- Email: [email protected]
Simultaneous quantification of ethylpurine adducts in human urine by stable isotope dilution nanoflow liquid chromatography nanospray ionization tandem mass spectrometry
Journal of Chromatography A, 1322 (2013) 69–73
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Simultaneous quantification of ethylpurine adducts in human urine by stable isotope dilution nanoflow liquid chromatography nanospray ionization tandem mass spectrometry Hauh-Jyun Candy Chen ∗ , Chao-Ray Lin Department of Chemistry and Biochemistry, National Chung Cheng University, 168 University Road, Ming-Hsiung, Chia-Yi 62142, Taiwan
a r t i c l e
i n f o
Article history: Received 3 September 2013 Received in revised form 29 October 2013 Accepted 29 October 2013 Available online 6 November 2013 Keywords: DNA adduct 3-Ethyladenine 7-Ethylguanine Nanoflow LC Mass spectrometry Urine
a b s t r a c t Ethylating agents contained in cigarette smoke can damage DNA producing ethylated DNA adducts, including N3 -ethyladenine (3-EtAde) and N7 -ethylguanine (7-EtGua). In this study, a highly specific and sensitive assay based on stable isotope dilution nanoflow liquid chromatography nanospray ionization tandem mass spectrometry (nanoLC-NSI/MS/MS) was used to measure 3-EtAde and 7-EtGua in human urine. These urinary adducts were enriched by a polymeric reversed phase solid-phase extraction column before the nanoLC-NSI/MS/MS analysis. The on-column detection limits (S/N ≥ 3) of 3-EtAde and 7-EtGua were 15 fg (92 amol) and 10 fg (56 amol), respectively, while the lower quantification limits of 3-EtAde and 7-EtGua were 930 and 840 amol, respectively. Urinary concentrations of 3-EtAde and 7-EtGua in 21 smokers were 68.6 ± 29.4 and 18.7 ± 13.8 pg/mL, respectively. In 20 nonsmokers, concentrations of 3EtAde and 7-EtGua were 3.5 ± 3.8 and 2.4 ± 3.0 pg/mL, respectively. The urinary concentrations of 3-EtAde and 7-EtGua were statistically significantly higher in smokers than in nonsmokers (p < 0.0001). Moreover, 3-EtAde and 7-EtGua concentrations are significantly correlated with the number of cigarettes smoked per day and with the smoking index. This highly specific and sensitive assay based on stable isotope dilution nanoLC-NSI/MS/MS assay should be clinically valuable in assessing the possibility of measuring urinary ethylpurines as noninvasive biomarkers for smoking-related cancers in humans. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Cigarette smoke contains thousands of chemicals and more than seventy of them are carcinogens [1]. The fact that more than 80% of lung cancer patients are tobacco smokers or ex-smokers renders smoking as a primary cause of lung cancer [2]. Because tobacco intake is a major risk factor in lung and oral cancers, DNA ethylation might be associated with carcinogenesis in these cancers. Furthermore, the finding of cigarette smoking increased levels of ethylated DNA adducts in tissues and urine strongly suggests that ethylating agents are contained in the smoke [3–9]. The recent finding of 7-EtGua in calf thymus DNA treated with extracts of areca nut suggests that ethylating agents are also present in areca nut extracts [10]. Treatment of rats with the mutagenic N-ethyl-N-nitrosourea or N-nitrosodiethylamine causes tumor formation in brain and liver, along with formation of ethylated DNA adducts [11,12]. Ethylated DNA bases identified in humans include N3 -ethyladenine (3-EtAde), O6 -ethylguanine, N7 -ethylguanine (7-EtGua) and
∗ Corresponding author. Tel.: +886 5 242 8176; fax: +886 5 272 1040. E-mail address: [email protected] (H.-J.C. Chen). 0021-9673/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.chroma.2013.10.092
O2 -, O4 -, and N3 -ethylthymine [3–5,8,9,13–16]. In human tissues, 7-EtGua was quantified in liver and leukocyte DNA [14,17,18], while 3-EtAde was analyzed in leukocyte DNA [17]. Although results from animal studies indicated association between DNA ethylation and cancer [19,20], studies in humans are limited. The only connection was the reports in lung cancer patients that higher levels of O4 -ethylthymidine in DNA from normal lung tissue adjacent to tumors in smokers than in nonsmokers [7,13]. Urinary DNA adducts have been used as biomarkers for carcinogen exposure [21]. The glycosidic linkages of 3-EtAde and 7-EtGua are labile, with the respective half-lives of ca. 20–30 and 100–150 h at 37 ◦ C under neutral pH [22,23]. Thus, 3-EtAde and 7-EtGua can undergo spontaneous depurination. At the same time, they can be removed and repaired by DNA repair enzymes. As a result, these ethylpurine adducts have been detected in smokers’ urine [3–5,8,16]. Furthermore, higher levels of urinary 3-EtAde and 7EtGua in smokers than in nonsmokers have been reported [3–5,8]. Therefore, 3-EtAde and 7-EtGua should be potential biomarkers for smoking-induced DNA damage. There is a practical need to establish analytical methods for DNA adducts as biomarkers in surrogate tissues or biological fluids to assess DNA damage in the body because DNA adducts are often present in trace amounts in human tissues which are inaccessible.
70
H.-J.C. Chen, C.-R. Lin / J. Chromatogr. A 1322 (2013) 69–73
The development of analytical methods allowed characterization and quantification of trace levles of DNA adducts in biological tissues and fluids [24]. Among these methods, liquid chromatography tandem mass spectrometry is the state of the art choice [25,26]. In this study, a highly accurate, specific and sensitive method based on stable isotope dilution nanoflow liquid chromatography nanospray ionization tandem mass spectrometry (nanoLC-NSI/MS/MS) is used for simultaneous quantification of 3-EtAde and 7-EtGua in human urine. 2. Materials and methods 2.1. Chemicals and reagents [13 C1 ,15 N2 ]Adenine and [15 N5 ]2 -deoxyguanosine were purchased from Cambridge Isotope Laboratories (Andover, MA). Standard 3-EtAde, 7-EtGua, [13 C1 ,15 N2 ]3-EtAde, and [15 N5 ]7-EtGua were synthesized and characterized as previously reported [17]. All reagents are of reagent grade or above. 2.2. Adduct enrichment by SPE −80 ◦ C
An aliquot of 24-h urine stored at was defrosted and centrifuged at 23,000 × g for 20 min at 4 ◦ C. To the supernatant (100 L) was added 100 pg each of [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7EtGua and the solution (140 L) was loaded on a polymeric reversed phase SPE column [Strata-X, 100 mg, 1 mL, Phenomenex (Torrance, CA)] after conditioning with 5 mL of methanol, followed by 5 mL of water. After sample loading, the SPE column was washed with 1 mL of 15% aqueous MeOH, and the adducts were eluted with 1 mL of 40% aqueous MeOH. The fraction containing these adducts was evaporated, redissolved in 10 L of 0.01% formic acid (pH 3.7), and passed through a 0.22 m Nylon syringe filter. A volume of 2 L of the processed sample was subjected to the nanoLC-NSI/MS/MS analysis described below. 2.3. nanoLC-NSI/MS/MS analysis A 2 L injection loop was connected to a six-port switching valve injector into an LC system consisting of an UltiMate 3000 Nano LC system (Dionex, Amsterdam, The Netherlands) and a reversed phase column packed in-house (180 m × 11 cm, Magic C18AQ, ˚ Michrom BioResource, Album, CA). The pump out5 m, 200 A, put (30 L/min) was split before the injection port to a flow rate of 700 nL/min. Mobile phase A was 0.01% formic acid (pH 3.7), and mobile phase B was 0.1% formic acid in acetonitrile. The elution started with a linear gradient of 10% mobile phase B to 100% mobile phase B from 0 to 17 min, and maintained at 100% B in the next 13 min before conditioning with 100% mobile phase A. The effluent was subjected to analysis by a triple quadrupole mass spectrometer, TSQ Quantum Ultra EMR mass spectrometer (Thermo Electron Corp., San Jose, CA), equipped with a nanospray ionization (NSI) interface under the positive-ion mode for the NSIMS/MS. The column effluent enters the spray chamber through a tapered emitter constructed from a 180-m i.d. fused-silica capillary and is directly electrosprayed into the mass spectrometer in the positive ion mode. The spray was monitored by a built-in CCD camera. The spray voltage was 1.5 kV, and the source temperature was at 270 ◦ C. In MS/MS experiments, argon was used as the collision gas and the pressure of the collision cell was 1.5 mTorr. The adductenriched sample was analyzed by nanoLC-NSI/MS/MS using the transition from the precursor ion [M+H]+ focused in quadrupole 1 (Q1) and dissociated in a collision cell (Q2), yielding the product ion, which was analyzed in quadrupole 3 (Q3). Under the highly selected reaction monitoring (H-SRM) mode, the mass width of Q1 and Q3 was 0.2 and 0.7 m/z, respectively, and the dwell time
was 0.1 s. Two H-SRM transitions were simultaneously employed: method 1 monitored 3-EtAde in Q1 for the precursor ion [M+H]+ at m/z 164.2 and in Q3 for the product ion [M+H−C2 H4 ]+ at m/z 136.0. For [13 C1 ,15 N2 ]3-EtAde, Q1 and Q3 were at m/z 167.2 and m/z 139.0, respectively. For 7-EtGua, Q1 and Q3 were for the precursor ion [M+H]+ at m/z 180.1 and for the product ion [M+H−C2 H4 ]+ at m/z 152.0, respectively, while those for [15 N5 ]7-EtGua were at m/z 185.1 (Q1) and m/z 157.0 (Q3). The collision energy was 20 eV for these four H-SRM transitions. The H-SRM method 2 monitored Q1 for the precursor ion [M+H]+ at m/z 164.2 and Q3 for the product ion [M+H−C2 H4 −NH3 ]+ at m/z 119.0 for 3-EtAde and from m/z 167.2 to m/z 122.0 for [13 C1 ,15 N2 ]3-EtAde, with a collision energy of 30 eV for both transitions. For 7-EtGua, Q1 and Q3 was for the precursor ion [M+H]+ at m/z 180.1 and for the product ion [M+H−NH3 ]+ at m/z 163.0, respectively, while that for [15 N5 ]7-EtGua was at m/z 185.1 (Q1) and m/z 167.0 (Q3), with a collision energy of 20 eV for both transitions.
2.4. Method validation The detection limits was defined by injecting the smallest amount of standard 3-EtAde and 7-EtGua on-column with S/N ≥ 3. The calibration curves were obtained by adding various amount of the standards (0, 0.01, 0.15, 2, 6, 15, and 20 pg) in the presence of 100 pg each of [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtGua to a urine sample with non-detectable amounts of 3-EtAde and 7-EtGua. Each solution went through the SPE column described in Section 2.2. The fraction containing these adducts was evaporated and reconstituted in 10 L of 0.01% formic acid (pH 3.7), and 2 L of the aliquot was subjected to the nanoLC-NSI/MS/MS analysis described above. The lowest amount of analyte showing linearity with S/N ≥ 10 was defined as the lower limit of quantification (LLOQ). The precision of the assay was assessed by the relative standard deviation (RSD) in analyzing 3-EtAde and 7-EtGua in a urine sample in triplicates, and the experiments were performed on three different days using the optimized assay conditions. The accuracy of this assay was evaluated by adding known amount of 3-EtAde and 7-EtGua (6, 12, 18 pg each) to 0.1 mL of a urine sample and measured the total amount. The y-intercept obtained by linear regression was compared with the adduct level in this sample without addition of the standards.
2.5. Study-subjects This study is approved by the Institutional Review Board of the National Chung Cheng University (IRB No. 100112902). The studysubjects were healthy individuals recruited from employees and students of the National Chung Cheng University, including 21 male smokers and 20 nonsmokers (10 male and 10 female). The mean (±standard deviation (SD)) age was 37 ± 16 for smokers and 36 ± 11 for nonsmokers. The mean (±SD) smoking index (number of cigarette per day × years smoked) of the study-subjects was 203 ± 201 in a range of 4–690. The subjects received a written warranty stating that the information was for research purposes only and that their personal information would be kept confidential.
2.6. Statistical analysis GraphPad InStat version 3.00 for Windows 95, GraphPad Software (San Diego, CA, www.graphpad. com) was used for statistical analysis. The nonparametric Mann–Whitney test was used to analyze levels of 3-EtAde and 7-EtGua in urine between the smokers and nonsmokers. The correlation between each adduct and the number of cigarettes smoked per day or the smoking index as well
H.-J.C. Chen, C.-R. Lin / J. Chromatogr. A 1322 (2013) 69–73
as between two adduct levels were performed by the nonparametric Spearman correlation. 3. Results and discussion 3.1. Method development As shown in Fig. 1, the assay procedures are straight forward, including (1) addition of [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtGua to urine as internal standards, (2) adduct enrichment by a disposable solid-phase extraction (SPE) column, and (3) nanoLC-NSI/MS/MS analysis under the H-SRM mode. Urine is a complex mixture composed of high contents of salts and metabolites. If the sample pretreatment is not appropriate, the interfering compounds result in suppression of ionization for the analytes of interest (i.e. matrix effect) and give complicated chromatograms, which make characterization and quantification of analytes difficult or impossible in some cases. Incorporation of isotope analog of the analyte as the internal standard adjusts for the loss of analytes during the assay procedures. It is critical that adducts are purified prior to analysis by LC–MS/MS to remove interfering compounds as much as possible and overcome problems encountered with suppression of ionization due to matrix effects. One approach to achieve this goal is to use antibody against the analyte to specifically capture it from the complex mixture. The drawback is that the antibody is expensive and the results could be irreproducible due to batch-to-batch preparation and the life-time of the antibody [27]. Another method is to collect the analyte from HPLC column, which is a labor-intensive process and there is a risk of cross-contamination between samples [28,29]. In this study, several solid-phase extraction (SPE) columns and elution conditions were tried for purification of 3-EtAde and 7EtGua in urine, including silica gel (hydrophilic interaction liquid chromatography, HILIC), mix-mode cation exchange (MCX), nonendcapped reversed phase (C18-OH), and a polymeric reversed phase (PRP) SPE column. Poor chromatograms were obtained for both 3-EtAde and 7-EtGua analysis after silica SPE cleanup. Using MCX or C18-OH alone resulted in severe ion suppression and the added [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtGua were barely detectable. Employing MCX followed by C18-OH SPE, the resulting fraction gave satisfactory chromatograms for analysis of 3-EtAde, but not for 7-EtGua (data not shown). Polymeric reversed phase SPE provides the most satisfactory results in terms of removing most of the interference from the matrix. The urinary ethylpurine adducts were eventually enriched by the PRP SPE column which leads to clean nanoLC-NSI/MS/MS chromatograms for monitoring of 3-EtAde, 7-EtGua, [13 C1 ,15 N2 ]3-EtAde, and [15 N5 ]7EtAGua.
71
After urine was enriched by the SPE column, there are still many compounds in the fraction collected together with the adducts and the internal standards. High sensitivity can be achieved when a column with a small inner diameter (i.d.) eluting at low flow rate is employed. We and others have used a 75 m-i.d. column for analysis of DNA adducts in tissue and leukocyte DNA [9,18,30,31]. The matrix effect in DNA hydrolysate is smaller than that in the complex mixture of urine partly because DNA is purified to a certain extent before enzyme hydrolysis. Although the small i.d. column can provide high sensitivity in the MS, a 180 m-i.d. column was used in this study for analysis of urinary 3-EtAde and 7-EtGua to avoid residues in the urine sample affecting the durability of the analytical column. Compared to electrospray ionization (ESI), nanospray or nanoelectrospray ionization (NSI) method operates at sufficiently low flow rates and analyte concentrations, which results in small droplets and high ionization efficiency in the mass spectrometry [32]. The coupling of nanoflow LC to NSI-MS leads to further gains in sensitivity. Among mass spectrometers, the use of static modes on both quadrupole 1 (Q1) and quadrupole 3 (Q3) of a triple quadrupole mass spectrometer offers the highest sensitivity possible. To increase the specificity of the measurement, the highly selective reaction monitoring (H-SRM) mode was employed in this study with the mass width at Q1 and Q3 being at 0.2 and 0.7 Da, respectively. 3.2. Method performance and validation The detection limits (S/N ≥ 3) of 3-EtAde and 7-EtGua injected on-column were 15 fg (92 amol) and 10 fg (56 amol), respectively. The calibration curves were obtained by adding various amount of the standards (0–20 pg) to a urine sample (No. 37) with non-detectable amounts of 3-EtAde and 7-EtGua (Fig. 1S, Supplementary material). The lowest amount of analyte showing linearity, i.e. the lower limit of quantification (LLOQ) with S/N ≥ 10, was 150 fg for both 3-EtAde and 7-EtGua. This gives the concentration quantification limit of 1.5 pg/mL, starting with 0.1 mL of urine. The precision of the assay was assessed by analyzing 3-EtAde and 7-EtGua in a urine sample (No. 21) in triplicates, and the experiments were performed on three different days using the optimized assay conditions. The results revealed that the relative standard deviations for both intraday and interday analyses were within 10% (Table 1S, Supplementary material), showing the high reproducibility of the assay. The accuracy of this assay was evaluated by adding known amount of 3-EtAde and 7-EtGua (6, 12, 18 pg each) to 0.1 mL of a urine sample (No. 6) and measured the total amount. The yintercept obtained by linear regression was 6.20 and 1.77 pg for 3-EtAde and 7-EtGua, respectively. The levels are very close to the concentration of 3-EtAde and 7-EtGua in this sample without addition of standards, i.e. 6.13 and 1.67 pg in 0.1 mL (or 61.3 and 16.7 pg/mL), respectively (Fig. 1S, Supplementary material). The matrix effect for 3-EtAde and 7-EtGua in urine was 56% and 34%, respectively, determined by the decrease in the peak area of the internal standards [13 C1 ,15 N2 ]3-EtAde or [15 N5 ]7-EtGua after addition of a SPE-enriched urine sample (No. 6). 3.3. nanoLC-NSI/MS/MS analysis of 3-EtAde and 7-EtGua in human urine
Fig. 1. Assay procedure for urinary 3-EtAde and 7-EtGua by stable isotope dilution nanoLC-NSI/MS/MS.
Fig. 2 shows representative nanoLC-NSI/MS/MS chromatograms of 3-EtAde and 7-EtGua in a smoker’s urine (No. 2) under the HSRM mode. Using H-SRM method 1, 3-EtAde and 7-EtGua eluted at 10.7 and 13.4 min, co-eluting with their isotope-labeled standards [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtAGua, respectively (Fig. 2A).
72
H.-J.C. Chen, C.-R. Lin / J. Chromatogr. A 1322 (2013) 69–73 Table 1 Spearman correlation coefficient (␥) between urinary 3-EtAde and 7-EtGua levels and numbers of cigarettes smoked per day and smoking index.a , b 3-EtAde
7-EtGua
Number of cigarettes smoked per day 0.6600 (<0.0001) 0.9011(< 0.0001) Smoking index 0.8670(< 0.0001) a b
0.6824 (<0.0001)
The two-tailed p value is shown in the parenthesis. n = 41.
coefficient of 0.8670 (p < 0.0001) and 0.6824 (p < 0.0001), respectively (Table 1). 3.4. Comparison of 3-EtAde and 7-EtGua levels
Fig. 2. nanoLC-NSI/MS/MS chromatograms of 3-EtAand 7-EtG in a smoker’s urine (No. 2) under H-SRM (A) method 1 and (B) method 2.
Co-elution of 3-EtAde and 7-EtGua with their isotope-labeled standards was also observed using H-SRM method 2. However, H-SRM method 2 was less sensitive than method 1, based on the smaller peak intensity for both [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtAGua. In addition, the interference in the channels of 3-EtAde and 7EtGua was more severe using H-SRM method 2 (Fig. 2B). Therefore, H-SRM method 1 was used for quantification of adducts, while HSRM method 2 was used for qualitative assurance of the presence of these adducts. This assay requires only 0.1 mL of urine and an equivalent of 20 L of processed urine was injected on-column. Concentrations of urinary 3-EtAde and 7-EtGua in smokers and nonsmokers were determined and listed in Table 2S, Supplementary material. Adduct concentrations of 3-EtAde and 7-EtGua in 21 smokers were 68.6 ± 29.4 and 18.7 ± 13.8 pg/mL, respectively. In 20 nonsmokers, adduct concentrations of 3-EtAde and 7-EtGua were 3.5 ± 3.8 and 2.4 ± 3.0 pg/mL, respectively. Levels in 4 out of 20 nonsmokers were not detectable. As shown in Fig. 3, the concentrations of 3-EtAde and 7-EtGua are statistically significantly higher in smokers than in nonsmokers (p < 0.0001) performed by the nonparametric Mann–Whitney test. Moreover, there are positive associations between the concentrations of 3-EtAde and 7-EtGua and the number of cigarettes smoked per day with a correlation coefficient of 0.9011 and 0.6600, respectively (p < 0.0001) using the nonparametric Spearman correlation. Additionally, there is a statistically significant association between the smoking index (number of cigarettes smoked per day × years smoked) and the urinary concentrations of 3-EtAde and 7-EtGua with a correlation
The results indicated that the average urinary concentration of 3-EtAde is ca. 3.6 times higher than that of 7-EtGua, while our previous study reported a 1.7 times higher mean level of 3-EtAde compared to 7-EtGua in smokers’ leukocytes DNA [17]. Urinary excretion of 3-EtAde and 7-EtGua attributed from individual’s removal and repair activity in addition to nonenzymatic spontaneous depurination. These adducts are recognized by different enzymatic activities which remove and repair at various rates with efficiencies depending on the adduct levels and enzyme contents in the cell [33]. The half-lives of 3-EtAde and 7-EtGua induced by Nethyl-N-nitrosourea were reported to be ca. 12 and 75 h in rat liver, respectively, and 16 and 90 h in rat brain, respectively [34]. However, there is no report on the half-lives of 3-EtAde and 7-EtGua in human tissues so far. While the removal and repair efficiency of 3-EtAde and 7-EtGua in human body is not yet understood, this assay provides a valuable tool in measuring urinary 3-EtAde and 7-EtGu. Combination of data from urinary concentration and DNA adduct level in tissues from an individual should shed light in our understanding of formation and repair of 3-EtAde and 7-EtGua in the whole body. 3.5. Potential dietary effect on urinary 3-EtAde and 7-EtGua excretion Factors other than smoking, such as diet, in contribution to ethylpurine adducts in tissues and in urine have been investigated. In contrary to 3-methyladenine, change of diet only affects urinary excretion of 3-EtAde very slightly in smoking subjects and in subjects ingesting 3-EtAde [3,5]. N-nitrosodiethylamine and Nethyl-N-nitrosourea are the possible dietary sources of ethylating agent humans expose to. N-Nitrosodiethylamine occurs in human diet, such as in cured meat [35]. Hepatocarcinogenesis in rainbow trout treated with N-nitrosodiethylamine is modulated by indole-3-carbinol and beta-naphthoflavone. Liver 7-EtGua levels are reduced in indole-3-carbinol-pretreated fish and increase in -naphthoflavone-pretreated fish [36]. Vitamin supplementation can modulate the tumorigenicity of N-ethyl-N-nitrosourea. For instance, the N-ethyl-N-nitrosourea-induced tumorigenicity in rat brain is affected by dietary vitamin A or N-acetylcystein [37]. Niacin decreases the incidence of nonlymphocytic leukemia in rats induced by N-ethyl-N-nitrosourea [38]. Furthermore, fatty acid composition in the diet influences on N-ethyl-N-nitrosoureainduced mammary tumor incidence in the rat [39]. Recently, ketogenic diet is shown to significantly extend the lifespan of Nethyl-N-nitrosourea-treated mice which could otherwise die from lethal cardiomyopathy due to mutation in a critical gene [40]. The assay developed in this study should provide a valuable tool in evaluation of specific dietary ingredients in urinary excretion of these ethylpurine adducts.
H.-J.C. Chen, C.-R. Lin / J. Chromatogr. A 1322 (2013) 69–73
73
Fig. 3. Comparison of urinary (A) 3-EtAde and (B) 7-EtGua concentrations in smokers and in nonsmokers.
4. Conclusion To our knowledge, this the first study for simultaneous quantification of 3-EtAde and 7-EtGua in human urine, which allows comparison of levels of these two ethylpurine adducts. This nanoLC-NSI/MS/MS assay is highly accurate, sensitive and specific, which requires as little as 0.1 mL of urine to provide two ethylpurine adducts as noninvasive biomarkers of smoking-associated DNA damage. Conflict of interest statement The authors declare that there are no conflicts of interest. Funding sources This work was supported by National Science Council of Taiwan (Grant NSC 100-2113-M-194-002-MY3) and National Chung Cheng University (to H.-J.C.C.). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma. 2013.10.092. References [1] IARC, IARC Monogr. Eval. Carcinog. Risks Hum., International Agency for Research on Cancer, Lyon, France, 2004, pp. 53. [2] S.S. Hecht, Int. J. Cancer 131 (2012) 2724. [3] V. Prevost, D.E. Shuker, M.D. Friesen, G. Eberle, M.F. Rajewsky, H. Bartsch, Carcinogenesis 14 (1993) 199. [4] A. Kopplin, G. Eberle-Adamkiewicz, K.H. Glusenkamp, P. Nehls, U. Kirstein, Carcinogenesis 16 (1995) 2637. [5] V. Prevost, D.E. Shuker, Chem. Res. Toxicol. 9 (1996) 439. [6] R. Godschalk, J. Nair, H.C. Kliem, M. Wiessler, G. Bouvier, H. Bartsch, Chem. Res. Toxicol. 15 (2002) 433.
[7] R. Godschalk, J. Nair, F.J. van Schooten, A. Risch, P. Drings, K. Kayser, H. Dienemann, H. Bartsch, Carcinogenesis 23 (2002) 2081. [8] M.R. Chao, C.J. Wang, L.W. Chang, C.W. Hu, Carcinogenesis 27 (2006) 146. [9] H.J. Chen, Y.C. Wang, W.P. Lin, Anal. Chem. 84 (2012) 2521. [10] C.W. Hu, M.R. Chao, Chem. Res. Toxicol. 25 (2012) 2386. [11] L. Verna, J. Whysner, G.M. Williams, Pharmacol. Ther. 71 (1996) 57. [12] L. Den Engelse, G.J. Menkveld, R.J. De Brij, A.D. Tates, Carcinogenesis 7 (1986) 393. [13] L. Anna, K. Kovacs, E. Gyorffy, B. Schoket, J. Nair, Mutagenesis 26 (2011) 523. [14] L. Chen, M. Wang, P.W. Villalta, S.S. Hecht, Chem. Res. Toxicol. 20 (2007) 1498. [15] S.M. Bronstein, T.R. Skopek, J.A. Swenberg, Cancer Res. 52 (1992) 2008. [16] S. Feng, H.J. Roethig, Q. Liang, R. Kinser, Y. Jin, G. Scherer, M. Urban, J. Engl, K. Riedel, Biomarkers 11 (2006) 28. [17] H.J. Chen, Y.F. Liu, J. Chromatogr. A 1271 (2013) 86. [18] S. Balbo, P.W. Villalta, S.S. Hecht, Chem. Res. Toxicol. 24 (2011) 1729. [19] J.A. Swenberg, M.C. Dyroff, M.A. Bedell, J.A. Popp, N. Huh, U. Kirstein, M.F. Rajewsky, Proc. Natl. Acad. Sci. U. S. A. 81 (1984) 1692. [20] J. Thomale, N.H. Huh, P. Nehls, G. Eberle, M.F. Rajewsky, Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 9883. [21] D.E. Shuker, P.B. Farmer, Chem. Res. Toxicol. 5 (1992) 450. [22] W.J. Bodell, B. Singer, G.H. Thomas, J.E. Cleaver, Nucleic Acids Res. 6 (1979) 2819. [23] B. Singer, T.P. Brent, Proc. Natl. Acad. Sci. U. S. A. 78 (1981) 856. [24] M.W. Himmelstein, P.J. Boogaard, J. Cadet, P.B. Farmer, J.H. Kim, E.A. Martin, R. Persaud, D.E. Shuker, Crit. Rev. Toxicol. 39 (2009) 679. [25] H.J. Chen, Mol. Nutr. Food Res. 55 (2011) 1391. [26] R. Singh, P.B. Farmer, Carcinogenesis 27 (2006) 178. [27] T.Y. Yen, S. Holt, R. Sangaiah, A. Gold, J.A. Swenberg, Chem. Res. Toxicol. 11 (1998) 810. [28] F. Bianchini, F. Donato, H. Faure, J.L. Ravanat, J. Hall, J. Cadet, Int. J. Cancer 77 (1998) 40. [29] J.L. Ravanat, P. Guicherd, Z. Tuce, J. Cadet, Chem. Res. Toxicol. 12 (1999) 802. [30] H.J. Chen, W.P. Lin, Anal. Chem. 83 (2011) 8543. [31] H.J. Chen, G.J. Lin, W.P. Lin, Anal. Chem. 82 (2010) 4486. [32] R.D. Smith, Y. Shen, K. Tang, Acc. Chem. Res. 37 (2004) 269. [33] B. Singer, Environ. Health Perspect. 62 (1985) 41. [34] R. Goth, M.F. Rajewsky, Proc. Natl. Acad. Sci. U. S. A. 71 (1974) 639. [35] A.R. Tricker, R. Preussmann, Mutat. Res. 259 (1991) 277. [36] A.T. Fong, J.D. Hendricks, R.H. Dashwood, S. Van Winkle, B.C. Lee, G.S. Bailey, Toxicol. Appl. Pharmacol. 96 (1988) 93. [37] D.A. Ross, P. Kish, K.M. Muraszko, M. Blaivas, M. Strawderman, J. Neurooncol. 40 (1998) 29. [38] A.P. Bartleman, R. Jacobs, J.B. Kirkland, Nutr. Cancer 60 (2008) 251. [39] C.M. Williams, K. Maunder, Br. J. Nutr. 71 (1994) 543. [40] P. Krebs, W. Fan, Y.H. Chen, K. Tobita, M.R. Downes, M.R. Wood, L. Sun, X. Li, Y. Xia, N. Ding, J.M. Spaeth, E.M. Moresco, T.G. Boyer, C.W. Lo, J. Yen, R.M. Evans, B. Beutler, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 19678.
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Simultaneous quantification of ethylpurine adducts in human urine by stable isotope dilution nanoflow liquid chromatography nanospray ionization tandem mass spectrometry Hauh-Jyun Candy Chen ∗ , Chao-Ray Lin Department of Chemistry and Biochemistry, National Chung Cheng University, 168 University Road, Ming-Hsiung, Chia-Yi 62142, Taiwan
a r t i c l e
i n f o
Article history: Received 3 September 2013 Received in revised form 29 October 2013 Accepted 29 October 2013 Available online 6 November 2013 Keywords: DNA adduct 3-Ethyladenine 7-Ethylguanine Nanoflow LC Mass spectrometry Urine
a b s t r a c t Ethylating agents contained in cigarette smoke can damage DNA producing ethylated DNA adducts, including N3 -ethyladenine (3-EtAde) and N7 -ethylguanine (7-EtGua). In this study, a highly specific and sensitive assay based on stable isotope dilution nanoflow liquid chromatography nanospray ionization tandem mass spectrometry (nanoLC-NSI/MS/MS) was used to measure 3-EtAde and 7-EtGua in human urine. These urinary adducts were enriched by a polymeric reversed phase solid-phase extraction column before the nanoLC-NSI/MS/MS analysis. The on-column detection limits (S/N ≥ 3) of 3-EtAde and 7-EtGua were 15 fg (92 amol) and 10 fg (56 amol), respectively, while the lower quantification limits of 3-EtAde and 7-EtGua were 930 and 840 amol, respectively. Urinary concentrations of 3-EtAde and 7-EtGua in 21 smokers were 68.6 ± 29.4 and 18.7 ± 13.8 pg/mL, respectively. In 20 nonsmokers, concentrations of 3EtAde and 7-EtGua were 3.5 ± 3.8 and 2.4 ± 3.0 pg/mL, respectively. The urinary concentrations of 3-EtAde and 7-EtGua were statistically significantly higher in smokers than in nonsmokers (p < 0.0001). Moreover, 3-EtAde and 7-EtGua concentrations are significantly correlated with the number of cigarettes smoked per day and with the smoking index. This highly specific and sensitive assay based on stable isotope dilution nanoLC-NSI/MS/MS assay should be clinically valuable in assessing the possibility of measuring urinary ethylpurines as noninvasive biomarkers for smoking-related cancers in humans. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Cigarette smoke contains thousands of chemicals and more than seventy of them are carcinogens [1]. The fact that more than 80% of lung cancer patients are tobacco smokers or ex-smokers renders smoking as a primary cause of lung cancer [2]. Because tobacco intake is a major risk factor in lung and oral cancers, DNA ethylation might be associated with carcinogenesis in these cancers. Furthermore, the finding of cigarette smoking increased levels of ethylated DNA adducts in tissues and urine strongly suggests that ethylating agents are contained in the smoke [3–9]. The recent finding of 7-EtGua in calf thymus DNA treated with extracts of areca nut suggests that ethylating agents are also present in areca nut extracts [10]. Treatment of rats with the mutagenic N-ethyl-N-nitrosourea or N-nitrosodiethylamine causes tumor formation in brain and liver, along with formation of ethylated DNA adducts [11,12]. Ethylated DNA bases identified in humans include N3 -ethyladenine (3-EtAde), O6 -ethylguanine, N7 -ethylguanine (7-EtGua) and
∗ Corresponding author. Tel.: +886 5 242 8176; fax: +886 5 272 1040. E-mail address: [email protected] (H.-J.C. Chen). 0021-9673/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.chroma.2013.10.092
O2 -, O4 -, and N3 -ethylthymine [3–5,8,9,13–16]. In human tissues, 7-EtGua was quantified in liver and leukocyte DNA [14,17,18], while 3-EtAde was analyzed in leukocyte DNA [17]. Although results from animal studies indicated association between DNA ethylation and cancer [19,20], studies in humans are limited. The only connection was the reports in lung cancer patients that higher levels of O4 -ethylthymidine in DNA from normal lung tissue adjacent to tumors in smokers than in nonsmokers [7,13]. Urinary DNA adducts have been used as biomarkers for carcinogen exposure [21]. The glycosidic linkages of 3-EtAde and 7-EtGua are labile, with the respective half-lives of ca. 20–30 and 100–150 h at 37 ◦ C under neutral pH [22,23]. Thus, 3-EtAde and 7-EtGua can undergo spontaneous depurination. At the same time, they can be removed and repaired by DNA repair enzymes. As a result, these ethylpurine adducts have been detected in smokers’ urine [3–5,8,16]. Furthermore, higher levels of urinary 3-EtAde and 7EtGua in smokers than in nonsmokers have been reported [3–5,8]. Therefore, 3-EtAde and 7-EtGua should be potential biomarkers for smoking-induced DNA damage. There is a practical need to establish analytical methods for DNA adducts as biomarkers in surrogate tissues or biological fluids to assess DNA damage in the body because DNA adducts are often present in trace amounts in human tissues which are inaccessible.
70
H.-J.C. Chen, C.-R. Lin / J. Chromatogr. A 1322 (2013) 69–73
The development of analytical methods allowed characterization and quantification of trace levles of DNA adducts in biological tissues and fluids [24]. Among these methods, liquid chromatography tandem mass spectrometry is the state of the art choice [25,26]. In this study, a highly accurate, specific and sensitive method based on stable isotope dilution nanoflow liquid chromatography nanospray ionization tandem mass spectrometry (nanoLC-NSI/MS/MS) is used for simultaneous quantification of 3-EtAde and 7-EtGua in human urine. 2. Materials and methods 2.1. Chemicals and reagents [13 C1 ,15 N2 ]Adenine and [15 N5 ]2 -deoxyguanosine were purchased from Cambridge Isotope Laboratories (Andover, MA). Standard 3-EtAde, 7-EtGua, [13 C1 ,15 N2 ]3-EtAde, and [15 N5 ]7-EtGua were synthesized and characterized as previously reported [17]. All reagents are of reagent grade or above. 2.2. Adduct enrichment by SPE −80 ◦ C
An aliquot of 24-h urine stored at was defrosted and centrifuged at 23,000 × g for 20 min at 4 ◦ C. To the supernatant (100 L) was added 100 pg each of [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7EtGua and the solution (140 L) was loaded on a polymeric reversed phase SPE column [Strata-X, 100 mg, 1 mL, Phenomenex (Torrance, CA)] after conditioning with 5 mL of methanol, followed by 5 mL of water. After sample loading, the SPE column was washed with 1 mL of 15% aqueous MeOH, and the adducts were eluted with 1 mL of 40% aqueous MeOH. The fraction containing these adducts was evaporated, redissolved in 10 L of 0.01% formic acid (pH 3.7), and passed through a 0.22 m Nylon syringe filter. A volume of 2 L of the processed sample was subjected to the nanoLC-NSI/MS/MS analysis described below. 2.3. nanoLC-NSI/MS/MS analysis A 2 L injection loop was connected to a six-port switching valve injector into an LC system consisting of an UltiMate 3000 Nano LC system (Dionex, Amsterdam, The Netherlands) and a reversed phase column packed in-house (180 m × 11 cm, Magic C18AQ, ˚ Michrom BioResource, Album, CA). The pump out5 m, 200 A, put (30 L/min) was split before the injection port to a flow rate of 700 nL/min. Mobile phase A was 0.01% formic acid (pH 3.7), and mobile phase B was 0.1% formic acid in acetonitrile. The elution started with a linear gradient of 10% mobile phase B to 100% mobile phase B from 0 to 17 min, and maintained at 100% B in the next 13 min before conditioning with 100% mobile phase A. The effluent was subjected to analysis by a triple quadrupole mass spectrometer, TSQ Quantum Ultra EMR mass spectrometer (Thermo Electron Corp., San Jose, CA), equipped with a nanospray ionization (NSI) interface under the positive-ion mode for the NSIMS/MS. The column effluent enters the spray chamber through a tapered emitter constructed from a 180-m i.d. fused-silica capillary and is directly electrosprayed into the mass spectrometer in the positive ion mode. The spray was monitored by a built-in CCD camera. The spray voltage was 1.5 kV, and the source temperature was at 270 ◦ C. In MS/MS experiments, argon was used as the collision gas and the pressure of the collision cell was 1.5 mTorr. The adductenriched sample was analyzed by nanoLC-NSI/MS/MS using the transition from the precursor ion [M+H]+ focused in quadrupole 1 (Q1) and dissociated in a collision cell (Q2), yielding the product ion, which was analyzed in quadrupole 3 (Q3). Under the highly selected reaction monitoring (H-SRM) mode, the mass width of Q1 and Q3 was 0.2 and 0.7 m/z, respectively, and the dwell time
was 0.1 s. Two H-SRM transitions were simultaneously employed: method 1 monitored 3-EtAde in Q1 for the precursor ion [M+H]+ at m/z 164.2 and in Q3 for the product ion [M+H−C2 H4 ]+ at m/z 136.0. For [13 C1 ,15 N2 ]3-EtAde, Q1 and Q3 were at m/z 167.2 and m/z 139.0, respectively. For 7-EtGua, Q1 and Q3 were for the precursor ion [M+H]+ at m/z 180.1 and for the product ion [M+H−C2 H4 ]+ at m/z 152.0, respectively, while those for [15 N5 ]7-EtGua were at m/z 185.1 (Q1) and m/z 157.0 (Q3). The collision energy was 20 eV for these four H-SRM transitions. The H-SRM method 2 monitored Q1 for the precursor ion [M+H]+ at m/z 164.2 and Q3 for the product ion [M+H−C2 H4 −NH3 ]+ at m/z 119.0 for 3-EtAde and from m/z 167.2 to m/z 122.0 for [13 C1 ,15 N2 ]3-EtAde, with a collision energy of 30 eV for both transitions. For 7-EtGua, Q1 and Q3 was for the precursor ion [M+H]+ at m/z 180.1 and for the product ion [M+H−NH3 ]+ at m/z 163.0, respectively, while that for [15 N5 ]7-EtGua was at m/z 185.1 (Q1) and m/z 167.0 (Q3), with a collision energy of 20 eV for both transitions.
2.4. Method validation The detection limits was defined by injecting the smallest amount of standard 3-EtAde and 7-EtGua on-column with S/N ≥ 3. The calibration curves were obtained by adding various amount of the standards (0, 0.01, 0.15, 2, 6, 15, and 20 pg) in the presence of 100 pg each of [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtGua to a urine sample with non-detectable amounts of 3-EtAde and 7-EtGua. Each solution went through the SPE column described in Section 2.2. The fraction containing these adducts was evaporated and reconstituted in 10 L of 0.01% formic acid (pH 3.7), and 2 L of the aliquot was subjected to the nanoLC-NSI/MS/MS analysis described above. The lowest amount of analyte showing linearity with S/N ≥ 10 was defined as the lower limit of quantification (LLOQ). The precision of the assay was assessed by the relative standard deviation (RSD) in analyzing 3-EtAde and 7-EtGua in a urine sample in triplicates, and the experiments were performed on three different days using the optimized assay conditions. The accuracy of this assay was evaluated by adding known amount of 3-EtAde and 7-EtGua (6, 12, 18 pg each) to 0.1 mL of a urine sample and measured the total amount. The y-intercept obtained by linear regression was compared with the adduct level in this sample without addition of the standards.
2.5. Study-subjects This study is approved by the Institutional Review Board of the National Chung Cheng University (IRB No. 100112902). The studysubjects were healthy individuals recruited from employees and students of the National Chung Cheng University, including 21 male smokers and 20 nonsmokers (10 male and 10 female). The mean (±standard deviation (SD)) age was 37 ± 16 for smokers and 36 ± 11 for nonsmokers. The mean (±SD) smoking index (number of cigarette per day × years smoked) of the study-subjects was 203 ± 201 in a range of 4–690. The subjects received a written warranty stating that the information was for research purposes only and that their personal information would be kept confidential.
2.6. Statistical analysis GraphPad InStat version 3.00 for Windows 95, GraphPad Software (San Diego, CA, www.graphpad. com) was used for statistical analysis. The nonparametric Mann–Whitney test was used to analyze levels of 3-EtAde and 7-EtGua in urine between the smokers and nonsmokers. The correlation between each adduct and the number of cigarettes smoked per day or the smoking index as well
H.-J.C. Chen, C.-R. Lin / J. Chromatogr. A 1322 (2013) 69–73
as between two adduct levels were performed by the nonparametric Spearman correlation. 3. Results and discussion 3.1. Method development As shown in Fig. 1, the assay procedures are straight forward, including (1) addition of [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtGua to urine as internal standards, (2) adduct enrichment by a disposable solid-phase extraction (SPE) column, and (3) nanoLC-NSI/MS/MS analysis under the H-SRM mode. Urine is a complex mixture composed of high contents of salts and metabolites. If the sample pretreatment is not appropriate, the interfering compounds result in suppression of ionization for the analytes of interest (i.e. matrix effect) and give complicated chromatograms, which make characterization and quantification of analytes difficult or impossible in some cases. Incorporation of isotope analog of the analyte as the internal standard adjusts for the loss of analytes during the assay procedures. It is critical that adducts are purified prior to analysis by LC–MS/MS to remove interfering compounds as much as possible and overcome problems encountered with suppression of ionization due to matrix effects. One approach to achieve this goal is to use antibody against the analyte to specifically capture it from the complex mixture. The drawback is that the antibody is expensive and the results could be irreproducible due to batch-to-batch preparation and the life-time of the antibody [27]. Another method is to collect the analyte from HPLC column, which is a labor-intensive process and there is a risk of cross-contamination between samples [28,29]. In this study, several solid-phase extraction (SPE) columns and elution conditions were tried for purification of 3-EtAde and 7EtGua in urine, including silica gel (hydrophilic interaction liquid chromatography, HILIC), mix-mode cation exchange (MCX), nonendcapped reversed phase (C18-OH), and a polymeric reversed phase (PRP) SPE column. Poor chromatograms were obtained for both 3-EtAde and 7-EtGua analysis after silica SPE cleanup. Using MCX or C18-OH alone resulted in severe ion suppression and the added [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtGua were barely detectable. Employing MCX followed by C18-OH SPE, the resulting fraction gave satisfactory chromatograms for analysis of 3-EtAde, but not for 7-EtGua (data not shown). Polymeric reversed phase SPE provides the most satisfactory results in terms of removing most of the interference from the matrix. The urinary ethylpurine adducts were eventually enriched by the PRP SPE column which leads to clean nanoLC-NSI/MS/MS chromatograms for monitoring of 3-EtAde, 7-EtGua, [13 C1 ,15 N2 ]3-EtAde, and [15 N5 ]7EtAGua.
71
After urine was enriched by the SPE column, there are still many compounds in the fraction collected together with the adducts and the internal standards. High sensitivity can be achieved when a column with a small inner diameter (i.d.) eluting at low flow rate is employed. We and others have used a 75 m-i.d. column for analysis of DNA adducts in tissue and leukocyte DNA [9,18,30,31]. The matrix effect in DNA hydrolysate is smaller than that in the complex mixture of urine partly because DNA is purified to a certain extent before enzyme hydrolysis. Although the small i.d. column can provide high sensitivity in the MS, a 180 m-i.d. column was used in this study for analysis of urinary 3-EtAde and 7-EtGua to avoid residues in the urine sample affecting the durability of the analytical column. Compared to electrospray ionization (ESI), nanospray or nanoelectrospray ionization (NSI) method operates at sufficiently low flow rates and analyte concentrations, which results in small droplets and high ionization efficiency in the mass spectrometry [32]. The coupling of nanoflow LC to NSI-MS leads to further gains in sensitivity. Among mass spectrometers, the use of static modes on both quadrupole 1 (Q1) and quadrupole 3 (Q3) of a triple quadrupole mass spectrometer offers the highest sensitivity possible. To increase the specificity of the measurement, the highly selective reaction monitoring (H-SRM) mode was employed in this study with the mass width at Q1 and Q3 being at 0.2 and 0.7 Da, respectively. 3.2. Method performance and validation The detection limits (S/N ≥ 3) of 3-EtAde and 7-EtGua injected on-column were 15 fg (92 amol) and 10 fg (56 amol), respectively. The calibration curves were obtained by adding various amount of the standards (0–20 pg) to a urine sample (No. 37) with non-detectable amounts of 3-EtAde and 7-EtGua (Fig. 1S, Supplementary material). The lowest amount of analyte showing linearity, i.e. the lower limit of quantification (LLOQ) with S/N ≥ 10, was 150 fg for both 3-EtAde and 7-EtGua. This gives the concentration quantification limit of 1.5 pg/mL, starting with 0.1 mL of urine. The precision of the assay was assessed by analyzing 3-EtAde and 7-EtGua in a urine sample (No. 21) in triplicates, and the experiments were performed on three different days using the optimized assay conditions. The results revealed that the relative standard deviations for both intraday and interday analyses were within 10% (Table 1S, Supplementary material), showing the high reproducibility of the assay. The accuracy of this assay was evaluated by adding known amount of 3-EtAde and 7-EtGua (6, 12, 18 pg each) to 0.1 mL of a urine sample (No. 6) and measured the total amount. The yintercept obtained by linear regression was 6.20 and 1.77 pg for 3-EtAde and 7-EtGua, respectively. The levels are very close to the concentration of 3-EtAde and 7-EtGua in this sample without addition of standards, i.e. 6.13 and 1.67 pg in 0.1 mL (or 61.3 and 16.7 pg/mL), respectively (Fig. 1S, Supplementary material). The matrix effect for 3-EtAde and 7-EtGua in urine was 56% and 34%, respectively, determined by the decrease in the peak area of the internal standards [13 C1 ,15 N2 ]3-EtAde or [15 N5 ]7-EtGua after addition of a SPE-enriched urine sample (No. 6). 3.3. nanoLC-NSI/MS/MS analysis of 3-EtAde and 7-EtGua in human urine
Fig. 1. Assay procedure for urinary 3-EtAde and 7-EtGua by stable isotope dilution nanoLC-NSI/MS/MS.
Fig. 2 shows representative nanoLC-NSI/MS/MS chromatograms of 3-EtAde and 7-EtGua in a smoker’s urine (No. 2) under the HSRM mode. Using H-SRM method 1, 3-EtAde and 7-EtGua eluted at 10.7 and 13.4 min, co-eluting with their isotope-labeled standards [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtAGua, respectively (Fig. 2A).
72
H.-J.C. Chen, C.-R. Lin / J. Chromatogr. A 1322 (2013) 69–73 Table 1 Spearman correlation coefficient (␥) between urinary 3-EtAde and 7-EtGua levels and numbers of cigarettes smoked per day and smoking index.a , b 3-EtAde
7-EtGua
Number of cigarettes smoked per day 0.6600 (<0.0001) 0.9011(< 0.0001) Smoking index 0.8670(< 0.0001) a b
0.6824 (<0.0001)
The two-tailed p value is shown in the parenthesis. n = 41.
coefficient of 0.8670 (p < 0.0001) and 0.6824 (p < 0.0001), respectively (Table 1). 3.4. Comparison of 3-EtAde and 7-EtGua levels
Fig. 2. nanoLC-NSI/MS/MS chromatograms of 3-EtAand 7-EtG in a smoker’s urine (No. 2) under H-SRM (A) method 1 and (B) method 2.
Co-elution of 3-EtAde and 7-EtGua with their isotope-labeled standards was also observed using H-SRM method 2. However, H-SRM method 2 was less sensitive than method 1, based on the smaller peak intensity for both [13 C1 ,15 N2 ]3-EtAde and [15 N5 ]7-EtAGua. In addition, the interference in the channels of 3-EtAde and 7EtGua was more severe using H-SRM method 2 (Fig. 2B). Therefore, H-SRM method 1 was used for quantification of adducts, while HSRM method 2 was used for qualitative assurance of the presence of these adducts. This assay requires only 0.1 mL of urine and an equivalent of 20 L of processed urine was injected on-column. Concentrations of urinary 3-EtAde and 7-EtGua in smokers and nonsmokers were determined and listed in Table 2S, Supplementary material. Adduct concentrations of 3-EtAde and 7-EtGua in 21 smokers were 68.6 ± 29.4 and 18.7 ± 13.8 pg/mL, respectively. In 20 nonsmokers, adduct concentrations of 3-EtAde and 7-EtGua were 3.5 ± 3.8 and 2.4 ± 3.0 pg/mL, respectively. Levels in 4 out of 20 nonsmokers were not detectable. As shown in Fig. 3, the concentrations of 3-EtAde and 7-EtGua are statistically significantly higher in smokers than in nonsmokers (p < 0.0001) performed by the nonparametric Mann–Whitney test. Moreover, there are positive associations between the concentrations of 3-EtAde and 7-EtGua and the number of cigarettes smoked per day with a correlation coefficient of 0.9011 and 0.6600, respectively (p < 0.0001) using the nonparametric Spearman correlation. Additionally, there is a statistically significant association between the smoking index (number of cigarettes smoked per day × years smoked) and the urinary concentrations of 3-EtAde and 7-EtGua with a correlation
The results indicated that the average urinary concentration of 3-EtAde is ca. 3.6 times higher than that of 7-EtGua, while our previous study reported a 1.7 times higher mean level of 3-EtAde compared to 7-EtGua in smokers’ leukocytes DNA [17]. Urinary excretion of 3-EtAde and 7-EtGua attributed from individual’s removal and repair activity in addition to nonenzymatic spontaneous depurination. These adducts are recognized by different enzymatic activities which remove and repair at various rates with efficiencies depending on the adduct levels and enzyme contents in the cell [33]. The half-lives of 3-EtAde and 7-EtGua induced by Nethyl-N-nitrosourea were reported to be ca. 12 and 75 h in rat liver, respectively, and 16 and 90 h in rat brain, respectively [34]. However, there is no report on the half-lives of 3-EtAde and 7-EtGua in human tissues so far. While the removal and repair efficiency of 3-EtAde and 7-EtGua in human body is not yet understood, this assay provides a valuable tool in measuring urinary 3-EtAde and 7-EtGu. Combination of data from urinary concentration and DNA adduct level in tissues from an individual should shed light in our understanding of formation and repair of 3-EtAde and 7-EtGua in the whole body. 3.5. Potential dietary effect on urinary 3-EtAde and 7-EtGua excretion Factors other than smoking, such as diet, in contribution to ethylpurine adducts in tissues and in urine have been investigated. In contrary to 3-methyladenine, change of diet only affects urinary excretion of 3-EtAde very slightly in smoking subjects and in subjects ingesting 3-EtAde [3,5]. N-nitrosodiethylamine and Nethyl-N-nitrosourea are the possible dietary sources of ethylating agent humans expose to. N-Nitrosodiethylamine occurs in human diet, such as in cured meat [35]. Hepatocarcinogenesis in rainbow trout treated with N-nitrosodiethylamine is modulated by indole-3-carbinol and beta-naphthoflavone. Liver 7-EtGua levels are reduced in indole-3-carbinol-pretreated fish and increase in -naphthoflavone-pretreated fish [36]. Vitamin supplementation can modulate the tumorigenicity of N-ethyl-N-nitrosourea. For instance, the N-ethyl-N-nitrosourea-induced tumorigenicity in rat brain is affected by dietary vitamin A or N-acetylcystein [37]. Niacin decreases the incidence of nonlymphocytic leukemia in rats induced by N-ethyl-N-nitrosourea [38]. Furthermore, fatty acid composition in the diet influences on N-ethyl-N-nitrosoureainduced mammary tumor incidence in the rat [39]. Recently, ketogenic diet is shown to significantly extend the lifespan of Nethyl-N-nitrosourea-treated mice which could otherwise die from lethal cardiomyopathy due to mutation in a critical gene [40]. The assay developed in this study should provide a valuable tool in evaluation of specific dietary ingredients in urinary excretion of these ethylpurine adducts.
H.-J.C. Chen, C.-R. Lin / J. Chromatogr. A 1322 (2013) 69–73
73
Fig. 3. Comparison of urinary (A) 3-EtAde and (B) 7-EtGua concentrations in smokers and in nonsmokers.
4. Conclusion To our knowledge, this the first study for simultaneous quantification of 3-EtAde and 7-EtGua in human urine, which allows comparison of levels of these two ethylpurine adducts. This nanoLC-NSI/MS/MS assay is highly accurate, sensitive and specific, which requires as little as 0.1 mL of urine to provide two ethylpurine adducts as noninvasive biomarkers of smoking-associated DNA damage. Conflict of interest statement The authors declare that there are no conflicts of interest. Funding sources This work was supported by National Science Council of Taiwan (Grant NSC 100-2113-M-194-002-MY3) and National Chung Cheng University (to H.-J.C.C.). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma. 2013.10.092. References [1] IARC, IARC Monogr. Eval. Carcinog. Risks Hum., International Agency for Research on Cancer, Lyon, France, 2004, pp. 53. [2] S.S. Hecht, Int. J. Cancer 131 (2012) 2724. [3] V. Prevost, D.E. Shuker, M.D. Friesen, G. Eberle, M.F. Rajewsky, H. Bartsch, Carcinogenesis 14 (1993) 199. [4] A. Kopplin, G. Eberle-Adamkiewicz, K.H. Glusenkamp, P. Nehls, U. Kirstein, Carcinogenesis 16 (1995) 2637. [5] V. Prevost, D.E. Shuker, Chem. Res. Toxicol. 9 (1996) 439. [6] R. Godschalk, J. Nair, H.C. Kliem, M. Wiessler, G. Bouvier, H. Bartsch, Chem. Res. Toxicol. 15 (2002) 433.
[7] R. Godschalk, J. Nair, F.J. van Schooten, A. Risch, P. Drings, K. Kayser, H. Dienemann, H. Bartsch, Carcinogenesis 23 (2002) 2081. [8] M.R. Chao, C.J. Wang, L.W. Chang, C.W. Hu, Carcinogenesis 27 (2006) 146. [9] H.J. Chen, Y.C. Wang, W.P. Lin, Anal. Chem. 84 (2012) 2521. [10] C.W. Hu, M.R. Chao, Chem. Res. Toxicol. 25 (2012) 2386. [11] L. Verna, J. Whysner, G.M. Williams, Pharmacol. Ther. 71 (1996) 57. [12] L. Den Engelse, G.J. Menkveld, R.J. De Brij, A.D. Tates, Carcinogenesis 7 (1986) 393. [13] L. Anna, K. Kovacs, E. Gyorffy, B. Schoket, J. Nair, Mutagenesis 26 (2011) 523. [14] L. Chen, M. Wang, P.W. Villalta, S.S. Hecht, Chem. Res. Toxicol. 20 (2007) 1498. [15] S.M. Bronstein, T.R. Skopek, J.A. Swenberg, Cancer Res. 52 (1992) 2008. [16] S. Feng, H.J. Roethig, Q. Liang, R. Kinser, Y. Jin, G. Scherer, M. Urban, J. Engl, K. Riedel, Biomarkers 11 (2006) 28. [17] H.J. Chen, Y.F. Liu, J. Chromatogr. A 1271 (2013) 86. [18] S. Balbo, P.W. Villalta, S.S. Hecht, Chem. Res. Toxicol. 24 (2011) 1729. [19] J.A. Swenberg, M.C. Dyroff, M.A. Bedell, J.A. Popp, N. Huh, U. Kirstein, M.F. Rajewsky, Proc. Natl. Acad. Sci. U. S. A. 81 (1984) 1692. [20] J. Thomale, N.H. Huh, P. Nehls, G. Eberle, M.F. Rajewsky, Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 9883. [21] D.E. Shuker, P.B. Farmer, Chem. Res. Toxicol. 5 (1992) 450. [22] W.J. Bodell, B. Singer, G.H. Thomas, J.E. Cleaver, Nucleic Acids Res. 6 (1979) 2819. [23] B. Singer, T.P. Brent, Proc. Natl. Acad. Sci. U. S. A. 78 (1981) 856. [24] M.W. Himmelstein, P.J. Boogaard, J. Cadet, P.B. Farmer, J.H. Kim, E.A. Martin, R. Persaud, D.E. Shuker, Crit. Rev. Toxicol. 39 (2009) 679. [25] H.J. Chen, Mol. Nutr. Food Res. 55 (2011) 1391. [26] R. Singh, P.B. Farmer, Carcinogenesis 27 (2006) 178. [27] T.Y. Yen, S. Holt, R. Sangaiah, A. Gold, J.A. Swenberg, Chem. Res. Toxicol. 11 (1998) 810. [28] F. Bianchini, F. Donato, H. Faure, J.L. Ravanat, J. Hall, J. Cadet, Int. J. Cancer 77 (1998) 40. [29] J.L. Ravanat, P. Guicherd, Z. Tuce, J. Cadet, Chem. Res. Toxicol. 12 (1999) 802. [30] H.J. Chen, W.P. Lin, Anal. Chem. 83 (2011) 8543. [31] H.J. Chen, G.J. Lin, W.P. Lin, Anal. Chem. 82 (2010) 4486. [32] R.D. Smith, Y. Shen, K. Tang, Acc. Chem. Res. 37 (2004) 269. [33] B. Singer, Environ. Health Perspect. 62 (1985) 41. [34] R. Goth, M.F. Rajewsky, Proc. Natl. Acad. Sci. U. S. A. 71 (1974) 639. [35] A.R. Tricker, R. Preussmann, Mutat. Res. 259 (1991) 277. [36] A.T. Fong, J.D. Hendricks, R.H. Dashwood, S. Van Winkle, B.C. Lee, G.S. Bailey, Toxicol. Appl. Pharmacol. 96 (1988) 93. [37] D.A. Ross, P. Kish, K.M. Muraszko, M. Blaivas, M. Strawderman, J. Neurooncol. 40 (1998) 29. [38] A.P. Bartleman, R. Jacobs, J.B. Kirkland, Nutr. Cancer 60 (2008) 251. [39] C.M. Williams, K. Maunder, Br. J. Nutr. 71 (1994) 543. [40] P. Krebs, W. Fan, Y.H. Chen, K. Tobita, M.R. Downes, M.R. Wood, L. Sun, X. Li, Y. Xia, N. Ding, J.M. Spaeth, E.M. Moresco, T.G. Boyer, C.W. Lo, J. Yen, R.M. Evans, B. Beutler, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 19678.