triggered by exposure to exogenous factors or by

Corresponding author at: Ortaklar Cd. Butan Sk. No: 2, 34394 Mecidiyek oy, Istanbul, Turkey. Fax: +90 2125861548. E-mail address: ilonaran@istanbul.edu.tr (I. Onaran).

1383-5718/$ see front matter 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.mrgentox.2006.06.036

I. Onaran et al. / Mutation Research 611 (2006) 18

chemicals producing reactive oxygen species (ROS) and is associated with an overproduction of ROS, as well as an impairment of antioxidant defense capacity. In diabetes, oxidative stress seems to arise primarily from an increase in free radical concentrations in plasma and a reduction in antioxidant defense [3,4]. ROS are well known to cause DNA damage and induce cytotoxicity. They induce a variety of lesions in DNA, including oxidized bases, abasic sites, DNA strand-breaks and cross-links between DNA and proteins. Growing evidence indicates that oxidative stress increases DNA damage in diabetics [5,6]. Similarly, long-lived animals including humans are known to accumulate aberrations in the genome and in cellular components during the aging process, which are thought to be caused by the cumulative effects of oxidative damage [7]. On the other hand, it has been pointed out that vitamin antioxidants administered in vitro and in vivo prevent DNA damage and increase the DNA-repair capacity of individuals subject to oxidative stress [8]. Therefore, agents with antioxidant activity may offer a benet to diabetic patients and could be useful in preventing or delaying the development of diabetic complications. Metformin (1-(diaminomethylidene)-3,3-dimethylguanidine) is an anti-hyperglycaemic drug commonly used for the management of type-2 diabetes. Protective effects against diabetic complications have been observed with metformin monotherapy [9]. Previous in vitro and in vivo studies have demonstrated that metformin causes an improvement in antioxidant activities in various tissues and acts to limit lipid peroxidation [1013]. Bonnefont-Rousselot et al. [14] also suggested that metformin could directly scavenge ROS or indirectly act by modulating the intracellular production of superoxide radicals (O2 ). Thus, metformin may help protect against free radical-induced DNA damage. However, the ability of metformin to modulate the DNA-damaging effects of oxidative stress is not known. Therefore, the aim of the present study is to investigate the in vitro ability of metformin to protect against oxidative stressinduced DNA damage in peripheral blood lymphocytes obtained from both elderly and younger subjects, and to compare it with the activity of N-acetylcysteine (NAC), a known antioxidant and ROS scavenger. It has been pointed out that age and diabetes are factors that inuence the generation of DNA damage. However, attempts to correlate levels of DNA damage with age or diabetes have led to contradictory results ranging from no signicant changes [15,16] to either positive [5,6,17] or negative relationships [18]. In the case of diabetes, all these reports should be interpreted with caution for various reasons. Firstly, diabetes is a disease that may

implicate various disturbances with an unknown impact on DNA. Secondly, glycaemic control seems to play a role in DNA-damage processing. Another problem follows from the age onset of this disease. In this study, the use of cells from diabetics may complicate the evaluation of possible protective effects of metformin on DNA damage induced in vitro. On the other hand, the incidence of type-2 diabetes mellitus increases with age and leads to signicant morbidity and mortality. It has been suggested that metformin decelerates aging in mice [19], and that it is the most suitable strategy to prevent diabetes in the elderly [20]. Hence, studies were undertaken using lymphocytes in from healthy aged and young subjects.
2. Materials and methods 2.1. Chemicals Metformin hydrochloride, N-acetylcysteine, cumene hydroperoxide, RPMI 1640 growth medium, penicillin, streptomycin, dimethyl sulfoxide, normal melting point agarose, low melting point agarose and ethidium bromide were obtained from Sigma chemicals, Saint Louis, MO, USA. Giemsa and trypan-blue were obtained from Merck, Darmstadt, Germany and l-glutamine, fetal calf serum, phytohaemagglutinin and cytochalasin-B were purchased from Biological Industries, Israel. Ficoll-Paque was from Pharmacia, Uppsala, Sweden. All other chemicals and solvents used were of the highest purity grade available. 2.2. Lymphocyte preparation, cell culture and treatment Following informed consent, 10 healthy young donors (5 male and 5 female; mean age, 29 5 years; range, 2039) and 10 healthy elderly volunteers (5 male and 5 female; mean age, 79 6 years; range, 7087) were included in the study. Blood was collected by venipuncture in heparinised tubes, diluted 1:1 in phosphate-buffered saline (PBS) and separated by a Ficoll gradient for isolation of lymphocytes. Cell viability was measured by use of the trypan-blue exclusion assay. Approximately 1 106 cells were seeded in RPMI medium containing l-glutamine (2 mM), fetal calf serum (20%), penicillin (100 UI/ml) and streptomycin (100 g/ml) and incubated at 37 C for 72 h. Lymphocyte stimulation was done by addition of phytohaemagglutinin (1.5%). At 48 h, various concentrations of metformin or NAC were added to cultures from each individual. After pre-incubation with metformin or NAC for 1 h, oxidative stress was induced by addition of cumene hydroperoxide (CumOOH) (nal concentration 15 M) dissolved in dimethyl sulfoxide (DMSO). All cell suspensions not treated with CumOOH were brought to a concentration of DMSO equivalent to that delivered with CumOOH. Since pharmacological metformin concentrations are close to 20 M [21], we studied metformin in the concentration range of

I. Onaran et al. / Mutation Research 611 (2006) 18

3 A is number of cells classied to group A B is number of cells classied to group B C is number of cells classied to group C D is number of cells classied to group D E is number of cells classied to group E

1050 M. NAC (50 M), dissolved in PBS and neutralized with sodium hydroxide, was added to the culture medium 1 h before CumOOH exposure. 2.3. Cytokinesis-block micronucleus assay The cytokinesis-block micronucleus (MN) assay was carried out according to Fenech [22]. Cytochalasin-B (6 g/ml) was added to cultures at 44 h to prevent cytokinesis. Cells were xed in 3:1 methanol:acetic acid without hypotonic treatment, and the suspension was dropped onto clean slides and stained with 5% Giemsa for approximately 10 min. In accordance with standard criteria [23], 1000 binucleated lymphocytes were scored for MN identication for each subject. To provide data regarding proliferation kinetics, the frequencies of mono-, bi-, tri- and tetra-nucleated cells were determined. 2.4. Comet assay A slightly modied Comet assay [24], implemented according to recent guidelines [25], was used to measure the extent of DNA damage. After cell culture, the cells were centrifuged (1000 g, 10 min) and taken up in PBS. The cell suspension was mixed with low melting-point agarose and placed on a microscope slide pre-coated with 1% normal melting-point agarose. The slides with the agarose-embedded cells were then subjected to a lysis step (1 h incubation at 4 C in 1% N-lauroylsarcosine, 2.5 M NaCl, 100 mM Na2 EDTA, 1% Triton X-100, 10% DMSO, pH 10.0). After lysis the slides were washed three times for 5 min in endonuclease buffer (40 mM HepesKOH, 0.1 M KCl, 0.5 mM EDTA, 0.2 mg/ml bovine serum albumin, pH 8.0) and incubated for 30 min with a mixture of repair enzymes (Fpg) at 37 C. The slides (processed either without or with Fpg) were then transferred to an electrophoresis box containing an alkaline solution (300 mM NaOH, 1 mM Na2 EDTA, pH 13). They were kept in this solution for a 40 min DNAunwinding period at 4 C. A current of 25 V (300 mA) was then applied for 30 min. The slides were removed and neutralized with TrisHCl (0.4 M, pH 7.5). Cells were stained with 20 l ethidium bromide (5 g/ml) and the extent of DNA migration was evaluated by visual scoring. Slides were scored without knowledge of the group by only one well-trained scorer. A Nikon Eclip E600 uoroscence microscope was used to identify comets. One thousand cells were graded by eye into ve categories, according to Anderson et al. [26]. This method has the advantage of speed; it is calibrated by reference to computer-image analysis based on uorometric measurement of DNA intensities in head and tail. The results were expressed as comet assay tail factors calculated according to Diem et al. [27], corresponding to the following amount of DNA fragments in the tail: classication group A < 5%, B < 520%, C 2040%, D 4095% and E > 95%. Tail factors were then calculated according to the following formula: AFA + BFB + CFC + DFD + EFE tail factor (%) = 1000

FA is average of group A (=2.5) FB is average of group B (=12.5) FC is average of group C (=30) FD is average of group D (=67.5) FE is average of group E (=97.5)

2.5. Estimation of DNA fragmentation A cellular DNA fragmentation assay was done by an ELISA method, using a kit (Roche Molecular Biochemicals, Mannheim, Germany) according to the recommended procedure. Cells proliferating in the culture were incubated for 2 h at 37 C with the non-radioactive thymidine analog BrdU, which is incorporated into genomic DNA. After this procedure, cells were treated with individual agents as described above. After 48 h, cells were lysed and transferred to a microtiter plate coated with an anti-DNA antibody, and incubated for 75 min at room temperature. After washing the plate three times, substrate solution was added and the plate was incubated in dark until color development was sufcient. The reaction was then stopped by addition 0.56 M H2 SO4 and the absorbance of the samples were measured with a microplate reader at 450 nm and 655 nm. 2.6. Lipid peroxidation assay Immediately after the incubation with individual agents of lymphocytes in culture, the cell suspensions in medium were centrifuged for 10 min at 1000 g. The amounts of malondialdehyde and 4-hydroxynonenal (MDA + 4-HNE) in supernatants were determined with CALBIOCHEM Lipid Peroxidation Assay kit, exactly as described by the manufacturer. The level of lipid peroxidation was expressed as the amount (in 106 cells) of MDA + 4-HNE, as major lipid peroxidation end products. In the assay, we added 2.5 mM (nal concentration) butylated hydroxytoluene to prevent sample auto-oxidation. 2.7. Statistical analysis and data presentation Values reported are means S.D. All data were normally distributed and underwent equal variance testing. The experiments were analyzed with the general linear model of SPSS 11.5 for Windows (SPSS, Chicago, IL). Signicance (p < 0.05) was determined with a one-way ANOVA, Tukeys HSD test.

3. Results To investigate the protective effects of metformin against DNA damage in lymphocytes when exposed to in vitro-induced oxidative stress, we focused on cells

I. Onaran et al. / Mutation Research 611 (2006) 18

treated at low oxidant levels that do not decrease cell viability. In preliminary studies, when lymphocytes were exposed to different doses of CumOOH for 48 h, the dose response curves showed that a concentration of 15 M CumOOH was the highest at which cell death among lymphocytes of elderly (n = 4) and younger (n = 5) subjects did not occur. Exposure of cells to this concentration revealed insignicant changes in cell-death rate of 3.4 0.5%. Concentrationresponse determination for further experiments was not practical and therefore all samples were studied at a xed concentration of CumOOH. In all subjects, 15 M CumOOH induced signicant increases in MN frequency compared with control values (p < 0.05). However, a statistically signicant difference was observed in the sensitivity of young and elderly individuals, as seem by differences in the mean frequency of MN (p < 0.05) (Table 1). In addition, there was a signicant difference in the MN frequency observed in control cultures from young and elderly persons. We also determined the comet assay tail-factor (DNA damage level) in peripheral lymphocytes. As shown in Table 1, the levels of CumOOH-induced DNA damage in lymphocytes were again signicantly higher in old subjects compared with younger ones. In all lymphocyte samples, pre-treatment with metformin concentrations ranging from 10 M to 50 M did not lead to a signicant decrease in CumOOHinduced micronucleus-forming activity (p > 0.05). The results obtained by the Comet assay on lymphocytes conrmed the data of the MN test (Table 1). Under the same experimental conditions, pre-treatment of the cells with 50 M NAC produced a signicant reduction in DNA damage measured by Comet assay and in the frequency of micronuclei, when the agent was present with CumOOH throughout the incubation period. Table 1

shows the inuence of metformin and NAC on MN frequency and DNA damage induced by CumOOH. In addition, experiments with Fpg post-treatment further indicated that there was no reduction in the amount of oxidative base damage after supplementation with metformin (data not shown). Although under our experimental conditions the in vitro treatments of cells did not signicantly induce cell death, the concentration of CumOOH that was applied may induce apoptosis. To examine this possibility, we used a cellular DNA fragmentation ELISA assay that quanties DNA fragmentation caused by apoptosis. The basal and CumOOH-induced levels of apoptosis in lymphocytes from donors were quantied using this ELISA procedure, which measures the release of mono- and oligonucleosomal fragments into the cytoplasm. Treatment of lymphocytes with 15 M CumOOH for 48 h caused a signicant increase in DNA fragmentation compared with that in untreated cells. The DNA fragmentation in lymphocytes from aging individuals was approximately 1.65 times the amount in untreated cells, similar to the increase seen with lymphocytes from young individuals, which was 1.67-fold. As shown in Fig. 1, treatment with the highest dose of metformin (50 M) did not signicantly affect the levels of DNA fragmentation produced by CumOOH, while NAC treatment at the same concentration signicantly inhibited the DNA fragmentation. MDA + 4-HNE levels were used to measure the oxidative damage in lymphocytes treated with individual agents. Despite the differences in inhibition proles of the MDA + 4-HNE formation in lymphocytes from aged and young subjects, the increased lipid peroxidation induced by CumOOH was attenuated by co-incubation with metformin or NAC. Treatment of the cells with the highest dose of metformin

Table 1 Effect of metformin on CumOOH-induced DNA damage estimated by MN and Comet assay in peripheral lymphocytes from young and elderly individuals Treatment MN/1000 binucleated lymphocytes Aging group (n = 10) Basal 0 M Metformin + CumOOH 10 M Metformin + CumOOH 20 M Metformin + CumOOH 50 M Metformin + CumOOH 50 M l-NAC + CumOOH 14.80 25.20 24.57 25.21 24.80 17.34 4.98a 8.59a,b 4.42a,b 4.88a,b 8.02a,b 5.44a Young group (n = 10) 8.80 16.80 15.97 17.01 17.30 9.71 2.39 5.59b 3.99b 3.74b 4.87b 2.02 Comet assay tail factor (%) Aging group (n = 10) 6.53 11.99 10.52 10.28 9.65 6.01 1.88a 3.21a,b 2.61a,b 2.87a,b 2.77a,b 1.31a Young group (n = 10) 3.19 7.32 7.08 7.15 7.54 3.70 0.33 1.90b 2.14b 2.43b 1.61b 0.67

Cells were treated with metformin (1050 M) or NAC (50 M) for 1 h before addition of 15 M CumOOH and were incubated in culture medium for 48 h. a Signicant difference (p < 0.05) with regard to young group. b Signicant difference (p < 0.05) with regard to the basal values.

I. Onaran et al. / Mutation Research 611 (2006) 18

Fig. 1. Effect of metformin and NAC on CumOOH-induced cellular DNA fragmentation in lymphocytes from elderly (n = 10) and young (n = 10) individuals as assessed by ELISA. The following conditions were tested: Basal; 15 M CumOOH; 50 M Metformin + 15 M CumOOH; 50 M N-acetylcysteine + 15 M CumOOH. The results are the mean S.D. * Signicant difference (p < 0.05) with regard to the cells incubated with CumOOH alone.

(50 M) signicantly decreased CumOOH-induced lipid peroxidation compared with CumOOH alone (Table 2). At this dose of metformin, the inhibition percentages of CumOOH-induced lipid peroxidation in lymphocytes from aged and young subjects were 61% and 56%, respectively. Under the same experimental conditions, the protective effect of metformin against lipid peroxidation in lymphocytes was smaller than that in lymphocytes treated with NAC (Table 2).

4. Discussion To test whether metformin could inhibit DNA damage and apoptosis induced by oxidative stress in lymphocytes as target cells we used cumene hydroperoxide, a known genotoxic agent. CumOOH, which is a strong non-polar oxidizing agent used in industry, is easily taken up by cells and is not metabolized by catalase [28]. Oxidative stress generated by CumOOH caused certain aspects of aging as well as induction of DNA damage [29]. In the present study, in vitro anti-genotoxic effects of metformin were compared with those of NAC, commonly used by the pharmaceutical industry. Numerous studies have demonstrated that NAC is able to inhibit chemically induced oxidative stress and DNA damage [30,31]. The mammalian cytokinesis-block micronucleus assay and the alkaline Comet assay (or single-cell gel electrophoresis assay) were used to investigate the modifying potential of metformin on CumOOH-induced DNA damage. The Comet assay is a simple, sensitive and reliable method for detecting DNA single- and double-strand breaks and alkali-labile sites at the singlecell level. However, the Comet assay is not a direct indicator of the amount of DNA adducts formed. On the other hand, MN is a well-known cytogenetic technique to quantify DNA damage induced by chemical compounds and complex mixtures [23]. A good correlation

Table 2 Effect of metformin on lipid peroxidation (expressed as amount of MDA + 4-HNE in M/106 cells) in 15 M CumOOH-induced oxidative stress in peripheral lymphocytes from elderly and young individuals In vitro treatment M MDA + 4-HNE/106 cells Aging group (n = 10) Basal 0 M Metformin + CumOOH 50 M Metformin + CumOOH 50 M NAC + CumOOH 0.969 1.989 1.363 1.182 0.162 0.270a 0.208a,b 0.182b Young group (n = 10) 0.851 1.768 1.255 1.095 0.088 0.224a 0.178a,b 0.178b

MDA, malondialdehyde, lipid peroxidation product; 4-HNE, 4hydroxynonenal, lipid peroxidation product. a Signicant difference (p < 0.05) with regard to the basal values. b Signicant difference (p < 0.05) with regard to the cells incubated with CumOOH.

I. Onaran et al. / Mutation Research 611 (2006) 18

between micronucleus formation used as a biomarker of DNA damage and oxidative stress in cells was obtained in several experiments [32,33]. It is documented that one of the effects of oxidative stress-induced cytotoxicity in cells is DNA fragmentation caused by apoptosis [3436]. Therefore, this parameter was also determined to see if metformin affects this event associated with DNA damage. From results reported in Tables 1 and 2 and Fig. 1 it is clear that exposure of peripheral lymphocytes in culture to 15 M CumOOH leads to substantial DNA damage and an increase in lipid peroxidation, although there were signicant differences between elderly and younger groups of donors. Using MN and Comet assay techniques, evaluation of DNA damage of 48 h cultures of cells treated with CumOOH and three different concentrations of metformin (1050 M) failed to reveal any signicant differences by comparison with CumOOH alone, in either group. However, the highest dose of metformin showed a partial inhibition of lipid peroxidation as manifested by the decreased concentrations of MDA + 4-HNE. Higher concentrations of metformin in preliminary experiments not only did not protect against DNA damage, but even increased the CumOOH-induced DNA damage (data not shown). In contrast, treatment with 50 M NAC resulted in a marked reduction of DNA damage and in a more effective protection against lipid peroxidation. In the present study, DNA fragmentation showed a similar trend. No signicant difference in DNA fragmentation in lymphocytes from either aged and young individuals was observed between 50 M metformin-treated and untreated cells under oxidative stress. This indicates that metformin did not protect the lymphocytes from apoptotic cell death. The present study conrms previous ndings that NAC provides protection against diverse oxidative insults [3739], which correlates with a reduction in chemical-induced apoptosis. Therefore, the results from all three widely used assays indicate that we were unable to nd dosedependent effects of metformin against oxidative DNA damage in cultured lymphocytes of young and elderly individuals. However, we should be aware of the limitations of this in vitro study, which does not analyze: (a) the protective effects of metformin on DNA damage induced by other free radical-producing agents such as H2 O2 , which has a different action mechanism, and (b) short-term protective effects of metformin against free radical-induced DNA damage. Also, this study was conducted on peripheral blood cells, which may not fully represent changes that occur in all tissues. Furthermore, our experimental conditions may not necessarily reect the in vivo situation.

At present, the basis for the lack of effect of metformin on oxidatively-induced DNA damage remains unknown, although the compound has partial protective effect on lipid peroxidation. Lack of protective effect of metformin may be simply attributed to the fact that it has a partial inhibitory effect on oxidative stress. Studies about the action of metformin on free radicals have shown that it has a direct scavenging effect on the hydroxyl free radicals ( OH), but no direct scavenging effect on O2 free radicals, which have an indirect role in DNA damage [14,40]. Therefore, it may not have a sufciently protective effect against damage in an environment in which we know that CumOOH has the potential to produce free radicals such as OH and O2 [41]. Since NAC is a wellknown O2 scavenger [42], under our experimental conditions it may appear as a protective anti-genotoxic factor. Another possibility is that metformin can modulate the anti-oxidant systems through an increased activity of certain genes involved in the stress response. However, under our experimental conditions it may not modulate DNA repair against CumOOH-induced DNA damage. This possibility could have been checked by alterations in the protocol, as noted by Glei et al. [43]. On the other hand, it has been reported that the concentration of metformin recovered in the cytoplasm after an incubation period of 60 min with Xenopus oocytes was only <0.1% [44,45]. At this low intracellular concentration, it is possible that the expected protective capacity of metformin on the known damaging effects of CumOOH, such as oxidative stress induction, is curtailed. Therefore, a further study is desired to assess whether metformin is able to inhibit DNA damage and apoptosis induced by oxidative compounds in a larger study. Within the limitations of this in vitro study, the following conclusion was drawn: metformin at pharmacological concentrations has no modifying potential against chemically induced DNA damage in cultured human lymphocytes, but has a partial protective effect on lipid peroxidation. References
[1] R.S. Sohal, The free radical hypothesis of aging: an appraisal of the current status, Aging (Milano) 5 (1993) 317. [2] F. Mercuri, L. Quagliaro, A. Ceriello, Oxidative stress evaluation in diabetes, Diabet. Technol. Ther. 2 (2000) 589600. [3] A. Ceriello, Oxidative stress and glycaemic regulation, Metabolism 49 (2000) 2729. [4] S.L. Nuttall, F. Dunne, M.J. Kendall, U. Martin, Age-independent oxidative stress in elderly patients with non-insulin-dependent diabetes mellitus, Quart. J. Med. 92 (1999) 3338. [5] Y. Dincer, T. Akcay, Z. Alademir, H. Ilkova, Assessment of DNA base oxidation and glutathione level in patients with type 2 diabetes, Mutat. Res. 29 (2002) 7581.

I. Onaran et al. / Mutation Research 611 (2006) 18 [6] V. Pitozzi, L. Giovannelli, G. Bardini, C.M. Rotella, P. Dolara, Oxidative DNA damage in peripheral blood cells in type 2 diabetes mellitus: higher vulnerability of polymorphonuclear leukocytes, Mutat. Res. 529 (2003) 129133. [7] B.N. Ames, M.K. Shigenaga, Oxidants are a major contributor to aging, Ann. N.Y. Acad. Sci. 663 (1992) 8596. [8] A. Ceriello, New insights on oxidative stress and diabetic complications may lead to a causal antioxidant therapy, Diabet. Care 26 (2003) 15891596. [9] N.F. Wiernsperger, Metformin: intrinsic vasculoprotective properties, Diabet. Technol. Ther. 2 (2000) 259272. [10] P. Faure, E. Rossini, N. Wiernsperger, M.J. Richard, A. Favier, S. Halimi, An insulin sensitizer improves the free radical defense system potential and insulin sensitivity in high fructose-fed rats, Diabetes 48 (1999) 353357. [11] G. Kanigur-Sultuybek, M. Guven, I. Onaran, V. Tezcan, A. Cenani, H. Hatemi, The effect of metformin on insulin receptors and lipid peroxidation in alloxan and streptozotocin induced diabetes, J. Basic Clin. Physiol. Pharmacol. 6 (1995) 271 280. [12] D. Tessier, P. Maheux, A. Khalil, T. Fulop, Effects of gliclazide versus metformin on the clinical prole and lipid peroxidation markers in type 2 diabetes, Metabolism 48 (1999) 897 903. [13] S. Srividhya, C.V. Anuradha, Metformin improves liver antioxidant potential in rats fed a high-fructose diet, Asia Pac. J. Clin. Nutr. 11 (2002) 319322. [14] D. Bonnefont-Rousselot, B. Raji, S. Walrand, M. Gardes-Albert, D. Jore, J. Peynet, M.P. Vasson, An intracellular modulation of free radical production could contribute to the benecial effects of metformin towards oxidative stress, Metabolism 52 (2003) 586589. [15] S.I. Rattan, DNA damage and repair during cellular aging, Int. Rev. Cytol. 116 (1989) 4788. [16] M.P.A. Hannon-Fletcher, M.J. OKane, K.W. Moles, C. Weatherup, C.R. Barnett, Y.A. Barnett, Levels of peripheral blood cell DNA damage in insulin dependent diabetes mellitus human subjects, Mutat. Res. 460 (2000) 5360. [17] M.C. Chicca, C. Nesti, M. Muzzoli, P. Pasetti, S. Pinamonti, Correlation between age and DNA damage detected by FADU in human peripheral blood lymphocytes, Mutat. Res. 316 (1996) 201208. [18] R.R. Tice, R.B. Setlow, DNA repair and replication in aging organisms and cells, in: J.R. Florini, R.C. Adelman, G.S. Roth (Eds.), CRC Handbook of Biochemistry in Aging, CRC Press, Boca Raton, Florida, 1985, pp. 2548. [19] V.N. Anisimov, L.M. Berstein, P.A. Egormin, T.S. Piskunova, I.G. Popovich, M.A. Zabezhinski, I.G. Kovalenko, T.E. Poroshina, A.V. Semenchenko, M. Re, F. Provinciali, C. Franceschi, Effect of metformin on life span and on the development of spontaneous mammary tumors in HER-2/neu transgenic mice, Exp. Gerontol. 40 (2005) 685693. [20] M. Nagata, Prevention of the manifestation of diabetes in the elderly in the presymptomatic stage, Nippon Ronen Igakkai Zasshi 43 (2006) 8486. [21] K.H. Yuen, K.K. Peh, Simple high-performance liquid chromatographic method for the determination of metformin in human plasma, J. Chromatogr 12 (1998) 243246. [22] M. Fenech, The cytokinesis-block micronucleus technique: a detailed description of the method and its application to genotoxicity studies in human populations, Mutat. Res. 285 (1993) 3544.

[23] M. Fenech, The in vitro micronucleus technique, Mutat. Res. 455 (2000) 8195. [24] N.P. Singh, M.T. McCoy, R.R. Tice, E.L. Schneider, A simple technique for quantitation of low levels of DNA damage in individual cells, Exp. Cell Res. 175 (1988) 184191. [25] R.R. Tice, E. Agurell, D. Anderson, B. Burlinson, A. Hartmann, H. Kobayashi, Y. Miyamae, E. Rojas, J.C. Ryu, Y.F. Sasaki, Single cell gel/comet assay: guidelines for in vitro and in vivo genetic toxicology testing, Environ. Mol. Mutagen. 35 (2000) 206221. [26] D. Anderson, T.W. Yu, B.J. Phillips, P. Schmezer, The effect of various antioxidants and other modifying agents on oxygenradical-generated DNA damage in human lymphocytes in the COMET assay, Mutat. Res. 307 (1994) 261271. [27] E. Diem, S. Ivancsits, H.W. Rudiger, Basal levels of DNA strand breaks in human leukocytes determined by comet assay, J. Toxicol. Environ. Health 65 (2002) 641648. [28] B.G. Taffe, N. Takahashi, T.W. Kensler, R.P. Mason, Generation of free radicals from organic hydroperoxide tumor promoters in isolated mouse keratinocytes. Formation of alkyl and alkoxyl radicals from tert-butyl hydroperoxide and cumene hydroperoxide, J. Biol. Chem. 262 (1987) 1214312149. [29] N.W. Gaikwad, W.J. Bodell, Formation of DNA adducts by microsomal and peroxidase activation of p-cresol: role of quinone methide in DNA adduct formation, Chem. Biol. Interact. 38 (2001) 217229. [30] J.E. Park, J.H. Yang, S.J. Yoon, J.H. Lee, E.S. Yang, J.W. Park, Lipid peroxidation-mediated cytotoxicity and DNA damage in U937 cells, Biochimie 84 (2002) 11991205. [31] R. Reliene, E. Fischer, R.H. Schiestl, Effect of N-acetylcysteine on oxidative DNA damage and the frequency of DNA deletions in atm-decient mice, Cancer Res. 64 (2004) 51485153. [32] C. Mayer, P. Schmezer, R. Freese, M. Mutanen, E. Hietanen, G. Obe, S. Basu, H. Bartsch, Lipid peroxidation status, somatic mutations and micronuclei in peripheral lymphocytes: a case observation on a possible interrelationship, Cancer Lett. 152 (2000) 169173. [33] P. Vineis, Molecular epidemiology: low dose carcinogens and genetic susceptibility, Int. J. Cancer 7 (1997) 13. [34] T. Buttke, P.A. Sandstr om, Oxidative stress as a mediator of apoptosis, Immunol. Today 15 (1994) 710. [35] D. Beyersmann, S. Hechtenberg, Cadmium, gene regulation, and cellular signalling in mammalian cells, Toxicol. Appl. Pharmacol. 144 (1997) 247261. [36] J.E. Park, K.J. Lee, C. Kim, Thermotolerance inhibits various stress-induced apoptosis in NIH3T3 cells, Arch. Pharm. Res. 21 (1998) 4653. [37] K.B. Atkins, I.J. Lodhi, L.L. Hurley, D.B. Hinshaw, NAcetylcysteine and endothelial cell injury by sulfur mustard, J. Appl. Toxicol. (Suppl. 1) (2000) S125S128. [38] W.C. Lee, C.H. Choi, S.H. Cha, H.L. Oh, Y.K. Kim, Role of ERK in hydrogen peroxide-induced cell death of human glioma cells, Neurochem. Res. 30 (2005) 263270. [39] I. Manov, M. Hirsh, T.C. Iancu, Acetaminophen hepatotoxicity and mechanisms of its protection by N-acetylcysteine: a study of Hep3B cells, Exp. Toxicol. Pathol. 53 (2002) 489500. [40] H. Khouri, F. Collin, D. Bonnefont-Rousselot, A. Legrand, D. Jore, M. Gardes-Albert, Radical-induced oxidation of metformin, Eur. J. Biochem. 271 (2004) 47454752. [41] X. Shi, N.S. Dalal, K.S. Kasprzak, Generation of free radicals in reactions of Ni(II)thiol complexes with molecular oxygen and model lipid hydroperoxides, J. Inorg. Biochem. 50 (1993) 211225.

I. Onaran et al. / Mutation Research 611 (2006) 18 [44] D. Detaille, N. Wiernsperger, P. Devos, Cellular and molecular mechanisms involved in insulins potentiation of glycogen synthase activity by metformin, Biochem. Pharmacol. 58 (1999) 14751486. [45] D. Detaille, B. Guigas, X. Leverve, N. Wiernsperger, P. Devos, Obligatory role of membrane events in the regulatory effect of metformin on the respiratory chain function, Biochem. Pharmacol. 63 (2002) 12591272.

[42] O.I. Aruoma, B. Halliwell, B.M. Hoey, J. Butler, The antioxidant action of N-acetylcysteine: its reaction with hydrogen peroxide, hydroxyl radical, superoxide, and hypochlorous acid, Free Radic. Biol. Med. 6 (1989) 593597. [43] M. Glei, U.M. Liegibel, M.N. Ebert, V. Bohm, B.L. PoolZobel, Beta-carotene reduces bleomycin-induced genetic damage in human lymphocytes, Toxicol. Appl. Pharmacol. 179 (2002) 6573.