ARTICLE pubs.acs.org/ac
Quantitative Analysis of Multiple Exocyclic DNA Adducts in Human Salivary DNA by Stable Isotope Dilution Nanoflow Liquid ChromatographyNanospray Ionization Tandem Mass Spectrometry Hauh-Jyun Candy Chen* and Wen-Peng Lin Department of Chemistry and Biochemistry, National Chung Cheng University, 168 University Road, Ming-Hsiung, Chia-Yi 62142, Taiwan
bS Supporting Information ABSTRACT: Exocyclic DNA adducts, including 1,N2-propano-20 -deoxyguanosine derived from acrolein (AdG) and crotonaldehyde (CdG) and the three lipid peroxidation-related etheno adducts 1,N6-etheno-20 -deoxyadenosine (εdAdo), 3, N4-etheno-20 -deoxycytidine (εdCyt), and 1,N2-etheno-20 -deoxyguanosine (1,N2-εdGuo), play an important role in cancer formation and they are associated with oxidative-stress-induced DNA damage. Saliva is an easily accessible and available biological fluid and a potential target of noninvasive biomarkers. In this study, a highly sensitive and specific assay based on isotope dilution nanoflow LCnanospray ionization tandem mass spectrometry (nanoLCNSI/MS/MS) is developed for simultaneous detection and quantification of these five adducts in human salivary DNA. The levels of AdG, CdG, εdAdo, εdCyd, and 1,N2-εdGuo, measured in 27 human salivary DNA samples from healthy volunteers, were determined as 104 ( 50, 7.6 ( 12, 99 ( 50, 72 ( 49, 391 ( 198 (mean ( SD) in 108 normal nucleotides, respectively, starting with 25 μg of DNA isolated from an average of 3 mL of saliva. Statistically significant correlations were found between levels of εdAdo and εdCyd (γ = 0.8007, p < 0.0001), between levels of εdAdo and 1,N2-εdGuo (γ = 0.6778, p = 0.0001), between levels of εdCyd and 1,N2-εdGuo (γ = 0.5643, p = 0.0022), between levels of AdG and 1,N2-εdGuo (γ = 0.5756, p = 0.0017), and between levels of AdG and εdAdo (γ = 0.3969, p = 0.0404). Only 5 μg of DNA sample was analyzed for simultaneous quantification of these adducts. The easy accessibility and availability of saliva and the requirement for the small amount of DNA samples make this nanoLCNSI/MS/MS assay clinically feasible in assessing the possibility of measuring 1,N2-propano-20 deoxyguanosine and etheno adducts levels in human salivary DNA as noninvasive biomarkers for DNA damage resulting from oxidative stress and for evaluating their roles in cancer formation and prevention.
D
NA adduction plays an important role in the initiation stage of the multistage carcinogenesis processes.1,2 It is, however, an analytical challenge to elucidate the role of DNA adducts in cancer formation, progression, and diagnosis by accurately quantifying DNA adducts, because they are frequently present in very low amounts. Clinical DNA samples are normally not available in large quantities, except when they are obtained from surgical procedures. Therefore, an assay with high accuracy, sensitivity, and specificity is demanded. To achieve these goals, liquid chromatography mass spectrometry (LCMS)-based methodology allows direct analyses of polar DNA adducts without derivatization. In recent years, great advancements in LC interfaces and MS ionization technologies have been achieved to enhance the sensitivity. The increase in sensitivity of the analysis reduced the amount of DNA samples required for the assay and is required for biomonitoring studies. To further increase the sensitivity of LCMS analysis, the coupling of capillary or nanoflow LC with nanoelectrospray or nanospray (NSI) ionization MS has shown promise for DNA adduct analysis.3,4 Furthermore, MS-based assays can provide accurate r 2011 American Chemical Society
quantification of the analytes with the use of stable isotopomers as internal standards, which are structurally identical to the analytes. Therefore, errors due to the recovery in each step of the assay procedure and due to the matrix effect can be corrected by using stable isotope-labeled internal standards. Most importantly, stable-isotope internal standards can serve as carriers for the ultratrace amounts of the analytes present in the biological samples; such small amounts can be easily lost during the multistep assay procedures. For these reasons, isotopic dilution mass spectrometry (IDMS) has been widely applied for the specific detection and accurate quantification of low-abundance DNA adducts in vivo.5 Humans are constantly exposed to reactive α,β-unsaturated aldehydes, such as acrolein and crotonaldehyde, which are present in the environment, including the atmosphere, cigarette smoke, and diesel exhaust.68 Acrolein is also generated during Received: July 21, 2011 Accepted: September 26, 2011 Published: September 29, 2011 8543
dx.doi.org/10.1021/ac201874d | Anal. Chem. 2011, 83, 8543–8551
Analytical Chemistry Scheme 1. Structures of AdG, CdG, εdAdo, εdCyd, and 1,N2εdGuo
cooking of carbohydrates, oils, fats, and amino acids.9 Both acrolein and crotonaldehyde are derived endogenously from lipid peroxidation, forming the mutagenic exocyclic 1,N2-propano-20 -deoxyguanosine derived from acrolein (AdG) and crotonaldehyde (CdG) (Scheme 1).1013 In the brain tissue DNA of Alzheimer patients, levels of AdG is found to be significantly higher than those of control subjects.14 The three etheno adducts 1,N6-etheno-20 -deoxyadenosine (εdAdo), 3,N4-etheno-20 -deoxycytidine (εdCyd), and 1,N2etheno-20 -deoxyguanosine (1,N2-εdGuo) (Scheme 1) are mutagenic DNA lesions derived from exogenous industrial chemicals or environmental pollutants.1517 These adducts are also generated from endogenous lipid peroxidation, and they play an important role in cancer formation associated with oxidative stress, inflammation, and atherosclerosis.10,1821 In our previous studies, separate assays for 1,N2-propano-20 deoxyguanosine adducts and for etheno adducts were developed and applied for analyses in human placental and leukocyte DNA using isotope dilution nanoflow LCnanospray ionization tandem mass spectrometry (nanoLCNSI/MS/MS).22,23 To save time and resources, these five adducts can be simultaneously detected in human salivary DNA, and their levels were accurately quantified in this study. Saliva is a readily available source of DNA that has been used as a target of biomarkers associated with oral cancer.2427 DNA isolated from saliva is predominantly from leukocytes and epithelial buccal cells.28,29 The turnover time for oral epithelia is between 5 and 12 days,30 and the gingival crevice-originated leukocytes are also short-lived.31 Thus, saliva may be a practical and useful target of noninvasive biomarkers for monitoring recent status on carcinogen- or oxidative-stress-associated DNA damage, which could be incorporated as tools in intervention studies. The aim of this study is to develop an assay for simultaneous analysis of multiple adducts so that levels of these adducts can be compared in a single experiment. It is also economical because labor, consumable materials, and instrument time are saved compared to performing analysis for individual adducts. This assay is applied to human DNA samples isolated from saliva, a convenient and noninvasive source of DNA for evaluation of DNA damage in vivo.
’ EXPERIMENTAL SECTION Materials. All the enzymes used for DNA hydrolysis were obtained from Sigma Chemical Co. (St. Louis, MO), except alkaline phosphatase, which was from Calbiochem Chemical Co.
ARTICLE
(La Jolla, CA). The isotope standards [15N5]εdAdo, [15N3] εdCyd, [13C1,15N2]1,N2-εdGuo, [15N5]AdG, and [15N5]CdG were synthesized from the 20 -deoxyribonucleosides [15N5]dAdo, [15N3]dCyd, [13C1,15N2]dGuo, and [15N5]dGuo (Cambridge Isotope Laboratories, Andover, MA) as reported previously.3236 All reagents are of reagent grade or higher. Human Salivary DNA Isolation. The subjects of this study were healthy volunteers. They received a written warranty stating that the information was for research purposes only and that their personal information would be kept confidential. The subjects were asked not to eat any food 1 h before saliva collection. They bushed their teeth with toothpaste and rinsed their mouths thoroughly before saliva collection. The saliva was subjected to a Blood DNA Extraction Midiprep System (Viogen, Sunnyvale, CA) following the previously reported procedures with modifications.22,23 Typically, to every 3 mL of saliva was added proteinase K (25 mg/mL) in 3 mL of the extraction buffer. The mixture was vortexed for 20 s and incubated at 60 °C for 84 min, followed by at 70 °C for 56 min. The mixture was vortexed every 7 min during the incubation. Cold ethanol (99.5%, 3 mL) was added to the mixture and the mixture was vortexed. Then, the mixture was loaded onto the Genomic DNA Midi column and centrifuged at 2500g for 3 min. The filtrate was discarded and the precipitate was washed three times with the wash buffer. The column was centrifuged at 4900g for 5 min to dry the column. To the column was added 3 mL of 70 °C double distilled water, the column was incubated at 70 °C for 5 min, and salivary DNA was eluted by centrifugation at 2500g for 10 min. The elution process was repeated three times to maximize the yield of DNA. The amount of DNA was quantified by a NanoDrop 1000 photometer (J&H Technology Co., Ltd., Wilmington, DE). The purity of DNA was checked by the absorbance ratio A260/A280 being between 1.7 and 2.0. The entire extraction procedures yielded an average of 9.3 ( 4.5 μg of DNA per milliliter saliva (mean ( SD) with a range of 2.420 μg/mL. For samples with low DNA content, more saliva was collected. Enzyme Hydrolysis of DNA. To the solution containing salivary DNA (25 μg) were added aliquots containing 100 pg each of [15N5]AdG, [15N5]CdG, [15N5]εdAdo, [15N3]εdCyd, and [13C1,15N2]1,N2-εdGuo, and the mixture was subjected to the following enzyme hydrolysis conditions A or B. Method A.22. A solution containing DNA (25 μg) with the five stable isotope-labeled internal standards was dissolved in 10 mM sodium succinate/5 mM CaCl2 buffer (pH 7.0) and the mixture was incubated at 100 °C for 30 min and cooled to room temperature. Enzyme hydrolysis was performed by incubation with 3 units of micrococcal nuclease (from Staphylococcus aureus) and 0.01 unit of phosphodiesterase II (from bovine spleen) at 37 °C for 6 h. Then, 0.9 unit of adenosine deaminase (from bovine spleen) and 6 units of alkaline phosphatase (from calf intestine) were added, and the mixture was incubated at 37 °C overnight. Method B.23. A solution containing DNA (25 μg) with the internal standards was dissolved in 50 mM ammonium acetate buffer (pH 5.3) and then 2.4 units of nuclease P1 were added in the mixture. The mixture was incubated at 45 °C for 2 h. Enzymatic hydrolyte was neutralized by addition of Tris-HCl buffer (pH 7.4) and incubation with 0.0024 units of phosphodiesterase I (from Crotalus adamanteus venom) at 37 °C for 2 h. Finally, 0.9 units of adenosine deaminase (from bovine spleen) and 0.6 units of alkaline phosphatase were added, and the mixture was incubated at 37 °C for 1 h. 8544
dx.doi.org/10.1021/ac201874d |Anal. Chem. 2011, 83, 8543–8551
Analytical Chemistry
ARTICLE
AdG, εdAdo, and 1,N2-εdGuo, the product ions are [M + H 116]+ ([M + H dR]+) ions. As to CdG and εdCyd, the product ion is [M + H 160]+ ([M + H dR C2H4O]+) and [M + H 171]+ ([M + H dR CO HCN]+) ion, respectively. The structures of fragment ions of AdG, CdG, εdAdo, εdCyd, 1,N2-εdGuo and their stable isotope-labeled standards employed for the MRM transitions are summarized in Figure 1S of Supporting Information. Statistical Analysis. GraphPad InStat version 3.00 for Windows 95 (GraphPad Software, San Diego, CA; www.graphpad. com) was used for statistical analysis. The correlation between two adduct levels was calculated using Pearson linear correlation.
Adduct Enrichment. The hydrolyte was filtered through a
0.22 μm Nylon syringe filter and then purified using a solid-phase extraction (SPE) column [Bond Elut C18, 100 mg, 1 mL, Varian (Harbor City, CA)] after conditioning with methanol, followed by water. The SPE column was washed with 3 mL of H2O, 1 mL of 5% aqueous MeOH, and collected with 1 mL of 25% aqueous MeOH. The fraction containing five adducts was dried and redissolved in 10 μL of 0.1% acetic acid and passed through a 0.22 μm Nylon syringe filter. Two microliters of the processed sample was subjected to the nanoLCNSI/MS/MS analysis described below. NanoLCNSI/MS/MS Analysis. A 2-μL injection loop was connected to a six-port switching valve injector in a LC system consisting of an UltiMate 3000 Nano LC system (Dionex, Amsterdam, Netherlands) and a reversed phase tip column (75 μm 130 mm, 5 μm) packed in-house (MAGIC C18AQ, 200 Å, 5 μm, Michrom BioResource, Auburn, CA). The pump output (30 μL/min) was split before the injection port to a flow rate of 300 nL/min. Mobile phase A was 1% acetic acid (pH 3.2), and mobile phase B was acetonitrile containing 0.1% acetic acid. The elution started with a linear gradient of 5% mobile phase B to 30% mobile phase B in the 20 min. 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. The column effluent enters the spray chamber through a tapered emitter constructed from a 75-μm i.d. fused-silica capillary and is directly electrosprayed into the mass spectrometer under the positive ion mode for the NSI-MS/MS. The spray was monitored by a built-in CCD camera. The spray voltage was 1.6 kV, and the source temperature was at 220 °C. Argon was used as the collision gas in MS/MS experiments. The adduct enriched sample was analyzed by nanoLCNSI/MS/MS with the transition from the parent 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 selective reaction monitoring (H-SRM) mode with the mass width of Q1 and Q3 being 0.2 and 0.7 m/z, respectively, and a dwell time of 0.1 s. The transition from the parent ion [M + H]+ focused in quadrupole 1 (Q1) and dissociated in a collision cell (quadrupole 2, Q2) yielding the product ion in quadrupole 3 (Q3). The ions chosen at Q1 and Q3 and the collision energy at Q2 are listed in Table 1. For
Scheme 2. Procedures for Simultaneous Analysis of AdG, CdG, εdAdo, εdCyd, and 1,N2-εdGuo by NanoLCNSI/ MS/MS
Table 1. MRM Transitions for the Five Adducts and Their Corresponding Isotopes MRM condition 2a
MRM condition 1
a
Q1
Q3
collision energy (eV)
Q3
collision energy (eV)
AdG
324.0
207.8
15
164.0
35
[15N5]AdG
329.0
212.8
15
169.0
35
CdG
338.0
178.0
35
[15N5]CdG
343.0
183.0
35
εdAdo
276.1
160.0
20
119.1
45
[15N5]εdAdo
281.1
165.0
20
123.1
45
εdCyd
252.1
81.0
45
108.1
35
[15N3]εdCyd 1,N2-εdGuo
255.1 292.1
83.0 176.0
45 25
111.1 148.1
35 35
[15N2,13C1]1,N2-εdGuo
295.1
179.0
25
151.1
35
Same m/z values at Q1 were used in MRM condition 2. 8545
dx.doi.org/10.1021/ac201874d |Anal. Chem. 2011, 83, 8543–8551
Analytical Chemistry
ARTICLE
’ RESULTS AND DISCUSSION
the DNA solution was added five internal standards, namely, [15N5]AdG, [15N5]CdG, [15N5]εdAdo, [15N3]εdCyd, and [13C1,15N2]1,N2-εdGuo, and the mixture was subjected to the optimized enzyme hydrolysis conditions. These adducts were enriched from the enzyme hydrolysate by a reversed-phase solidphase extraction (SPE) column and analyzed by the nanoLCNSI/MS/MS under the highly selective reaction monitoring (H-SRM) mode (Scheme 2). The optimized collision energy and H-SRM transitions are listed in Table 1 as MRM condition 1. Except CdG and εdCyd, the daughter ions are mostly the protonated base adduct ions after losing the deoxyribse moiety from the molecular ions. Interferences in the channels of protonated base adduct ion of CdG and εdCyd forbid accurate quantification.22,23 The structures of fragment ions employed for the optimized MRM transitions used in this assay were summarized in Figure 1S of Supporting Information. The detection limits of these five adducts are comparable to those reported.22,23 Enzyme Hydrolysis. Our earlier studies demonstrated that hydrolysis of the adducted nucleosides from DNA greatly depends on the types and amounts of the hydrolytic enzymes used as well as the pH and incubation time of the hydrolysis.22,23,37 In this study, we compare the enzyme hydrolysis
Method Development. The saliva donors brushed their teeth beforehand and saliva was collected in an unstimulated manner by expectorating (spitting) into the container. Salivary DNA was isolated from fresh saliva using modified procedures with the DNA extraction kit originally designed for isolating leukocyte DNA from blood. The yield of DNA was quantified by spectrophotometer, and 2.420 μg of DNA per milliliter saliva was obtained, which is comparable to the reported methods.24,25 To
Table 2. Comparison of the Five Exocyclic DNA Adduct Levels Obtained by Two Enzyme Hydrolysis Methods mean adduct levels (adducts/108 nucleotides) ( SD (RSD, %) adduct
a
method Aa
method Bb
AdG
218 ( 11 (5.3)
43.4 ( 3.6 (8.3)
CdG
6.33 ( 0.43 (6.8)
5.83 ( 0.56 (9.6)
εdA
99.2 ( 3.8 (3.8)
117 ( 2 (1.7)
εdC
67.7 ( 6.2 (9.2)
66.1 ( 2.7 (4.1)
1,N2-εdG
672 ( 11 (1.7)
333 ( 9 (2.5)
Adapted from ref 22. b Adapted from ref 23.
Table 3. Precision in Quantification of the Five Adducts by NanoLCNSI/MS/MS Analysis mean adducts levels (adducts/108 nucleotides) ( SD (RSD, %) a,b day 1
day 2
108 ( 3.8 (3.6) 3.4 ( 0.2 (6.5)
111 ( 0.9 (0.8) 3.9 ( 0.2 (5.1)
day 3
interday variation
111 ( 0.5 (0.5) 3.3 ( 0.07 (2.2)
110 ( 3.0 (2.8) 3.6 ( 0.32 (9.1)
No. 26 AdG CdG εdAdo
111 ( 0.8 (0.7)
121 ( 2.8 (2.3)
119 ( 4.0 (3.3)
117 ( 5.6 (4.8)
εdCyd
70.7 ( 6.6 (9.3)
70.3 ( 2.5 (3.5)
68.1 ( 4.8 (7.0)
69.7 ( 4.7 (6.7)
1,N2-εdGuo
456 ( 44 (9.7)
441 ( 17 (3.9)
432 ( 9.8 (2.3)
443 ( 26 (5.8)
AdG
156 ( 4.2 (2.7)
No. 27 163 ( 2.6 (1.6)
163 ( 3.8 (2.3)
161 ( 5.4 (3.4)
CdG
9.4 ( 0.1 (1.5)
9.7 ( 0.2 (2.2)
9.1 ( 0.2 (2.1)
9.4 ( 0.4 (3.7)
εdAdo
219 ( 4.9 (2.2)
221 ( 1.7 (0.7)
210 ( 3.9 (1.8)
219 ( 4.9 (2.2)
εdCyd 1,N2-εdGuo
118 ( 5.4 (4.6) 518 ( 41 (8.0)
109 ( 6.6 (6.1) 543 ( 4.2 (0.8)
106 ( 1.0 (0.9) 536 ( 12 (2.3)
118 ( 5.4 (4.6) 519 ( 41 (8.0)
Each experiment started with 25 μg of human salivary DNA, and an equivalent of 5 μg of DNA hydrolysate was subjected to the nanoLCNSI/MS/ MS analysis. b Adduct levels are presented as mean ( standard deviation (SD) from triplicate experiments. The percentage standard deviation (RSD) is expressed in parentheses. a
Table 4. Validation of Adduct Levels under Different MRM Conditions mean adducts levels (adducts/108 nucleotides) ( SD (RSD, %) a,b sample no. 26 MRM condition 1
sample no. 27 MRM condition 2
MRM condition 1
MRM condition 2
AdG
111 ( 0.5 (0.5)
106 ( 0.3 (0.3)
163 ( 3.8 (2.3)
145 ( 3.6 (2.5)
εdAdo
119 ( 4.0 (3.3)
128 ( 1.2 (1.0)
210 ( 3.9 (1.8)
195 ( 4.4 (2.2)
εdCyd
68.1 ( 4.8 (7.0)
60.4 ( 0.6 (1.0)
106 ( 1.0 (0.9)
102 ( 3.3 (3.3)
1,N2-εdGuo
432 ( 9.8 (2.3)
448 ( 10 (2.3)
536 ( 12 (2.3)
523 ( 20 (3.8)
Each experiment started with 25 μg of human salivary DNA, and an equivalent of 5 μg of DNA hydrolysate was subjected to the nanoLCNSI/MS/ MS analysis. b Adduct levels are presented as mean ( standard deviation (SD) from triplicate experiments. The percentage standard deviation (RSD) is expressed in parentheses. a
8546
dx.doi.org/10.1021/ac201874d |Anal. Chem. 2011, 83, 8543–8551
Analytical Chemistry
Figure 1. NanoLCNSI/MS/MS analysis under the H-SRM mode of AdG, CdG, εdAdo, εdCyd, and 1,N2-εdGuo in a human salivary DNA sample (no. 27) in the absence of isotope-labeled internal standards.
method previously optimized for AdG and CdG (method A)22 and that for the three etheno adducts (method B) on human salivary DNA.23 The results showed that method A is much more effective in releasing 5 times AdG and 2 times 1,N2-εdGuo compared to method B. The levels of CdG and εdCyd were comparable using both methods A and B, whereas the level of εdAdo is 18% higher using method B than using method A (Table 2). The latter is not surprising because method B was originally used for εdAdo analysis.32 Method A is chosen for this assay considering that the sensitivity of εdAdo is the highest, with a detection limit of 0.73 amol, among the five adducts analyzed. Method Validation. The calibration curves were constructed by plotting response of the LC/MS chromatogram, i.e., the peak area ratios of the adduct added (0.1300 pg) versus the corresponding isotope-labeled internal standard (100 pg), after enzyme hydrolysis and adduct enrichment procedures. The correlation coefficient (R2) for the five adducts was 0.9997, 0.9992, 0.9999, 0.9996, and 0.9999 for AdG, CdG, εdAdo, εdCyd, and 1,N2-εdGuo, respectively (Figure 2S, Supporting Information). The lowest amount of the analyte in the calibration curve showing linearity, i.e., the limit of quantification, was 0.1 pg (0.31 fmol), 0.5 pg (1.5 fmol), 0.1 pg (0.36 fmol), 0.5 pg (2.0 fmol), and 0.5 pg (1.7 fmol) for AdG, CdG, εdAdo, εdCyd, and 1,N2-εdGuo, respectively. In the control experiments containing only the isotope standards and the positive control experiments containing the isotopes and the enzymes, no signal of any adduct was detected (Figure 3S, Supporting Information), suggesting no interference or contamination in the analyte channels from the internal standards or from the hydrolytic enzymes.
ARTICLE
Figure 2. NanoLCNSI/MS/MS analysis of AdG, CdG, εdAdo, εdCyd, and 1,N2-εdGuo in human salivary DNA under the H-SRM mode. The level of AdG, CdG, εdAdo, εdCyd, and 1,N2-εdGuo in this sample (no. 27) was 152, 9.21, 223, 122, and 489 in 108 normal nucleotides, respectively.
The accuracy of the assay was confirmed by adding three known amounts of adduct standards to the human placental DNA (25 μg) and quantifying the total adduct levels. Extrapolation of the linearly regressed lines gave a y intercept of 55.2, 6.8, 17.3, 10.5, and 27.0 pg for AdG, CdG, εdAdo, εdCyd, and 1,N2εdGuo, respectively; the amount of adducts presented in the DNA without added standards was measured as 57.6, 7.7, 16.3, 9.7, and 26.3 pg, respectively (Figure 4S, Supporting Information). These results presented evidence for the excellent accuracy and quality control of this assay. The precision of the assay was evaluated by two human salivary DNA samples. Each sample was analyzed in triplicate per day for three separate days. The intraday %RSD of the five adducts in these two samples ranged from 0.7% to 9.7%, while the interday %RSD ranged from 2.9% to 9.1% (Table 3). The adduct levels, except of CdG, were further validated by analyzing two human salivary DNA samples using different MRM transitions. The daughter ion of CdG used in this assay (MRM conditions 1) was a fragment of the base adduct, which resulted in very small signal when further fragmented. Because levels of CdG were on average the lowest among the five adducts analyzed, further degradation of the product ion did not allow detection and quantification of CdG. Under MRM condition 2, higher collision energy, albeit with lower sensitivity, was used, except for εdCyd (Table 1). The calibration curves of MRM condition 2 also showed good linearity, with the correlation coefficient being 0.9982, 0.9999, 0.9994, and 0.9994 for AdG, εdAdo, εdCyd, and 1,N2-εdGuo, respectively (Figure 5S, Supporting Information). Consistent adduct 8547
dx.doi.org/10.1021/ac201874d |Anal. Chem. 2011, 83, 8543–8551
Analytical Chemistry
ARTICLE
Table 5. Adduct Levels in Human Salivary DNA mean adducts levels (adducts/108 nucleotides) a,b ( SD (RSD, %) sample
AdG
CdG
εdAdo
εdCyd
1,N2-εdGuo
1
13 ( 1 (10)
ND c
75.0 ( 3.3 (4.4)
40.5 ( 3.5 (8.6)
2
74 ( 1 (1.9)
2.51 ( 0.07 (2.8)
26.0 ( 1.7 (6.6)
ND
106 ( 0.9 (0.8)
3
35 ( 2 (6.6)
4.7 ( 0.4 (8.1)
55.2 ( 5.5 (10)
98.6 ( 1.4 (1.4)
162 ( 0.2 (0.2)
68.4 ( 2.4 (3.4)
4
32 ( 0.6 (1.8)
ND
31.2 ( 1.4 (4.5)
43.6 ( 5.1 (12)
144 ( 0.07 (0.05)
5
85 ( 1.9 (2.3)
ND
129 ( 1.4 (1.1)
107 ( 8 (7.8)
548 ( 11 (1.9)
6
155 ( 8 (5.2)
2.26 ( 0.05 (2.1)
100 ( 0.3 (0.3)
77.8 ( 6 (7.8)
560 ( 20 (3.5)
7
167 ( 4 (2.2)
1.55 ( 0.04 (2.5)
90.4 ( 1.8 (2.0)
95.7 ( 4.4 (4.6)
365 ( 3 (0.8)
8 9
132 ( 4.7 (3.6) 176 ( 19 (10.6)
6.4 ( 0.27 (4.3) 16.7 ( 0.6 (3.7)
153 ( 2.5 (1.6) 73.6 ( 3.0 (4.0)
139 ( 10 (7.4) 49.5 ( 4.3 (8.7)
752 ( 12 (1.6) 286 ( 4.7 (1.6) 290 ( 6.6 (2.3)
10
108 ( 9 (8.7)
5.22 ( 0.44 (8.3)
81.2 ( 4.0 (4.9)
52.4 ( 4.9 (9.3)
11
116 ( 3 (2.3)
ND
210 ( 3.1 (1.5)
229 ( 3.3 (1.4)
589 ( 5 (0.9)
12
158 ( 6 (3.5)
4.8 ( 0.2 (3.7)
155 ( 4.3 (2.8)
99.5 ( 1.2 (1.2)
634 ( 7 (1.2)
13
106 ( 1 (1.0)
1.1 ( 0.1 (11.1)
72.2 ( 0.8 (1.0)
10.8 ( 0.1 (1.2)
627 ( 17 (2.8)
14
108 ( 2 (1.5)
48.5 ( 3.0 (6.2)
65.6 ( 0.6 (0.9)
48.9 ( 2.8 (5.6)
269 ( 4 (1.4)
15
117 ( 3 (2.3)
4.0 ( 0.2 (5.8)
83.8 ( 1.7 (2.0)
25.2 ( 0.8 (3.3)
374 ( 2.1 (0.6)
16 17
114 ( 2 (2.1) 106 ( 8 (7.9)
6.7 ( 0.4 (5.9) 19.0 ( 0.8 (4.3)
63.1 ( 1.1 (1.8) 69.4 ( 0.6 (0.9)
29.7 ( 1.5 (5.0) 51.5 ( 3.3 (6.4)
233 ( 7 (3.2) 322 ( 6 (1.7)
18
81.5 ( 6.6 (8.1)
5.5 ( 0.4 (7.6)
113 ( 3 (2.4)
120 ( 11 (9.5)
670 ( 12 (1.7)
19
139 ( 2 (1.5)
43.0 ( 3.0 (7.0)
130 ( 6 (4.5)
105 ( 1.5 (1.4)
343 ( 7 (1.9)
20
81.7 ( 6.9 (8.4)
1.3 ( 0.1 (3.5)
94.3 ( 0.6 (0.6)
46.8 ( 3.8 (8.0)
377 ( 9 (2.5)
21
63.2 ( 3.4 (5.4)
2.8 ( 0.2 (8.2)
131 ( 7 (5.2)
44.4 ( 1.8 (4.1)
356 ( 17 (4.8)
22
66.2 ( 0.8 (1.2)
2.5 ( 0.1 (2.1)
173 ( 9 (5.2)
134 ( 15 (11)
466 ( 20 (4.3)
23
22.8 ( 0.3 (1.2)
ND
22.1 ( 1.6 (7.3)
11.6 ( 0.6 (5.5)
72.3 ( 3.5 (4.8)
24 25
65.5 ( 2.6 (4.0) 218 ( 11 (5.3)
4.5 ( 0.1 (2.5) 6.3 ( 0.4 (6.8)
49.0 ( 2.8 (5.7) 99.2 ( 3.8 (3.8)
43.6 ( 0.8 (1.9) 67.7 ( 6.2 (9.2)
397 ( 7 (1.7) 672 ( 11 (1.7)
26
111 ( 0.5 (0.5)
3.3 ( 0.07 (2.2)
119 ( 4.0 (3.3)
68.1 ( 4.8 (7.0)
432 ( 9.8 (2.3)
27
163 ( 3.8 (2.3)
9.1 ( 0.2 (2.1)
210 ( 3.9 (1.8)
106 ( 1.0 (0.9)
536 ( 12 (2.3)
mean ( SD
104 ( 50
7.5 ( 12
99 ( 50
72 ( 49
391 ( 198
range
13218
048.5
22210
0139
68752
Each experiment started with 25 μg of human salivary DNA, and an equivalent of 5 μg of DNA hydrolysate was subjected to the nanoLCNSI/MS/ MS analysis. The adduct levels from each sample were obtained from the average of triplicate experiments. b Mean ( standard deviation (SD) from 27 samples. c Not detectable. a
Table 6. Correlationa between Adduct Levels in 27 Human Salivary DNA Samples AdG AdG CdG εdAdo εdCyd
CdG
εdAdo
εdCyd
1,N2-εdG
0.2588
0.3969
0.2414
0.5756
(0.1923)
(0.0404)b
(0.2250)
(0.0017)c
0.1859
0.0073
0.9825
(0.9267)
(0.9707)
(0.6259)
0.8007
0.6778
(
