- Email: [email protected]
Determination of urinary oxidative DNA damage marker 8-hydroxy-2′-deoxyguanosine and the association with cigarette smoking
Talanta 63 (2004) 617–623
Determination of urinary oxidative DNA damage marker 8-hydroxy-2-deoxyguanosine and the association with cigarette smoking Qing-Hong Yao a , Su-Rong Mei b , Qian-Feng Weng a , Pu-duen Zhang a , Qing Yang a , Cai-ying Wu b , Guo-Wang Xu a,∗ a
National Chromatographic R&A Center, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, ZhongShan Road 161, Dalian 116011, PR China b College of Chemistry, Wuhan University, Wuhan 430072, PR China Received 15 June 2003; received in revised form 3 December 2003; accepted 3 December 2003 Available online 1 February 2004
Abstract 8-hydroxy-2 -deoxyguanosine (8OHdG) has been widely used as a biomarker of oxidative DNA damage in both animal models and human studies. To evaluate the effect of cigarette smoking on oxidative stress, we studied the levels of urinary 8OHdG from smokers and non-smokers and investigated the association with cigarette smoking. The urinary 8OHdG concentrations were determinated by capillary electrophoresis with end-column amprometric detection (CE-AD) after a single-step solid phase extraction (SPE), and then quantitatively expressed as a function of creatinine excretion. To increase the concentration sensitivity, a dynamic pH junction was used and the focusing effect was obvious when using 30 mM phosphate (pH 6.50) as sample matrix. The limit of detection is 4.3 nM (signal-to-noise ratio S/N = 3). The relative standard deviation (R.S.D.) was 1.1% for peak current, and 2.3% for migration time. Based on the selected CE-AD method, it was found that the mean value of urinary 8OHdG levels in the smokers significantly higher than that in non-smokers (31.4 ± 18.9 nM versus 14.4 ± 7.6 nM, P = 0.0004; 23.5 ± 21.3 g g− 1 creatinine versus 12.6 ± 13.2 g g− 1 creatinine, P = 0.028). © 2004 Elsevier B.V. All rights reserved. Keywords: 8-hydroxy-2 -deoxyguanosine; Urine; Cigarette smoking; Capillary electrophoresis; End-column amperometric detection
1. Introduction Reactive oxygen species (ROS) generated by-products of normal metabolism and of lipid peroxidation [1] may also be produced by some environmental carcinogens such as X-ray [2], polyphenols in cigarette smoke [3], chewing tobacco components [4], etc. These reactive oxidants that constantly cause irreversible oxidative modifications to DNA have been proposed to be extensive contributors to aging, cancer, and other age-related degenerative diseases [1,5–7]. Among more than 20 DNA oxidized adducts, 8-hydroxy-2 -deoxyguanosine (8OHdG) is a major products formed by hydroxyl radical attack on the guanine residues of ∗ Corresponding author. Tel.: +86-411-3693403; fax: +86-411-3693403. E-mail address: [email protected] (G.-W. Xu).
0039-9140/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2003.12.024
DNA [7,8]. When damaged DNA is repaired, 8OHdG produced is excreted in urine without further metabolism [7,9]. It’s reported that within 8OHdG viral genomes introduce G:C → T:A transitions in vivo [10,11]. Due to its mutagenic potential, 8OHdG has been widely suggested as a biomarker for various cancers [12–14]. The concentration of 8OHdG is also evaluated in smokers and non-smokers [9,15,16] since cigarette smoke contains a large amount of oxygen radical forming substances, such as catechol and hydroquinone that may enhance lung carcinogenesis by free radical-mediated reactions. Loft et al. [9] reported smokers excreted 50% more 8OHdG than non-smokers in 24 h urine, indicating a 50% increased rate of oxidative DNA damage from smoking. Asami et al. [16] found a positive correlation between the number of cigarette smoked per day and the 8OHdG levels in normal lung tissues. But there are overall very few data on oxidative DNA damage in smokers. Thus, we developed
618
Q.-H. Yao et al. / Talanta 63 (2004) 617–623
an optimized method for the analysis of 8OHdG levels in human urine in order to provide a direct evidence for the effect of cigarette smoking on oxidative DNA damage. Until now, several methods have been developed for analysis of urinary 8OHdG including high-performance liquid chromatography–electrochemical detection (HPLC–ECD) [5–7,17,18], gas chromatography (GC) [10,19,20], and immunochemical assays [21]. HPLC–ECD is the most common method for its high sensitivity and good selectivity. However, using HPLC for 8OHdG often employed complicated and time-consuming double or triple column switching [16,18,22] and two or three-step solid phase extraction (SPE) [16,18]. Capillary electrophoresis (CE) has been interested in life sciences due to its high efficiency, high speed, and small sample size requirement [23–27]. Combination of electrochemical detection with capillary electrophoresis (CE–ECD) will be more suitable to analyze complex biological samples. Purines and pyrimidines have been analyzed successfully by utilizing CE–ECD [28,29]. Jin et al. [30] detected bovine serum albumin in a serum sample. Weiss et al. [31] reported 8OHdG could be quantified in urine by off-column end-column amprometric detection (CE-AD) with a detection limit of 50 nM. A recent study of our group [43] developed a simple end-column CE-AD method without complex decoupler, which showed usefulness in analysis of 8OHdG in urine. However, preparation of carbon fiber microcolumn electrode is complex and life span of such an electrode is short. The main goal of this research was to use a carbon fiber microdisk electrode to replace carbon fiber microcolumn electrode to re-setup an end-column CE-AD instrument and use it to evaluate the effect of cigarette smoking on oxidative stress by analyzing urinary 8OHdG content in smokers and non-smokers.
Beijing Institute of New Technology, Beijing, China). Uncoated fused-silica capillaries (25 m i.d.) were obtained from Yongnian Optical Conductive Fiber Plant (Yongnian, China) and flushed with 0.1 M sodium hydroxide solution overnight before use. The high voltage was applied at the injection end of the capillary, while the outlet end also as electrochemical cell was held at ground potential. The electrochemical cell was shielded in a copper box. The amperometric detection mode at a constant potential was performed by using end-column approach with the electrochemical analyzer. Detection was carried out in the amperometric mode using the three-electrode system. 2.2. Reagent and solutions 8-hydroxy-2 -deoxyguanosine (C10 H13 N5 O5 , FW = 283.2) of analytical grade was obtained from Sigma (St. Louis, MO, USA) and used as received. Stock solutions of 8OHdG were stored at 4 ◦ C in refrigerator when not in use. Potassium dihydrogen phosphate (KH2 PO4 ), disodium hydrogen phosphate (Na2 HPO4 ), and disodium tetraborate decahydrate (Na2 B4 O7 ·10H2 O) were purchased from Shanghai Reagent (Shanghai, PR China) and all of them were analytical grade. Methanol (CH3 OH) was HPLC grade obtained from Tedia Company (USA). Bond Elut LCR solid phase extraction columns (C18 /OH, 500 mg, 6 ml) were purchased from Varian (Harbor City, CA, USA). A 12-port vacuum manifold was obtained from J&W Scientific (USA). All solutions were prepared with ultra-pure water from a Milli-Q water system (Millipore Corp., Bedford, MA, USA). 2.3. Operational procedure
2. Experimental 2.1. Apparatus 2.1.1. Cyclic voltammetry An electrochemical analyzer (Model CHI 800, CH Instruments, Austin, TX, USA) was used. In connection with a copper cell, the potentiostatic control of the electrode potential was used by means of a three-electrode system which consists of a carbon fiber microdisk electrode (about 30 carbon fibers with 6 m diameter) as the working electrode, a Pt wire as the auxiliary electrode and a saturated calomel electrode (SCE) as the reference electrode. The reference electrode was connected to the electrolytic cell containing the analyte via a salt bridge filled with the same supporting electrolyte as in the cell. Carbon fiber microdisk electrode was constructed as described in ref. [30]. 2.1.2. CE Electrophoresis in the capillary was driven by a high voltage dc (0–30 kV) power supply (Model NT-9123,
2.3.1. Cyclic voltammetry The carbon fiber microdisk electrode was cleaned in water for 2 min in the supersonic wave cleaner before detection. To obtain a steady cyclic voltammogram, the carbon fiber microdisk electrode was pre-scanned 4–5 times in the CE background electrolyte (BGE). The sweep was firstly carried out on the working electrode in BGE without 8OHdG. The voltammogram was recorded in the potential range of −0.5 to 1.50 V with a scan rate of 100 mV/s. Then, 0.05 mM 8OHdG was added and the voltammograms were recorded. 2.3.2. Collection of urine samples Healthy Chinese men volunteers between the ages of 23 and 70 years (21 smokers and 21 non-smokers) were enrolled in this study. Each donor was asked to answer a self-administered questionnaire regarding age, body weight, any drug therapy, and daily smoking habits. Subjects on medication were excluded. The urine samples were titrated with 1 M HCl to the pH 4–5, then were frozen at −20 ◦ C in refrigerator.
Q.-H. Yao et al. / Talanta 63 (2004) 617–623
619
2.3.3. Sample preparation Before analysis, urine was thawed at room temperature and was centrifuged at 1000×g for 5 min in order to remove precipitates, then was filtered through a 0.2 m micropore filter membrane. A 12-port vacuum manifold supplied the vacuum of the SPE column. The Bond Elut C18 /OH cartridge was preconditioned with 10 ml of methanol and 10 ml of water in turn, and then 2 ml of urine was applied. The column was washed with 5 ml of water. 8OHdG was eluted with 2 ml of 30% methanol. The eluate containing 8OHdG was placed in a water bath at a temperature of 39 ◦ C and evaporated into dryness. The concentrated sample was dissolved in 100 l 30 mM phosphate solution (pH 6.5), thus a 20-fold concentration of the sample was obtained for analysis. 2.3.4. CE The capillary was flushed at the beginning of the day with 1 M NaOH ultra-pure water and BGE for 10 min, respectively. Between runs, the capillary was flushed with 1 M NaOH and ultra-pure water for 2 min, respectively. And then filled with the BGE by means of a syringe. In addition, the electrolyte in the electrochemical cell was also replaced before each run. Before sample separation was carried out, the capillary was equilibrated in BGE at 20 kV for about 10 min until a steady of baseline. Electromigration injection was carried out at 20 kV for 10 s and the electropherogram was recorded. 2.3.5. Activation of the carbon fiber microdisk electrode As the surface of the carbon fiber microdisk electrode was easily vitiated by the sample, some measures should be taken to assure its reproducibility and well-defined electrochemical behavior. The working electrode was polished on an abrasive paper with ultra-fine granularity, and then cleaned in the supersonic cleaner for 2 min before use. A reactivated process was done by means of cyclic voltammetric scanning for 2 min, which was the same as in cyclic voltammetry, and then the detection potential was maintained for 2 min.
Fig. 1. Cyclic voltammogram of 30 mM Na-borate (pH 9.0) (a) without and (b) with 0.05 mM 8OHdG at the carbon fiber microdisk electrode. Scan rate: 100 mV/s.
analyte was investigated. It was found that when Ed was 0.45 V, the working electrode showed a good stability and high reproducibility. The results agreed well with previous report by off-column CE-AD [31]. 3.3. SPE In previous reports, urine sample preparation for detection of 8OHdG often employed two C18 SPE columns [18] or one C18 column followed by silica SPE column for purification [5], which sometimes leaded to a low recovery. So, a single-step SPE method was preferred [33]. In this article, we investigated a single-step SPE method. It was found that the extraction result of C18/OH solid phase (87.5 ± 5.8%, N = 2) was better than that of C18 materials (61.9 ± 6.8%, N = 2). Thus, C18 /OH was chosen as the packing of column, its content of hydroxyl group in the solid phase is 14.87%. Using the optimum SPE method, the recovery of 8OHdG were 87.5 ± 5.0% (N = 3) and 82.7 ± 3.4% (N = 3) for aqueous standards and spiked urine samples, respectively.
3. Results and discussion 3.4. Optimization of separation conditions 3.1. Cyclic voltammogram of 8OHdG Using cyclic voltammetry, the electrochemical behavior of 8OHdG on carbon microdisk electrode was initially investigated in 30 mM Na-borate (pH 9.0). A typical cyclic voltammogram was shown in Fig. 1. In the blank electrolyte solution, no oxidative peak was observed in the range of −0.5 to 1.5 V. After the addition of 8OHdG to the solution, an obvious oxidative peak was observed around 0.35 V. 3.2. Hydrodynamic voltammogram To obtain optimum detection conditions, the effect of working electrode potential (Ed ) on the peak current of the
Optimum electrophoretic conditions should be defined to obtain not only high separating efficiency and a high S/N ratio, but also a good resolution of 8OHdG from sample matrix interferences. 3.4.1. Optimization of the pH of background electrolyte BGE pH affects electroosmotic flow (EOF) as well as ionization of analyte. Therefore, optimization of the pH of BGE is important for separation. We investigated the effect of BGE pH ranging from 8.0 to 10.0 on the peak current (ip ) and theoretical plates (N) showed in Fig. 2. It could be seen that when the pH was 9.0, both column efficiency and peak current were highest. At this pH value, a urine
620
Q.-H. Yao et al. / Talanta 63 (2004) 617–623
spiked urine sample from a non-smoker under the optimized BGE pH condition.
Fig. 2. Effect of BGE pH on (a) ip and (b) N. Electrophoresis conditions: 65 cm × 25 m i.d. capillary; BGE, 30 mM Na-borate; separation voltage, 20 kV; injection, 20 kV, 10 s; detection potential, 0.45 V (vs. SCE).
sample was analyzed, and it was found that the peak of 8OHdG was on the shoulder of its adjacent peak. Therefore, a further adjustment to the BGE pH was needed. When pH of BGE increased to 9.10, a baseline resolution of 8OHdG from matrix interferences was obtained. The reason why the separation of 8OHdG was very sensitive to the pH value may be that the ionization of 8OHdG was easily affected by the pH of environmental solution. So, it is very important to keep a stable pH value of the BGE. But this didn’t mean the pH of BGE need to be adjusted in a same day again and again. Because the BGE solution used in the experiments of a day has been exactly adjusted to pH 9.10 during the preparation it was taken out newly from the BGE storage solution. The BGE in the buffer bottle was replaced after a urine sample was analyzed, In the meantime, the pH of urine samples was adjusted to pH 4–5 before analysis. Therefore, pH value of the BGE is stable. Fig. 3 shows the electropherogram of a
Fig. 3. The electropherogram of a spiked urine extract from a non-smoker under optimized BGE pH condition. Electrophoresis conditions: 85 cm × 25 m i.d. capillary; BGE, 30 mM Na-borate (pH 9.10); separation voltage, 22 kV; injection, 20 kV, 10 s; detection potential, 0.45 V (vs. SCE).
3.4.2. Choice of BGE Previous reports investigated the effect of pH on migration time and theoretical plate and indicated that borate used as running buffer could be successfully used to separate 8OHdG in DNA sample [9]. A recent report [31] found that though zwitterionic buffers exhibit lower electrophoretic currents than borate or phosphate buffers, no resolution of 8OHdG from matrix interferences was observed with the three selected zwitterionic buffers. We investigated the effects of several other buffers, such as Na2 CO3 , Na2 HPO4 , Na-borate, and Na-acetate on the peak current (ip ) and theoretical plates (N). Na-borate showed a highest ip and N, but a lowest noise. So borate buffer is a good BGE for analysis of urinary 8OHdG in CE. 3.4.3. Effect of sample matrix Dynamic pH junction is a novel sample focusing method reported by Chen and co-workers [34,35]. The method was proved valid in increasing the concentration sensitivity for analyzing urinary 8OHdG in the previous report of our group [43]. The effects of focusing of 8OHdG in different pH of BGE and different pH of sample matrix were also observed in this experiment. Optimal focusing was achieved with a high column efficiency (N > 2.9 × 105 ) when the pH of the sample solution was 6.5 and the pH of BGE was 9.00. Therefore, 30 mM phosphate solution (pH 6.5) was used as sample matrix to get the highest sensitivity. However, at pH 9.00 of BGE 8OHdG couldn’t be baseline-separated from other urine components (Section 3.4.1), thus, pH 9.10 was selected as acidity of BGE when analyzing urine sample with dynamic pH junction. 3.5. Reproducibility, limit of detection, and linear range The reproducibility of the peak current and the migration time was determined by repeatedly (n = 8) injecting 1 M 8OHdG standard into the capillary under the selected conditions. The relative standard deviation (R.S.D.) was found to be 2.3% for peak current, and 1.1% for migration time. The calibration curve exhibits a good linear behavior over the concentration range from 20 nM to10 M with correlation coefficient greater than 0.99. The limit of detection for an aqueous standard of 8OHdG was 4.3 nM (signal-to-noise ratio S/N = 3), which was 10-folds lower than that of the off-column CE-AD method reported by Weiss [31] (50 nM) and almost five-folds lower than that of our previous study [43] (20 nM) for analyzing 8OHdG. Since the concentration of a urine sample was increased 20-folds after the SPE step, the limit of detection for a urine sample was 0.22 nM. And the lower limit of the linear range also can be reduced to 1 nM that is enough to determine the low concentration of urinary 8OHdG.
Q.-H. Yao et al. / Talanta 63 (2004) 617–623 Table 1 Descriptive data and 8OHdG levels of the study groupsa
Number Age (years) Body weight (kg) Cigarettes per day 8OHdG (nM) 8OHdG ( g g−1 Cr) a
Table 3 Effect of smoking habits on 8OHdG content in urinea
Smokers
Non-smokers
21 48.2 69.9 15.5 31.4 23.5
21 30.7 67.7 – 14.4 12.6
± ± ± ± ±
12.5 13.3 5.5 18.9 21.3
621
± 9.3 ± 10.5 ± 7.6 ± 13.2
Data in the table are mean ± S.D., Cr: creatinine.
In this new method, the working electrode was a carbon fibre microdisk electrode and was laid at the end of the separation capillary rather than inserted into it as in a carbon fibre microcolumn electrode. Such a placement not only avoided interference of electric field brought from the high pressure applied on the capillary and greatly decreased the noise on the electrode, but also was more convenient for the adjustment to make the capillary and the electrode in a line. In addition, the electrode prepared in this study also showed a better stability, and the life increased one fold than the microcolumn electrode used in our former study [43]. 3.6. Analysis of urinary 8OHdG and study of the association with cigarette smoking To show the usefulness of the optimized method and provide a direct evidence for the association between cigarette smoking and oxidative stress, we evaluated urinary 8OHdG levels in smokers and non-smokers. The peak of 8OHdG was identified by migration time and spiking with 8OHdG standard. The levels of urinary 8OHdG were quantified by the five-point calibration and then were transformed into microgram per gram (g g−1 ) creatinine. Determination of urinary creatinine level [32] was performed with the Beckman P/ACE MDQ capillary electrophoretic system (Beckman Coulter, USA). Before analysis, urine was
Group Smokers
N
Less than 10 cigarettes per day More than 10 cigarettes per day
Non-smokers a
8OHdG (nM)
8OHdG ( g g−1 Cr)
7
28.6 ± 15.8
18.2 ± 12.7
14
32.8 ± 20.7
26.9 ± 24.5
21
14.4 ± 7.6
12.6 ± 13.2
Data in the table are mean ± S.D.
thawed at room temperature and diluted for five-folds, and then introduced into the capillary. The separation was carried out with 30 mM phosphate buffer (pH 6.0) at 20 kV and detected with diode array detector at 245 nm. Table 1 summarized levels of 8OHdG, age, sex, and body weight, smoking status, and cigarette consumption of every volunteer. The urinary 8OHdG excretion presented here is in the range of others reported previously showed in Table 2. It is found that there exist differences in some degrees among the data of different laboratories. This may be caused by different race of subjects chosen or the systemic difference brought from the method employed in a research. In one-way ANOVA there was significant difference both in the urinary 8OHdG excretion (P = 0.0004) and in the ratio of 8OHdG-to-creatinine (P = 0.028) between smokers and non-smokers. This result is in agreement with others in the literature [16,39], indicating that cigarette smoking induces oxidative damage to DNA, although the effect was not observed in all smokers. To further know the association of urinary 8OHdG with the smoking amount, smokers were divided into two groups according to the number of cigarettes smoked per day. As reported in Table 3, urinary 8OHdG contents in subjects smoking less than 10 cigarettes per day have no significant difference from those in non-smokers (P = 0.22), but significantly higher levels (P = 0.003) were observed
Table 2 Urinary 8OHdG levels in healthy controls in literaturesa Reference
Age (years)
Smoker status
Mean ± S.D.
N
Methods
[31] [43] [36]
23–43 – 33 ± 10 24–55
[41]
–
[42] [39] [40] [37]
– – – 22–60
[18]
25–52
42 ± 26.9 nM 13.51 ± 5.08 nM 26.7 ± 11.7 nmol/24 h 18.6 ± 7.8 nmol/24 h 20.1 ± 7.3 nmol/24 h 18.4 ± 5.7 nmol/24 h 24.7 ± 11.1 nmol/24 h 19.1 ± 11.1 nmol/24 h 14.9 ± 7.8 nmol/24 h 11.9 ± 7.3 g g−1 Cr 15.2 ± 5.4 g g−1 Cr 7.41 ± 1.47 g g−1 Cr 3.81 ± 1.93 g g−1 Cr 6.76 ± 4.71 g g−1 Cr
8 9 12 21 30 30 44 23 27 10 22 7 6 60
Off column CE-AD End-cloumn CE-AD HPLC–ECD
[38]
– NS S NS S NS S NS – – – S NS –
a
S: smoker; NS: non-smoker; Cr: creatinine; N: subject number.
HPLC–ECD HPLC–ECD HPLC–ECD ELISA – HPLC–ECD HPLC–ECD
622
Q.-H. Yao et al. / Talanta 63 (2004) 617–623
Acknowledgements This work has been supported by the Knowledge Innovation Program of the Chinese Academy of Sciences and the 863 project from State Ministry of Science and Technology of China (2003AA 223061).
References Fig. 4. Correlation between age and 8OHdG levels in urine.
in the ratio of urinary 8OHdG-to-creatinine of the two groups. There was a significant difference both in the urinary 8OHdG excretion (P = 0.0007) and in the ratio of urinary 8OHdG-to-creatinine (P = 0.002) between subjects smoking more than 10 cigarettes per day and non-smokers. From above results, we observed that cigarette smoking has a strong effect on 8OHdG content in urine, especially for those heavy smokers. There exists a significant correlation between age and 8OHdG levels in urine from totally studied crowd including smokers and non-smokers (r = 0.330, P = 0.032) (Fig. 4). This observation confirms the fact that oxidative DNA damage increases with age. No significant correlation was found between body mass and urinary 8OHdG levels from the crowd including smokers and non-smokers (r = 0.249, P = 0.11). 8OHdG is known as a sensitive biomarker of oxidative DNA damage also of oxidative stress. The high urinary 8OHdG levels in the smokers with heavy smoking habits investigated in this study suggested that smoking has an important role in increasing oxidative stress in human body. Although the biochemical–physiological basis is unknown, it may be related to smoke constituents, which include or generate reactive oxygen species or consume antioxidants or enhance the effect of smoking on the metabolic rate.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]
4. Conclusions In summary, the present study developed a CE-AD method to determine 8OHdG in urine. The electrochemical behaviors of 8OHdG on carbon fiber microdisk electrode were investigated in detail. And the effects of CE conditions, such as pH value of BGE solution, kinds of BGE and sample matrix effect, on the separation were discussed in the experiment. The optimized method showed a high sensitivity, a good selectivity, and a large linear range which made it suitable to analyze 8OHdG in urine. With the optimized method the effect of cigarette smoking on the extent of 8OHdG in urine was assessed, significantly positive relation was found. Further investigations are needed to understand the exact biologic and pathologic significance of cigarette smoking on oxidative DNA damage.
[22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37]
B.N. Ames, Free Rad. Res. Commun. 7 (3–6) (1989) 121. H. Kasai, H. Tanooka, S. Nishimura, Gann 75 (1984) 565. P. Leanderson, C. Tagesson, Chem. Biol. Interact. 75 (1990) 71. U.J. Nair, R.A. Floyd, V. Bussachini, M. Friesen, H. Bartsch, Chem. Biol. Interact. 63 (1987) 157. M.K. Shigenaga, J.W. Park, K.C. Cundy, C.J. Gimeno, B.N. Ames, Methods Enzymol. 186 (1990) 17. M.K. Shigenaga, E.N. Aboujaoude, Q. Chen, B.N. Ames, Methods Enzymol. 234 (1994) 521. M.K. Shigenaga, B.N. Ames, Free Rad. Bio. Med. 10 (1991) 211. R.A. Floyd, J.J. Watson, P.K. Wong, D.H. Altmiller, R.C. Richard, Free Rad. Res. Commun. 1 (1986) 163. S. Loft, K. Vistisen, M. Ewertz, A. Tjnne´øland, H.E. Poulsen, Carcinogenesis 13 (1992) 2241. S.R. Mei, G.W. Xu, J. Xing, C.Y. Wu, Anal. Sci. 17 (2001) 779. S.R. Mei, G.W. Xu, C.Y. Wu, Anal. Lett. 34 (2001) 2063. M. Erhola, S. Toyokuni, H. Ochi, K. Uchida, H. Alho, FEBS Lett. 409 (1997) 287. C. Tagesson, M. Källberg, C. Klinterberg, H. Starkhammar, Eur. J. Cancer 31A (1995) 934. Y. Takara, H. Ken-ichi, T. Takaya, K. Hiroshi, U. Mamoru, H.J. Hider, Obstet. Gynaecol. Res. 22 (4) (1996) 359. C.G. Fraga, P.A. Motchnik, A.J. Wyrobek, D.M. Rempel, B.N. Ames, Mutat. Res. 351 (2) (1996) 199. S. Asami, H. Manabe, J. Miyake, Y. Tsurudome, T. Hirano, R. Yamaguchi, H. Itoh, H. Kasai, Carcinogenesis 18 (9) (1997) 1763. M. Nakajima, T. Takeuchi, K. Morimoto, Carcinogenesis 17 (4) (1996) 787. D. Germadnik, A. Pilger, H.W. Rüdiger, J. Chromatogr. B 689 (1997) 399. A. Teixeira, H. Gommers-Ampt, G. Werken, J. Westra, J. Stavenuiter, Anal. Biochem. 214 (1993) 474. S.R. Mei, G.W. Xu, C.Y. Wu, Chin. J. Anal. Chem. 29 (2001) 1394. B. Yin, R. Whyatt, F. Perera, M. Randall, T. Cooper, R. Santella, Free Rad. Biol. Med. 18 (6) (1995) 1023. K. Vistisen, H.E. Poulsen, S. Loft, Carcinogenesis 13 (1992) 1561. Q. Weng, W. Jin, Anal. Chim. Acta 478 (2003) 199. A. Pluym, W. Ael, M. Smet, Trends Anal. Chem. 11 (1992) 27. K. Altria, M. Kelly, B. Clark, Trends Anal. Chem. 17 (1998) 204. K. Altria, M. Kelly, B. Clark, Trends Anal. Chem. 17 (1998) 214. Q. Dong, W. Jin, Electrophoresis 22 (2001) 2786. H. Lin, D. Xu, H. Chen, J. Chromatogr. A 760 (1997) 227. W. Jin, H. Wei, X. Zhao, Electroanalysis 9 (1997) 770. W. Jin, Q. Weng, J. Wu, Anal. Chim. Acta 342 (1997) 67. D. Weiss, C. Lunte, Electrophoresis 21 (2000) 2080. Y. Zheng, Y. Zhang, D. Liu, X. Guo, S. Mei, J. Xiong, H. Kong, C. Zhang, G. Xu, Chem. J. Chin. Univ. 22 (6) (2001) 912. T. Renner, T. Fechner, G. Scherer, J. Chromatogr. B. 738 (2000) 311. P. Britz-Mckibbin, D.D.Y. Chen, Anal. Chem. 72 (2000) 1242. P. Britz-Mckibbin, G. Bebault, D.D.Y. Chen, Anal. Chem. 72 (2000) 1729. S. Loft, A. Astrup, B. Buemann, H.E. Poulsen, FASEB J. 8 (1994) 534. J. Suzuki, Y. Inoue, S. Suzuki, Free Rad. Biol. Med. 18 (1995) 431.
Q.-H. Yao et al. / Talanta 63 (2004) 617–623 [38] S. Pourcelot, H. Faure, F. Firoozi, V. Ducros, M. Tripier, J. Hee, J. Cadet, A. Favier, Free Rad. Res. 30 (1999) 173. [39] M. Honda, Y. Yamada, M. Tomonaga, H. Ichinose, S. Kamihira, Leuk. Res. 24 (2000) 461. [40] H. Miyake, H. Eto, Y. Takechi, S. Kamidono, I. Hara, BJU Int. 91 (7) (2003) 657.
623
[41] A. Pilger, S. Ivancsits, D. Germadnik, H.W. Rudiger, J. Chromatogr. B. Analyt. Technol. Biomed. Life Sci. 778 (1–2) (2002) 393. [42] C. Tagesson, M. Kallberg, C. Klintenberg, H. Starkhammar, Eur. J. Cancer 31A (1995) 934. [43] S.R. Mei, Q.H. Yao, L.S. Cai, J. Xing, G.W. Xu, C.Y. Wu, Electrophoresis 24 (2003) 1411.
Determination of urinary oxidative DNA damage marker 8-hydroxy-2-deoxyguanosine and the association with cigarette smoking Qing-Hong Yao a , Su-Rong Mei b , Qian-Feng Weng a , Pu-duen Zhang a , Qing Yang a , Cai-ying Wu b , Guo-Wang Xu a,∗ a
National Chromatographic R&A Center, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, ZhongShan Road 161, Dalian 116011, PR China b College of Chemistry, Wuhan University, Wuhan 430072, PR China Received 15 June 2003; received in revised form 3 December 2003; accepted 3 December 2003 Available online 1 February 2004
Abstract 8-hydroxy-2 -deoxyguanosine (8OHdG) has been widely used as a biomarker of oxidative DNA damage in both animal models and human studies. To evaluate the effect of cigarette smoking on oxidative stress, we studied the levels of urinary 8OHdG from smokers and non-smokers and investigated the association with cigarette smoking. The urinary 8OHdG concentrations were determinated by capillary electrophoresis with end-column amprometric detection (CE-AD) after a single-step solid phase extraction (SPE), and then quantitatively expressed as a function of creatinine excretion. To increase the concentration sensitivity, a dynamic pH junction was used and the focusing effect was obvious when using 30 mM phosphate (pH 6.50) as sample matrix. The limit of detection is 4.3 nM (signal-to-noise ratio S/N = 3). The relative standard deviation (R.S.D.) was 1.1% for peak current, and 2.3% for migration time. Based on the selected CE-AD method, it was found that the mean value of urinary 8OHdG levels in the smokers significantly higher than that in non-smokers (31.4 ± 18.9 nM versus 14.4 ± 7.6 nM, P = 0.0004; 23.5 ± 21.3 g g− 1 creatinine versus 12.6 ± 13.2 g g− 1 creatinine, P = 0.028). © 2004 Elsevier B.V. All rights reserved. Keywords: 8-hydroxy-2 -deoxyguanosine; Urine; Cigarette smoking; Capillary electrophoresis; End-column amperometric detection
1. Introduction Reactive oxygen species (ROS) generated by-products of normal metabolism and of lipid peroxidation [1] may also be produced by some environmental carcinogens such as X-ray [2], polyphenols in cigarette smoke [3], chewing tobacco components [4], etc. These reactive oxidants that constantly cause irreversible oxidative modifications to DNA have been proposed to be extensive contributors to aging, cancer, and other age-related degenerative diseases [1,5–7]. Among more than 20 DNA oxidized adducts, 8-hydroxy-2 -deoxyguanosine (8OHdG) is a major products formed by hydroxyl radical attack on the guanine residues of ∗ Corresponding author. Tel.: +86-411-3693403; fax: +86-411-3693403. E-mail address: [email protected] (G.-W. Xu).
0039-9140/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2003.12.024
DNA [7,8]. When damaged DNA is repaired, 8OHdG produced is excreted in urine without further metabolism [7,9]. It’s reported that within 8OHdG viral genomes introduce G:C → T:A transitions in vivo [10,11]. Due to its mutagenic potential, 8OHdG has been widely suggested as a biomarker for various cancers [12–14]. The concentration of 8OHdG is also evaluated in smokers and non-smokers [9,15,16] since cigarette smoke contains a large amount of oxygen radical forming substances, such as catechol and hydroquinone that may enhance lung carcinogenesis by free radical-mediated reactions. Loft et al. [9] reported smokers excreted 50% more 8OHdG than non-smokers in 24 h urine, indicating a 50% increased rate of oxidative DNA damage from smoking. Asami et al. [16] found a positive correlation between the number of cigarette smoked per day and the 8OHdG levels in normal lung tissues. But there are overall very few data on oxidative DNA damage in smokers. Thus, we developed
618
Q.-H. Yao et al. / Talanta 63 (2004) 617–623
an optimized method for the analysis of 8OHdG levels in human urine in order to provide a direct evidence for the effect of cigarette smoking on oxidative DNA damage. Until now, several methods have been developed for analysis of urinary 8OHdG including high-performance liquid chromatography–electrochemical detection (HPLC–ECD) [5–7,17,18], gas chromatography (GC) [10,19,20], and immunochemical assays [21]. HPLC–ECD is the most common method for its high sensitivity and good selectivity. However, using HPLC for 8OHdG often employed complicated and time-consuming double or triple column switching [16,18,22] and two or three-step solid phase extraction (SPE) [16,18]. Capillary electrophoresis (CE) has been interested in life sciences due to its high efficiency, high speed, and small sample size requirement [23–27]. Combination of electrochemical detection with capillary electrophoresis (CE–ECD) will be more suitable to analyze complex biological samples. Purines and pyrimidines have been analyzed successfully by utilizing CE–ECD [28,29]. Jin et al. [30] detected bovine serum albumin in a serum sample. Weiss et al. [31] reported 8OHdG could be quantified in urine by off-column end-column amprometric detection (CE-AD) with a detection limit of 50 nM. A recent study of our group [43] developed a simple end-column CE-AD method without complex decoupler, which showed usefulness in analysis of 8OHdG in urine. However, preparation of carbon fiber microcolumn electrode is complex and life span of such an electrode is short. The main goal of this research was to use a carbon fiber microdisk electrode to replace carbon fiber microcolumn electrode to re-setup an end-column CE-AD instrument and use it to evaluate the effect of cigarette smoking on oxidative stress by analyzing urinary 8OHdG content in smokers and non-smokers.
Beijing Institute of New Technology, Beijing, China). Uncoated fused-silica capillaries (25 m i.d.) were obtained from Yongnian Optical Conductive Fiber Plant (Yongnian, China) and flushed with 0.1 M sodium hydroxide solution overnight before use. The high voltage was applied at the injection end of the capillary, while the outlet end also as electrochemical cell was held at ground potential. The electrochemical cell was shielded in a copper box. The amperometric detection mode at a constant potential was performed by using end-column approach with the electrochemical analyzer. Detection was carried out in the amperometric mode using the three-electrode system. 2.2. Reagent and solutions 8-hydroxy-2 -deoxyguanosine (C10 H13 N5 O5 , FW = 283.2) of analytical grade was obtained from Sigma (St. Louis, MO, USA) and used as received. Stock solutions of 8OHdG were stored at 4 ◦ C in refrigerator when not in use. Potassium dihydrogen phosphate (KH2 PO4 ), disodium hydrogen phosphate (Na2 HPO4 ), and disodium tetraborate decahydrate (Na2 B4 O7 ·10H2 O) were purchased from Shanghai Reagent (Shanghai, PR China) and all of them were analytical grade. Methanol (CH3 OH) was HPLC grade obtained from Tedia Company (USA). Bond Elut LCR solid phase extraction columns (C18 /OH, 500 mg, 6 ml) were purchased from Varian (Harbor City, CA, USA). A 12-port vacuum manifold was obtained from J&W Scientific (USA). All solutions were prepared with ultra-pure water from a Milli-Q water system (Millipore Corp., Bedford, MA, USA). 2.3. Operational procedure
2. Experimental 2.1. Apparatus 2.1.1. Cyclic voltammetry An electrochemical analyzer (Model CHI 800, CH Instruments, Austin, TX, USA) was used. In connection with a copper cell, the potentiostatic control of the electrode potential was used by means of a three-electrode system which consists of a carbon fiber microdisk electrode (about 30 carbon fibers with 6 m diameter) as the working electrode, a Pt wire as the auxiliary electrode and a saturated calomel electrode (SCE) as the reference electrode. The reference electrode was connected to the electrolytic cell containing the analyte via a salt bridge filled with the same supporting electrolyte as in the cell. Carbon fiber microdisk electrode was constructed as described in ref. [30]. 2.1.2. CE Electrophoresis in the capillary was driven by a high voltage dc (0–30 kV) power supply (Model NT-9123,
2.3.1. Cyclic voltammetry The carbon fiber microdisk electrode was cleaned in water for 2 min in the supersonic wave cleaner before detection. To obtain a steady cyclic voltammogram, the carbon fiber microdisk electrode was pre-scanned 4–5 times in the CE background electrolyte (BGE). The sweep was firstly carried out on the working electrode in BGE without 8OHdG. The voltammogram was recorded in the potential range of −0.5 to 1.50 V with a scan rate of 100 mV/s. Then, 0.05 mM 8OHdG was added and the voltammograms were recorded. 2.3.2. Collection of urine samples Healthy Chinese men volunteers between the ages of 23 and 70 years (21 smokers and 21 non-smokers) were enrolled in this study. Each donor was asked to answer a self-administered questionnaire regarding age, body weight, any drug therapy, and daily smoking habits. Subjects on medication were excluded. The urine samples were titrated with 1 M HCl to the pH 4–5, then were frozen at −20 ◦ C in refrigerator.
Q.-H. Yao et al. / Talanta 63 (2004) 617–623
619
2.3.3. Sample preparation Before analysis, urine was thawed at room temperature and was centrifuged at 1000×g for 5 min in order to remove precipitates, then was filtered through a 0.2 m micropore filter membrane. A 12-port vacuum manifold supplied the vacuum of the SPE column. The Bond Elut C18 /OH cartridge was preconditioned with 10 ml of methanol and 10 ml of water in turn, and then 2 ml of urine was applied. The column was washed with 5 ml of water. 8OHdG was eluted with 2 ml of 30% methanol. The eluate containing 8OHdG was placed in a water bath at a temperature of 39 ◦ C and evaporated into dryness. The concentrated sample was dissolved in 100 l 30 mM phosphate solution (pH 6.5), thus a 20-fold concentration of the sample was obtained for analysis. 2.3.4. CE The capillary was flushed at the beginning of the day with 1 M NaOH ultra-pure water and BGE for 10 min, respectively. Between runs, the capillary was flushed with 1 M NaOH and ultra-pure water for 2 min, respectively. And then filled with the BGE by means of a syringe. In addition, the electrolyte in the electrochemical cell was also replaced before each run. Before sample separation was carried out, the capillary was equilibrated in BGE at 20 kV for about 10 min until a steady of baseline. Electromigration injection was carried out at 20 kV for 10 s and the electropherogram was recorded. 2.3.5. Activation of the carbon fiber microdisk electrode As the surface of the carbon fiber microdisk electrode was easily vitiated by the sample, some measures should be taken to assure its reproducibility and well-defined electrochemical behavior. The working electrode was polished on an abrasive paper with ultra-fine granularity, and then cleaned in the supersonic cleaner for 2 min before use. A reactivated process was done by means of cyclic voltammetric scanning for 2 min, which was the same as in cyclic voltammetry, and then the detection potential was maintained for 2 min.
Fig. 1. Cyclic voltammogram of 30 mM Na-borate (pH 9.0) (a) without and (b) with 0.05 mM 8OHdG at the carbon fiber microdisk electrode. Scan rate: 100 mV/s.
analyte was investigated. It was found that when Ed was 0.45 V, the working electrode showed a good stability and high reproducibility. The results agreed well with previous report by off-column CE-AD [31]. 3.3. SPE In previous reports, urine sample preparation for detection of 8OHdG often employed two C18 SPE columns [18] or one C18 column followed by silica SPE column for purification [5], which sometimes leaded to a low recovery. So, a single-step SPE method was preferred [33]. In this article, we investigated a single-step SPE method. It was found that the extraction result of C18/OH solid phase (87.5 ± 5.8%, N = 2) was better than that of C18 materials (61.9 ± 6.8%, N = 2). Thus, C18 /OH was chosen as the packing of column, its content of hydroxyl group in the solid phase is 14.87%. Using the optimum SPE method, the recovery of 8OHdG were 87.5 ± 5.0% (N = 3) and 82.7 ± 3.4% (N = 3) for aqueous standards and spiked urine samples, respectively.
3. Results and discussion 3.4. Optimization of separation conditions 3.1. Cyclic voltammogram of 8OHdG Using cyclic voltammetry, the electrochemical behavior of 8OHdG on carbon microdisk electrode was initially investigated in 30 mM Na-borate (pH 9.0). A typical cyclic voltammogram was shown in Fig. 1. In the blank electrolyte solution, no oxidative peak was observed in the range of −0.5 to 1.5 V. After the addition of 8OHdG to the solution, an obvious oxidative peak was observed around 0.35 V. 3.2. Hydrodynamic voltammogram To obtain optimum detection conditions, the effect of working electrode potential (Ed ) on the peak current of the
Optimum electrophoretic conditions should be defined to obtain not only high separating efficiency and a high S/N ratio, but also a good resolution of 8OHdG from sample matrix interferences. 3.4.1. Optimization of the pH of background electrolyte BGE pH affects electroosmotic flow (EOF) as well as ionization of analyte. Therefore, optimization of the pH of BGE is important for separation. We investigated the effect of BGE pH ranging from 8.0 to 10.0 on the peak current (ip ) and theoretical plates (N) showed in Fig. 2. It could be seen that when the pH was 9.0, both column efficiency and peak current were highest. At this pH value, a urine
620
Q.-H. Yao et al. / Talanta 63 (2004) 617–623
spiked urine sample from a non-smoker under the optimized BGE pH condition.
Fig. 2. Effect of BGE pH on (a) ip and (b) N. Electrophoresis conditions: 65 cm × 25 m i.d. capillary; BGE, 30 mM Na-borate; separation voltage, 20 kV; injection, 20 kV, 10 s; detection potential, 0.45 V (vs. SCE).
sample was analyzed, and it was found that the peak of 8OHdG was on the shoulder of its adjacent peak. Therefore, a further adjustment to the BGE pH was needed. When pH of BGE increased to 9.10, a baseline resolution of 8OHdG from matrix interferences was obtained. The reason why the separation of 8OHdG was very sensitive to the pH value may be that the ionization of 8OHdG was easily affected by the pH of environmental solution. So, it is very important to keep a stable pH value of the BGE. But this didn’t mean the pH of BGE need to be adjusted in a same day again and again. Because the BGE solution used in the experiments of a day has been exactly adjusted to pH 9.10 during the preparation it was taken out newly from the BGE storage solution. The BGE in the buffer bottle was replaced after a urine sample was analyzed, In the meantime, the pH of urine samples was adjusted to pH 4–5 before analysis. Therefore, pH value of the BGE is stable. Fig. 3 shows the electropherogram of a
Fig. 3. The electropherogram of a spiked urine extract from a non-smoker under optimized BGE pH condition. Electrophoresis conditions: 85 cm × 25 m i.d. capillary; BGE, 30 mM Na-borate (pH 9.10); separation voltage, 22 kV; injection, 20 kV, 10 s; detection potential, 0.45 V (vs. SCE).
3.4.2. Choice of BGE Previous reports investigated the effect of pH on migration time and theoretical plate and indicated that borate used as running buffer could be successfully used to separate 8OHdG in DNA sample [9]. A recent report [31] found that though zwitterionic buffers exhibit lower electrophoretic currents than borate or phosphate buffers, no resolution of 8OHdG from matrix interferences was observed with the three selected zwitterionic buffers. We investigated the effects of several other buffers, such as Na2 CO3 , Na2 HPO4 , Na-borate, and Na-acetate on the peak current (ip ) and theoretical plates (N). Na-borate showed a highest ip and N, but a lowest noise. So borate buffer is a good BGE for analysis of urinary 8OHdG in CE. 3.4.3. Effect of sample matrix Dynamic pH junction is a novel sample focusing method reported by Chen and co-workers [34,35]. The method was proved valid in increasing the concentration sensitivity for analyzing urinary 8OHdG in the previous report of our group [43]. The effects of focusing of 8OHdG in different pH of BGE and different pH of sample matrix were also observed in this experiment. Optimal focusing was achieved with a high column efficiency (N > 2.9 × 105 ) when the pH of the sample solution was 6.5 and the pH of BGE was 9.00. Therefore, 30 mM phosphate solution (pH 6.5) was used as sample matrix to get the highest sensitivity. However, at pH 9.00 of BGE 8OHdG couldn’t be baseline-separated from other urine components (Section 3.4.1), thus, pH 9.10 was selected as acidity of BGE when analyzing urine sample with dynamic pH junction. 3.5. Reproducibility, limit of detection, and linear range The reproducibility of the peak current and the migration time was determined by repeatedly (n = 8) injecting 1 M 8OHdG standard into the capillary under the selected conditions. The relative standard deviation (R.S.D.) was found to be 2.3% for peak current, and 1.1% for migration time. The calibration curve exhibits a good linear behavior over the concentration range from 20 nM to10 M with correlation coefficient greater than 0.99. The limit of detection for an aqueous standard of 8OHdG was 4.3 nM (signal-to-noise ratio S/N = 3), which was 10-folds lower than that of the off-column CE-AD method reported by Weiss [31] (50 nM) and almost five-folds lower than that of our previous study [43] (20 nM) for analyzing 8OHdG. Since the concentration of a urine sample was increased 20-folds after the SPE step, the limit of detection for a urine sample was 0.22 nM. And the lower limit of the linear range also can be reduced to 1 nM that is enough to determine the low concentration of urinary 8OHdG.
Q.-H. Yao et al. / Talanta 63 (2004) 617–623 Table 1 Descriptive data and 8OHdG levels of the study groupsa
Number Age (years) Body weight (kg) Cigarettes per day 8OHdG (nM) 8OHdG ( g g−1 Cr) a
Table 3 Effect of smoking habits on 8OHdG content in urinea
Smokers
Non-smokers
21 48.2 69.9 15.5 31.4 23.5
21 30.7 67.7 – 14.4 12.6
± ± ± ± ±
12.5 13.3 5.5 18.9 21.3
621
± 9.3 ± 10.5 ± 7.6 ± 13.2
Data in the table are mean ± S.D., Cr: creatinine.
In this new method, the working electrode was a carbon fibre microdisk electrode and was laid at the end of the separation capillary rather than inserted into it as in a carbon fibre microcolumn electrode. Such a placement not only avoided interference of electric field brought from the high pressure applied on the capillary and greatly decreased the noise on the electrode, but also was more convenient for the adjustment to make the capillary and the electrode in a line. In addition, the electrode prepared in this study also showed a better stability, and the life increased one fold than the microcolumn electrode used in our former study [43]. 3.6. Analysis of urinary 8OHdG and study of the association with cigarette smoking To show the usefulness of the optimized method and provide a direct evidence for the association between cigarette smoking and oxidative stress, we evaluated urinary 8OHdG levels in smokers and non-smokers. The peak of 8OHdG was identified by migration time and spiking with 8OHdG standard. The levels of urinary 8OHdG were quantified by the five-point calibration and then were transformed into microgram per gram (g g−1 ) creatinine. Determination of urinary creatinine level [32] was performed with the Beckman P/ACE MDQ capillary electrophoretic system (Beckman Coulter, USA). Before analysis, urine was
Group Smokers
N
Less than 10 cigarettes per day More than 10 cigarettes per day
Non-smokers a
8OHdG (nM)
8OHdG ( g g−1 Cr)
7
28.6 ± 15.8
18.2 ± 12.7
14
32.8 ± 20.7
26.9 ± 24.5
21
14.4 ± 7.6
12.6 ± 13.2
Data in the table are mean ± S.D.
thawed at room temperature and diluted for five-folds, and then introduced into the capillary. The separation was carried out with 30 mM phosphate buffer (pH 6.0) at 20 kV and detected with diode array detector at 245 nm. Table 1 summarized levels of 8OHdG, age, sex, and body weight, smoking status, and cigarette consumption of every volunteer. The urinary 8OHdG excretion presented here is in the range of others reported previously showed in Table 2. It is found that there exist differences in some degrees among the data of different laboratories. This may be caused by different race of subjects chosen or the systemic difference brought from the method employed in a research. In one-way ANOVA there was significant difference both in the urinary 8OHdG excretion (P = 0.0004) and in the ratio of 8OHdG-to-creatinine (P = 0.028) between smokers and non-smokers. This result is in agreement with others in the literature [16,39], indicating that cigarette smoking induces oxidative damage to DNA, although the effect was not observed in all smokers. To further know the association of urinary 8OHdG with the smoking amount, smokers were divided into two groups according to the number of cigarettes smoked per day. As reported in Table 3, urinary 8OHdG contents in subjects smoking less than 10 cigarettes per day have no significant difference from those in non-smokers (P = 0.22), but significantly higher levels (P = 0.003) were observed
Table 2 Urinary 8OHdG levels in healthy controls in literaturesa Reference
Age (years)
Smoker status
Mean ± S.D.
N
Methods
[31] [43] [36]
23–43 – 33 ± 10 24–55
[41]
–
[42] [39] [40] [37]
– – – 22–60
[18]
25–52
42 ± 26.9 nM 13.51 ± 5.08 nM 26.7 ± 11.7 nmol/24 h 18.6 ± 7.8 nmol/24 h 20.1 ± 7.3 nmol/24 h 18.4 ± 5.7 nmol/24 h 24.7 ± 11.1 nmol/24 h 19.1 ± 11.1 nmol/24 h 14.9 ± 7.8 nmol/24 h 11.9 ± 7.3 g g−1 Cr 15.2 ± 5.4 g g−1 Cr 7.41 ± 1.47 g g−1 Cr 3.81 ± 1.93 g g−1 Cr 6.76 ± 4.71 g g−1 Cr
8 9 12 21 30 30 44 23 27 10 22 7 6 60
Off column CE-AD End-cloumn CE-AD HPLC–ECD
[38]
– NS S NS S NS S NS – – – S NS –
a
S: smoker; NS: non-smoker; Cr: creatinine; N: subject number.
HPLC–ECD HPLC–ECD HPLC–ECD ELISA – HPLC–ECD HPLC–ECD
622
Q.-H. Yao et al. / Talanta 63 (2004) 617–623
Acknowledgements This work has been supported by the Knowledge Innovation Program of the Chinese Academy of Sciences and the 863 project from State Ministry of Science and Technology of China (2003AA 223061).
References Fig. 4. Correlation between age and 8OHdG levels in urine.
in the ratio of urinary 8OHdG-to-creatinine of the two groups. There was a significant difference both in the urinary 8OHdG excretion (P = 0.0007) and in the ratio of urinary 8OHdG-to-creatinine (P = 0.002) between subjects smoking more than 10 cigarettes per day and non-smokers. From above results, we observed that cigarette smoking has a strong effect on 8OHdG content in urine, especially for those heavy smokers. There exists a significant correlation between age and 8OHdG levels in urine from totally studied crowd including smokers and non-smokers (r = 0.330, P = 0.032) (Fig. 4). This observation confirms the fact that oxidative DNA damage increases with age. No significant correlation was found between body mass and urinary 8OHdG levels from the crowd including smokers and non-smokers (r = 0.249, P = 0.11). 8OHdG is known as a sensitive biomarker of oxidative DNA damage also of oxidative stress. The high urinary 8OHdG levels in the smokers with heavy smoking habits investigated in this study suggested that smoking has an important role in increasing oxidative stress in human body. Although the biochemical–physiological basis is unknown, it may be related to smoke constituents, which include or generate reactive oxygen species or consume antioxidants or enhance the effect of smoking on the metabolic rate.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]
4. Conclusions In summary, the present study developed a CE-AD method to determine 8OHdG in urine. The electrochemical behaviors of 8OHdG on carbon fiber microdisk electrode were investigated in detail. And the effects of CE conditions, such as pH value of BGE solution, kinds of BGE and sample matrix effect, on the separation were discussed in the experiment. The optimized method showed a high sensitivity, a good selectivity, and a large linear range which made it suitable to analyze 8OHdG in urine. With the optimized method the effect of cigarette smoking on the extent of 8OHdG in urine was assessed, significantly positive relation was found. Further investigations are needed to understand the exact biologic and pathologic significance of cigarette smoking on oxidative DNA damage.
[22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37]
B.N. Ames, Free Rad. Res. Commun. 7 (3–6) (1989) 121. H. Kasai, H. Tanooka, S. Nishimura, Gann 75 (1984) 565. P. Leanderson, C. Tagesson, Chem. Biol. Interact. 75 (1990) 71. U.J. Nair, R.A. Floyd, V. Bussachini, M. Friesen, H. Bartsch, Chem. Biol. Interact. 63 (1987) 157. M.K. Shigenaga, J.W. Park, K.C. Cundy, C.J. Gimeno, B.N. Ames, Methods Enzymol. 186 (1990) 17. M.K. Shigenaga, E.N. Aboujaoude, Q. Chen, B.N. Ames, Methods Enzymol. 234 (1994) 521. M.K. Shigenaga, B.N. Ames, Free Rad. Bio. Med. 10 (1991) 211. R.A. Floyd, J.J. Watson, P.K. Wong, D.H. Altmiller, R.C. Richard, Free Rad. Res. Commun. 1 (1986) 163. S. Loft, K. Vistisen, M. Ewertz, A. Tjnne´øland, H.E. Poulsen, Carcinogenesis 13 (1992) 2241. S.R. Mei, G.W. Xu, J. Xing, C.Y. Wu, Anal. Sci. 17 (2001) 779. S.R. Mei, G.W. Xu, C.Y. Wu, Anal. Lett. 34 (2001) 2063. M. Erhola, S. Toyokuni, H. Ochi, K. Uchida, H. Alho, FEBS Lett. 409 (1997) 287. C. Tagesson, M. Källberg, C. Klinterberg, H. Starkhammar, Eur. J. Cancer 31A (1995) 934. Y. Takara, H. Ken-ichi, T. Takaya, K. Hiroshi, U. Mamoru, H.J. Hider, Obstet. Gynaecol. Res. 22 (4) (1996) 359. C.G. Fraga, P.A. Motchnik, A.J. Wyrobek, D.M. Rempel, B.N. Ames, Mutat. Res. 351 (2) (1996) 199. S. Asami, H. Manabe, J. Miyake, Y. Tsurudome, T. Hirano, R. Yamaguchi, H. Itoh, H. Kasai, Carcinogenesis 18 (9) (1997) 1763. M. Nakajima, T. Takeuchi, K. Morimoto, Carcinogenesis 17 (4) (1996) 787. D. Germadnik, A. Pilger, H.W. Rüdiger, J. Chromatogr. B 689 (1997) 399. A. Teixeira, H. Gommers-Ampt, G. Werken, J. Westra, J. Stavenuiter, Anal. Biochem. 214 (1993) 474. S.R. Mei, G.W. Xu, C.Y. Wu, Chin. J. Anal. Chem. 29 (2001) 1394. B. Yin, R. Whyatt, F. Perera, M. Randall, T. Cooper, R. Santella, Free Rad. Biol. Med. 18 (6) (1995) 1023. K. Vistisen, H.E. Poulsen, S. Loft, Carcinogenesis 13 (1992) 1561. Q. Weng, W. Jin, Anal. Chim. Acta 478 (2003) 199. A. Pluym, W. Ael, M. Smet, Trends Anal. Chem. 11 (1992) 27. K. Altria, M. Kelly, B. Clark, Trends Anal. Chem. 17 (1998) 204. K. Altria, M. Kelly, B. Clark, Trends Anal. Chem. 17 (1998) 214. Q. Dong, W. Jin, Electrophoresis 22 (2001) 2786. H. Lin, D. Xu, H. Chen, J. Chromatogr. A 760 (1997) 227. W. Jin, H. Wei, X. Zhao, Electroanalysis 9 (1997) 770. W. Jin, Q. Weng, J. Wu, Anal. Chim. Acta 342 (1997) 67. D. Weiss, C. Lunte, Electrophoresis 21 (2000) 2080. Y. Zheng, Y. Zhang, D. Liu, X. Guo, S. Mei, J. Xiong, H. Kong, C. Zhang, G. Xu, Chem. J. Chin. Univ. 22 (6) (2001) 912. T. Renner, T. Fechner, G. Scherer, J. Chromatogr. B. 738 (2000) 311. P. Britz-Mckibbin, D.D.Y. Chen, Anal. Chem. 72 (2000) 1242. P. Britz-Mckibbin, G. Bebault, D.D.Y. Chen, Anal. Chem. 72 (2000) 1729. S. Loft, A. Astrup, B. Buemann, H.E. Poulsen, FASEB J. 8 (1994) 534. J. Suzuki, Y. Inoue, S. Suzuki, Free Rad. Biol. Med. 18 (1995) 431.
Q.-H. Yao et al. / Talanta 63 (2004) 617–623 [38] S. Pourcelot, H. Faure, F. Firoozi, V. Ducros, M. Tripier, J. Hee, J. Cadet, A. Favier, Free Rad. Res. 30 (1999) 173. [39] M. Honda, Y. Yamada, M. Tomonaga, H. Ichinose, S. Kamihira, Leuk. Res. 24 (2000) 461. [40] H. Miyake, H. Eto, Y. Takechi, S. Kamidono, I. Hara, BJU Int. 91 (7) (2003) 657.
623
[41] A. Pilger, S. Ivancsits, D. Germadnik, H.W. Rudiger, J. Chromatogr. B. Analyt. Technol. Biomed. Life Sci. 778 (1–2) (2002) 393. [42] C. Tagesson, M. Kallberg, C. Klintenberg, H. Starkhammar, Eur. J. Cancer 31A (1995) 934. [43] S.R. Mei, Q.H. Yao, L.S. Cai, J. Xing, G.W. Xu, C.Y. Wu, Electrophoresis 24 (2003) 1411.