Abundance of Four Sulfur Mustard-DNA Adducts ex Vivo and in Vivo

Article pubs.acs.org/crt

Abundance of Four Sulfur Mustard-DNA Adducts ex Vivo and in Vivo Revealed by Simultaneous Quantification in Stable Isotope Dilution− Ultrahigh Performance Liquid Chromatography−Tandem Mass Spectrometry Lijun Yue,†,‡,§ Yuxia Wei,† Jia Chen,† Huiqin Shi,‡ Qin Liu,† Yajiao Zhang,† Jun He,‡ Lei Guo,*,† Tingfen Zhang,‡ Jianwei Xie,*,† and Shuangqing Peng‡ †

State Key Laboratory of Antitoxic Drugs and Toxicology, and Laboratory of Toxicant Analysis, Institute of Pharmacology and Toxicology, Academy of Military Medical Sciences, No. 27 Taiping Road, Haidian District 100850, Beijing, China ‡ Beijing Institute for Disease Control and Prevention, No. 20 Dongdajie Street, Fengtai District 100071, Beijing, China S Supporting Information *

ABSTRACT: Sulfur mustard (SM) is a highly reactive alkylating vesicant and causes blisters upon contact with skin, eyes, and respiratory organs. It covalently links with DNAs by forming four mono- or cross-link adducts. In this article, the reference standards of SM-DNA adducts and deuterated analogues were first synthesized with simplified procedures containing only one or two steps and using less toxic chemical 2-(2-chloroethylthio)ethanol or nontoxic chemical thiodiglycol as starting materials. A sensitive and high-throughput simultaneous quantification method of N7-[2-[(2-hydroxyethyl)thio]-ethyl]guanine (N7-HETEG), O6-[2-[(2-hydroxyethyl)thio]-ethyl]guanine (O6-HETEG), N3-[2-[(2-hydroxyethyl)thio]-ethyl]adenine (N3-HETEA), and bis[2-(guanin-7-yl)ethyl]sulfide (Bis-G) in the Sprague−Dawley rat derma samples was developed by stable isotope dilution− ultrahigh performance liquid chromatography−tandem mass spectrometry (ID-UPLC-MS/MS) with the aim of revealing the real metabolic behaviors of four adducts. The method was validated, the limit of detection (S/N ratio greater than 10) was 0.01, 0.002, 0.04, and 0.11 fmol on column for N7-HETEG, O6-HETEG, Bis-G, and N3-HETEA, respectively, and the lower limit of quantification (S/N ratio greater than 20) was 0.04, 0.01, 0.12, and 0.33 fmol on column for N7-HETEG, O6-HETEG, Bis-G, and N3-HETEA, respectively. The accuracy of this method was determined to be 76% to 129% (n = 3), and both the interday (n = 6) and intraday (n = 7) precisions were less than 10%. The method was further applied for the quantifications of four adducts in the derma of adult male Sprague−Dawley rats exposed to SM ex vivo and in vivo, and all adducts had time− and dose−effect relationships. To the best of our knowledge, this is the first time that the real presented status of four DNA adducts was simultaneously revealed by the MS-based method, in which Bis-G showed much higher abundance than the result previously reported and N3-HETEA showed much less. It should be noted that since the interstrand cross-linked adduct is believed to stall DNA replication and finally induce a double-strand break, the higher abundance of Bis-G is a great indication of a more serious DNA lesion by SM alkylation.

■

INTRODUCTION Sulfur mustard (bis(2-chloroethyl)sulphide, SM) is a wellknown, highly reactive alkylating vesicant and can cause blisters upon contact with skin, eyes, and respiratory organs. SM is also a predominant agent found in the chemical weapons abandoned in China by the Japanese, and has caused several civilian casualty incidents and environmental contamination.1−3 © 2014 American Chemical Society

As a highly reactive electrophile, it readily reacts with a variety of nucleophiles via the episulfonium ion.4−7 Previous studies showed that SM reacts with DNA by forming mono- or crosslinked adducts.8,9 As a consequence, DNA replication is Received: September 19, 2013 Published: January 27, 2014 490

dx.doi.org/10.1021/tx4003403 | Chem. Res. Toxicol. 2014, 27, 490−500

Chemical Research in Toxicology

Article

Figure 1. Chemical structures of four SM-DNA adducts and their corresponding deuterated analogues.

and accurate method is urgently required. In recent years, liquid chromatography−tandem mass spectrometry (LC-MS/MS) provides a powerful tool for DNA adduct quantification since such a technique can provide ultrahigh sensitivity as well as structural and unambiguous quantification information in an analysis cycle.16−24 In LC-MS/MS, accurate quantification of trace analytes in a complicated biological matrix requires an appropriate internal standard (IS). An isotopomer of identical structure to the analyte is an ideal IS because it has chemical and physical properties identical to those of the analyte except for the mass. A stable isotope-labeled IS can be used to precisely identify the peak of analyte in a complicated chromatogram based on almost the same retention times of the analyte and IS, which provides the highest possible accuracy and specificity for quantitative measurements in many reports.21−25 In the present work, N7-HETEG, N3-HETEA, O6-HETEG, Bis-G, and their corresponding deuterated analogues were successfully synthesized and characterized, and a stable isotope dilution−ultrahigh performance liquid chromatography−tandem mass spectrometry (ID-UPLC-MS/MS) method for simultaneous quantification of four SM-DNA adducts was developed and validated. The method was then applied to analyze SM-DNA adducts in the derma of male Sprague− Dawley rats by ex vivo dermal exposure of SM in three concentrations and in vivo exposure at four dosages, so as to further illustrate the dose and time dependence profiles with a good relevance to the abundance of four SM-DNA adducts.

blocked, leading to DNA breakage and cell cycle arrest. Eventually, the unrepaired lesions may lead to miscoding, altered gene expression, mutation, cancer, and thus, death. Therefore, identification and quantification of DNA adducts are essential for establishing the relationship between DNA adduct formation and other biological end points (mutations, DNA double-strand break etc.), and the evaluation of genetic damage.10−12 According to the multiple independent experiments ex vivo or in vivo summarized in a review, 13 it has been presumed that these adducts are formed preferentially at the N7 position of guanine (61%), the N3 position of adenine (16%), two N7 positions of guanine as intra- or interstrand cross-links (nearly 17%), and the O6 position of guanine (0.1%). With regard to the determination point of SM-DNA adducts, most research only focused on the detection of the frequently formed SMDNA adduct, i.e., N7-[2-[(2-hydroxyethyl)thio]-ethyl]guanine (N7-HETEG), as an excellent biomarker for internal exposure. However, N7-guanine adducts are chemically unstable and do not participate in Watson−Crick base pairing, indicating that N7-guanine maintained limited biological relevance.14 Even of comparatively much lower abundance, O 6 -[2-[(2hydroxyethyl)thio]-ethyl]guanine (O6-HETEG) is regarded as a critical DNA lesion because the formation of O6-HETEG may affect the hydrogen bonds between guanine and cytosine, and the human DNA repair mechanism fails to remove such SM adducts from this position.15 The interstrand cross-link, i.e., bis[2-(guanin-7-yl)ethyl]sulfide (Bis-G), is also believed to stall replication and finally induce a double-strand break. It is noteworthy that no simultaneous determination of four SMDNA adducts has been reported until now. Its development would definitely be helpful for clarifying the presumption on the presented status of four SM-DNA adducts and for a much further understanding of the SM toxicological mechanism. Considering the significant biological relevance and low abundance of N3-[2-[(2-hydroxyethyl)thio]-ethyl]adenine (N3HETEA), O6-HETEG, and Bis-G, a highly sensitive, specific,

■

MATERIALS AND METHODS

Caution: SM is a highly reactive alkylating vesicant and cytotoxic agent. This agent should be handled only in well-ventilated f ume hoods. The use of gloves and stringent protective measures should be adopted. Chemicals. SM was provided by China Institute of Chemical Defense, with purity higher than 95%. Methanol (HPLC grade) was obtained from J. T. Baker (New Jersey, USA). Guanosine and adenine were the Amresco packaging products from Beijing Xinjingke Biotechnology Limited Company (Beijing, China). Thiodiglycol 491



Article

Figure 2. Mass spectra of SM-DNA adducts and their corresponding deuterated analogues. purity greater than 98%. Guanosine-5′-monophosphate disodium (GMP) was from J&K Scientific LTD (Beijing, China), with a purity of 99%. 1,2-Dichloroethane-d4 was from Cambridge Isotope

(GC grade) was from Fluka (Buchs SG, Switzerland), at purity higher than 98.5%. 6-Chloroguanine was purchased from Beijing Coupling Science and Technology Limited Company (Beijing, China), with a 492



Article

O6-[2-[(2-Hydroxyethyl)thio]ethyl] guanine (O6-HETEG) and O6HETEG-d4. TDG or TDG-d4 of 1.64 mmol was dissolved in 10 mL of dry tetrahydrofuran. The resulting solution was added dropwise to a 60% oil suspension of sodium hydride (130 mg, 3.25 mmol) in 10 mL of dry tetrahydrofuran. After 10 min of stirring under N2 at room temperature, 6-chloroguanine (90 mg, 0.53 mmol) was added, and the resultant mixture refluxed for 3 h. The solvent was evaporated, and the residue was dissolved in 30 mL of water and extracted with ethyl acetate (30 mL, three times). The layer of water was acidified to pH 5.0 with 20% acetic acid and evaporated to dryness, then the mixture was placed in a silica gel column and eluted with dichloromethanemethanol-concentrated ammonia (10:1:0.05, v/v/v). The solvent was evaporated to dryness, and 40 mg of target compound was obtained. The yield was 29%, and the purity was more than 98%. O6-HETEG: lH NMR (DMSO-d6, 300 MHz) δ 12.44 (br, s, 1H), 7.83 (s, 1H), 6.25 (s, 2H), 4.86 (s, 1H), 4.53 (t, J = 7.2 Hz, 2H), 3.58 (d, J = 4.8 Hz, 2H), 2.94 (t, J = 7.2 Hz, 2H), 2.70 (t, J = 6.9 Hz, 2H). UPLC-MS/MS: m/z 256 [M+H]+, 210[M-HOCH2CH2]+, 105 [HOCH2CH2SCH2CH2]+, 87 [HOCH2CH2SCH2CH2−H2O]+. O6-HETEG-d4: lH NMR (CD3OD, 300 MHz) δ 8.09 (s, 2H), 4.74 (t, J = 6.9 Hz, 2H), 3.77 (t, J = 6.9 Hz, 2H), 3.04 (t, J = 6.9 Hz, 2H), 2.81(t, J = 6.9 Hz, 2H). UPLC-MS/MS: m/z 260 [M+H]+, 109 [HOCH2CH2SCD2CD2]+, 91 [HOCH2CH2SCD2CD2-H2O]+. Bis[2-(guanin-7-yl)ethyl] Sulfide (Bis-G) and Bis-G-d4. The procedures reported by Fidder and co-workers were followed.9 Yield was ca. 1%, and the purity was more than 97%. Bis-G: lH NMR (DMSO-d6, 300 MHz) δ 8.41 (s, 2H), 6.81 (s, 4H), 4.46−4.44(m, 4H), 3.04−3.02(m, 4H). UPLC-MS/MS: m/z 389 [M +H]+, 210 (for fragment ions, see Figure 2D). Bis-G-d4: lH NMR (DMSO-d6, 300 MHz) δ 8.26 (s, 2H), 6.56 (s, 4H), 4.42(t, J = 6.9 Hz, 2H), 3.02(t, J = 6.9 Hz, 2H). UPLC-MS/MS: m/z 393 [M+H]+, 214, 210 (for fragment ions, see Figure 2D). Mass spectra of synthetic SM-DNA adducts and their deuterated analogues are shown in Figure 2, and all 1H NMR spectra of these four SM-DNA adducts and their deuterated analogues are shown in Figures S1−S8 (Supporting Information). Animals and Treatment. Adult male Sprague−Dawley rats of specific pathogen free (SPF) grade were purchased from the Laboratory Animal Center of Beijing. The animal experiment was conducted in the Beijing Center for Toxicological Evaluation and Research, in accordance with the protocol approved by the Institutional Animal Care and Use Committee of the Center, which is in compliance with the guidelines of the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC). The animals were allowed to acclimate for at least 1 week prior to experimental use. They ate food and drank water freely. After acclimatization, Sprague−Dawley rats of 207 ± 8 g weight were assigned randomly into 21 groups of five each. The Sprague−Dawley rats were treated by dermal exposure, including a control and four dosage groups. After closely clipping the hair (the hair was clipped 24 h before the application), a fresh dilution of SM in 1,2-propanediol was uniformly applied by injecting 42 μL with a microsyringe onto the back of the rats onto a square area of approximately 1 cm2. SM was administrated as a single dose of 5.6, 11.2, 22.5, and 45.0 mg/kg body weight, equivalent to 0.25, 0.5, 1.0, and 2.0 times the median lethal dose (LD50), respectively (experimental determination of LD50, see Table S1, Supporting Information). Relatively higher doses were used so as to meet the detection needs of the lowest biomarker, i.e., O6HETEG. Control rats were only administered with the equivalent volume of 1,2-propanediol. Considering the mortalities for animals, the rats treated with a dosage of 45.0 mg/kg were euthanized at 1, 2, 4, and 5 days after dosing. The other rats were euthanized with urethane by intraperitoneal injection at 1, 2, 4, 7, and 14 days postdosing, and the derma of exposure spots, blood, spleen, liver, lung, kidney, and other tissues were immediately isolated and stored at −70 °C prior to DNA extraction. Only derma samples were examined in this research. Results of DNA adducts in other organ tissues samples will be reported later.

Laboratories, Inc. (MA, USA) with a deuterated ratio of 99%. 2Mercaptoethanol (biotechnological grade) was obtained from SigmaAldrich (MO, USA), and the purity was better than 98%. All other reagents were of analytical grade and were purchased from Beijing Chemical Works and Sinopharm Chemical Reagents Co. Ltd. (Beijing, China). Ultrapure water was produced in a Mill-Q water purification system (Millipore, MA, USA). Synthesis of Standards and Stable Isotope-Labeled Internal Standards. Molecular structures of standards and stable isotopelabeled ISs are shown in Figure 1. Chemicals of 2-(2-Chloroethylthio)ethanol (Semi-sulfur Mustard, Semi-SM) and Semi-SM-d4. Semi-SM or semi-SM-d4 was synthesized as the raw material for N7-HETEG and N3-HETEA. It was synthesized from equimolar amounts of sodium methoxide and 2-mercaptoethanol in the presence of 1,2-dichloroethane or 1,2-dichloroethane-d4 according to the procedures previously described (Rao et al.).26 Both of them were characterized by gas chromatography (GC)-MS. Thiodiglycol-d4 (TDG-d4). A mixture of 0.5 g of semi-SM-d4 and 8 mL of hydrochloric acid solution of 1 M was heated to 90 °C for 2 h, and the solvent was evaporated to dryness. The resultant compound was characterized by GC-MS, and the deuterated ratio was over 99%. Sulfur Mustard-d4 (SM-d4). Thionyl chloride of 2 mL was added to 0.2 g (1.4 mmol) of semi-SM-d4, refluxed at 80 °C for 2 h, and then evaporated to dryness. The product was characterized by GC-MS, and the deuterated ratio was over 99%. N7-[2-[(2-Hydroxyethyl)thio]ethyl]-guanine (N7-HETEG) and N7HETEG-d4. Guanosine of 1 g (3.5 mmol) was suspended in 12.5 mL of acetic acid, and 1.8 mmol semi-SM or semi-SM-d4 was added dropwise to the suspension. The resulting reaction mixture was stirred at 100 °C for 3 h, then cooled to room temperature. The unreacted guanosine was filtrated and discarded, and the filtrate was evaporated to dryness. The residue was dissolved in 19 mL of HCl solution of 1 M. After extraction of unreacted semi-SM with dichloromethane (19 mL, three times), the solution was heated at 100 °C for 1.5 h, and a pale-yellow clear solution resulted, which was cooled and neutralized with concentrated ammonia. A white crude product was achieved. After the mixture was filtered, the resultant precipitate was redissolved in acidic water, then purified by the preparative chromatography equipped with an ultraviolet (UV) detector (mode, Gilson GX-281, Gilson Inc., WI, USA; column, Phenomenex C18, 150 × 21.1 mm, 5 μm; mobile phase, 0.1% aqueous trifluoro acetic acid and acetonitrile; flow rate, 1 mL/min). The yield was 22%, and the purity was more than 96%. All purities of synthesized adducts and their deuterated analogues were provided by HPLC-UV detection with an area normalization method. N7-HETEG: lH NMR (DMSO-d6, 300 MHz) δ 8.57 (s, lH), 6.96 (br, s, 2H), 4.44−4.42 (m, 2H), 3.54 (t, J = 6.3 Hz, 2H), 3.02−2.98 (m, 2H), 2.61 (t, J = 6.3 Hz, 2H). UPLC-MS/MS: m/z 256 [M+H]+, 239[M+H-NH3]+, 105 [HOCH2CH2SCH2CH2]+, 87 [HOCH2CH2SCH2CH2−H2O]+. N7-HETEG-d4: lH NMR (CD3OD, 300 MHz) δ 8.71 (s, 1H), 3.70 (d, J = 1.2 Hz, 2H), 2.69 (d, J = 1.2 Hz, 2H). UPLC-MS/MS: m/z 260 [M+H]+, 109 [HOCH2CH2SCD2CD2]+, 91 [HOCH2CH2SCD2CD2H2O]+. N3-[2-[(2-Hydroxyethyl)thio]ethyl]adenine (N3-HETEA) and N3HETEA-d4. Adenine of 345 mg (2.5 mmol) was dissolved in 20 mL of dimethylacetamide, and 2.99 mmol semi-SM or semi-SM-d4 was added dropwise to the solution. The reaction mixture was stirred at 110 °C for 5 h and then cooled to room temperature. The crude product was purified by preparative chromatography equipped with a UV detector (the parameters are shown in the N7-HETEG purification section). The yield was 26%, and the purity was more than 98%. N3-HETEA: lH NMR (CD3OD, 300 MHz) δ 8.72 (s, 1H), 8.52 (s, 1H), 4.75−4.73(m, 2H), 3.74(t, J = 4.5 Hz, 2H), 3.30−3.20(m, 2H), 2.75(t, J = 4.5 Hz, 2H). UPLC-MS/MS: m/z 240 [M+H]+, 105 [HOCH2CH2SCH2CH2]+, 87 [HOCH2CH2SCH2CH2−H2O]+. N3-HETEA-d4: lH NMR (CD3OD, 300 MHz) δ 8.72 (s, 1H), 8.53 (s, 1H), 3.74(t, J = 6.3 Hz, 2H), 2.75(t, J = 6.3 Hz, 2H). UPLC-MS/ MS: m/z 244 [M+H] + , 109 [HOCH 2 CH 2 SCD 2 CD 2 ] + , 91 [HOCH2CH2SCD2CD2-H2O]+. 493



Article

Table 1. ID-UPLC-MS/MS Parameters for the Analysis of Four SM-DNA Adducts compound N7-HETEG

declustering potential (V)

collision energy (V)

collision cell exit potential (V)

entrance potential (V)

qualitative ion pairs

quantitative ion pairs

80

20

16

10

256→105 256→87 260→109 260→91 256→105 256→87 260→109 260→91 389→210 389→238 393→214 393→210 393→242 240→105 240→87 244→109 244→91

256→105

N7-HETEG-d4 O6-HETEG

80

20

16

10

O6-HETEG-d4 Bis-G

71

31

16

10

Bis-G-d4

N3-HETEA

66

21

10

10

N3-HETEA-d4

260→109 256→105 260→109 389→210 393→214

240→105 244→109

voltage was 2350 V, and the source temperature was 650 °C. All other mass spectrometric parameters were listed in Table 1. For qualitative analysis, two or three ion pairs were chosen, and for quantitative analysis, only the most abundant ion pair was chosen to achieve the highest sensitivity. In order to demonstrate the suitability of the developed method, systematic validation was carried out, including linearity, intra- and interassay accuracy and precision, limit of detection (LOD, S/N is greater than 10) and the lower limit of quantification (LLOQ, S/N is greater than 20), and the sensitivity, recovery, and selectivity were provided.

SM Administrated Sprague−Dawley Derma Sample Preparation ex Vivo. After closely clipping the hair, derma samples were taken from the untreated adult male Sprague−Dawley rats after euthanization. The frozen Sprague−Dawley rat derma samples were homogenized in a proportion of 0.1 g tissue per milliliter of cold physiological saline and stored at −70 °C prior to exposure ex vivo. Aliquots of 90 μL SM solutions (dissolved in 1,2-propanediol) at the concentrations of 0.5 mM, 0.1 M, and 0.99 M were added to 1.8 mL blank Sprague−Dawley rat derma (0.18g derma, DNA amounts is about 13 μg) homogenized solutions (n = 3) and incubated for 2 h at 37 °C with continuous mixing. The final low, medium, and high concentration levels are 0.024, 4.8, and 48 mM, respectively. Extraction and Neutral Thermal Hydrolysis of Sprague−Dawley Rat Derma DNA. The frozen Sprague−Dawley rat derma samples were homogenized in a proportion of 0.1 g of derma tissue (DNA amounts are about 7.3 μg) per milliliter of cold physiological saline, and they were subjected to the same procedures as those for the blood sample described previously.18 Amounts and purity of DNA were determined by UV−vis spectrometry according to the equation 20A260 nm = 1 mg/mL DNA. The A260 nm/A280 nm ratios were found to be between 1.7 and 1.9, ensuring minimal protein contamination. The yields of DNA are 73 μg per gram of Sprague−Dawley rat derma. The extracted DNA was dissolved in water, then the stable isotope ISs were spiked, and the final volumes were 50 μL in vivo and 100 μL ex vivo. The mixture was continuously shaken at 70 °C for 1 h. After hydrolysis, the sample was cooled down to room temperature for IDUPLC-MS/MS analysis. ID-UPLC-MS/MS Analysis. The qualitative and quantitative analysis of four adducts were carried out with stable isotope dilution UPLCMS/MS in multiple reaction monitoring (MRM) mode. The stock solutions of N7-HETEG, N3-HETEA, and Bis-G were prepared in formic acid solution, while the stock solution of O6-HETEG was prepared in methanol because it is unstable in acidic conditions. All UPLC-MS/MS analyses were performed in a triple quadrupole ion trap (QqQtrap) 5500 UPLC-MS/MS instrument (AB SCIEX, Framingham, MA, USA) by using a Waters ACQUITY UPLC BEH C18 column (2.1 mm ID × 50 mm, 1.7 μm, Waters Co., MA, USA). Mobile phases consisted of ultrapure water (A) and HPLC-grade methanol (B). The gradient elution started with 15% B, held for 1 min, then linearly increased to 60% B over 1 min, held 0.5 min, then decreased to 15% B in 0.1 min, and held for 1.4 min; thus, the total run time was 4 min. The flow rate was 0.35 mL/min, and the injection volume was 3 μL. MS detection was initiated from a positive turbo ion spray source. Curtain gas was 20 psi (1 psi = 6895 Pa), collision gas was 8 psi, ion source gas 1 and gas 2 were 40 and 50 psi, respectively, ionspray

■

RESULTS AND DISCUSSION Synthesis of DNA Adducts and the Deuterated Analogues. To establish a rapid and highly selective method for the simultaneous detection, identification, and characterization of four SM-DNA adducts, a stable ID method may be the best choice. Since the reference standards of adducts and the stable isotope ISs were not commercially available, four SMpurine alkylated products and their deuterated analogues were synthesized and characterized as an important issue to develop this method. The synthesis approaches are shown in Scheme S1 (Supporting Information). All obtained compounds were identified with MS and 1H NMR, and the results of four SMDNA adducts were in high accordance with the references.9 Purity and the deuterated ratio of all compounds meet the analysis demand well.27 In the synthesis of N7-HETEG and N3-HETEA, considering the toxicity of two alkylating agents, SM and semi-SM, semi-SM with much less toxicity was chosen as the original reactant. Another reactant used in the preparation of N7-HETEG is guanosine. Since both N7 and N9 of guanine were the most reactive positions in the presence of acetic acid, guanosine with a deoxyribose occupying its N9 position can make the alkylating agent selectively bind to the N7 position and can efficiently reduce byproduct. 9 Even though guanosine and semi-SM can react in equal moles theoretically, considering the poor solubility of guanosine in acetic acid, excessive guanosine was mixed with semi-SM to ensure complete reaction. Pure N7HETEG was easily isolated from the solution since it is not soluble in neutral pH. Both materials and the reaction condition in this approach were different from those described 494



Article

Scheme 1. Procedures of Simultaneous Analysis for N7HETEG, O6-HETEG, N3-HETEA, and Bis-G in Rat Derma Samples by ID-UPLC-MS/MS

but both phenomena did not affect the compounds acting as ISs. ID-UPLC-MS/MS Method Development and Validation. The whole procedure for the determination of N7HETEG, O6-HETEG, N3-HETEA, and Bis-G in derma samples by ID-UPLC-MS/MS is shown in Scheme 1. Here, we selected SM binding DNA nucleobases instead of nucleosides as SM-DNA adducts for three reasons. First, the altered molecular shape and hydrogen bonding characteristics occurred in the level of structurally modified bases.25 Second, MS-based analysis on nucleobases can provide better sensitivity than nucleosides since nucleosides have larger hydrophilicity due to the occurrence of glycosides and thus have a lower ionization efficiency than nucleobases. In the high temperature of the ion source in MS, some thermal labile nucleosides are inclined to decompose as the nucleobases. Third, prior to UPLC-MS/MS analysis, DNA is typically hydrolyzed to release SM-bound nucleotides, nucleosides, or nucleobases. Here, we used a mild neutral thermal hydrolysis condition, which causes hydrolytic release of the alkylated purines at neutral pH,8 and bis-G adducts can be quantitatively released,20,24 as well as thermally labile N7-HETEG and N3-HETEA. Besides, even O6HETE-deoxyguansine (dG) has better stability than N7-HETEdG and N3-HETE-deoxyadesine (dA), the O6-HETEG adduct is more thermal and acidic liable than N7-HETEG and N3HETEA during typical acidic and/or heating conditions after the enzymatic hydrolysis procedure. We need to pursue a balance between conditions to fully release O6-HETEG from SM-DNA adduction and to maintain the stability of such a compound. In our work, a relatively mild thermal hydrolysis condition, i.e., 70 °C for 1 h was employed because a certain amount of O6-HETEG can only be preserved in such a condition in comparison with other acidic hydrolysis (70 °C, 1 h at pH 2) and stronger thermal hydrolysis (100 °C, 1 h) conditions (Figure S9, Supporting Information). In this research, biological samples spiked with the deuterium-labeled ISs were subjected to UPLC-MS/MS. This method achieved high sensitivity, fast data-collection speed, and high efficiency. Several mobile phases were tested, such as water, 0.1% formic acid (pH 2.7), ammonium formate/formic acid buffer (pH 4.2), ammonium formate/formic acid buffer (pH 5.2), ammonium formate solution (10 mM, pH 6.1), ammonium acetate solution (10 mM, pH 6.8), etc. When the mobile phase was acidic, N3-HETEA had a good peak shape, while the other three adducts showed poor peak shape and sensitivity. The best sensitivity for all four adducts was only achieved with a gradient of water (solvent A) and methanol (solvent B) without any acidic adjustment. Under such a condition, the peak symmetry of N3-HETEA was slightly poorer, but its detection sensitivity was the same as the sensitivity in the acidic mobile phase. Since the four SM-DNA adducts have high polarity and similar properties, several different stationary phases and mobile phases were tested. N3-HETEA was never separated with the other three adducts, especially with N7-HETEG (Figure 3). However, it did not affect their qualitative and quantitative results in MRM mode. MRM mode offers a great advantage here since only selected multiple pairs of precursor ion of correct molecular weight and its product ion are detected,25 and it promises superior specificity from the point of view of m/z values. To validate this new method, blank DNAs were isolated from Sprague−Dawley rat derma, digested, and spiked with certain

in previous literature;9 this preparation way is much simpler and easier. In the preparation of N3-HETEA, adenine was reacted directly with semi-SM in the presence of N,N-dimethylacetamide since the N3 position of adenine was most reactive in neutral conditions, which was found in 1979 by Fujii and his co-workers.28 The optimized synthesis procedure was simpler and safer than the earlier reported methods.9 Since the O6 position of guanine or guanosine is less reactive, the synthesis of O6-HETEG was quite difficult. The only available reference was reported by Fidder et al., which had tedious and time-consuming procedures.9 However, Gundersen et al. successfully synthesized 2-amino-6-(2-methoxyethoxy) purine from 2-amino-6-chloropurine in strong basic conditions.29 Enlightened by the latter research, a facile method from nontoxic raw material adenine was designed and successfully achieved with only a one-step reaction. In the synthesis of Bis-G, the main byproduct is N7-HETEG. In order to decrease the output of this monoadduct, the reactants were added according to the theoretic ratio, namely, the ratio of GMP to SM was 2 to 1. The acidity of the Bis-G dissolving solution utilized was slightly higher than that of N7HETEG; benefiting from this minor difference, pure Bis-G was produced successfully in this approach. Compared with stable isotope-labeled agents originated from nitrogen, sulfur, oxygen, and carbon atoms, deuterium-labeled agents were more cost-effective. Except for the fact that the synthesis began from 1,2-dichloroethane-d4 instead of 1,2dichloroethane, the procedures were exactly same as those for the four SM-DNA adducts. Because the reactant involves four deuterium atoms, i.e., only occupying half of all eight hydrogen atoms, O6-HETEG-d4 has two isomeric compounds (see Figure 1, compound H1 and H2) and Bis-G-d4 has two main fragment ions (m/z 210 and 214) in the mass spectrum (see Figure 2D), 495



Article

Figure 3. Chromatograms of four SM-DNA adducts, N7-HETEG and O6‑HETEG (panel A), N3-HETEA (panel B), and Bis-G (panel C) in IDUPLC-MS/MS using MRM mode. The chromatograms of the control, the sample in vivo, and the sample spiked with IS are shown from top to bottom. Transition m/z: N7-HETEG, 256→105; O6-HETEG, 256→105; N3-HETEA, 240→105; Bis-G, 389→210; N7-HETEG-d4, 260→109; O6HETEG-d4, 260→109; N3-HETEA-d4, 244→109; Bis-G-d4, 393→214.

amounts of N7-HETEG, N3-HETEA, O6-HETEG, Bis-G, and corresponding deuterium-labeled ISs, and UPLC-MS/MS analysis was followed. An excellent correlation was observed between the expected and observed concentrations of adducts, and the linearity was good in 5 to 7 orders of magnitude with the value of correlation coefficient (r2) over 0.99, i.e., N7HETEG (0.003−198 ng/mL), N3-HETEA (0.026−260 ng/ mL), O6-HETEG (0.0007−149 ng/mL), and Bis-G (0.016− 1550 ng/mL). The LOD values (S/N ratio greater than 10) were 1, 0.2, 5, and 9 pg/mL (0.01, 0.002, 0.04, and 0.11 fmol on column) for N7-HETEG, O6-HETEG, Bis-G, and N3-HETEA, respectively, and the LLOQ values (S/N ratio greater than 20) were 3, 0.7, 16, and 26 pg/mL (0.04, 0.01, 0.12, and 0.33 fmol on column) for N7-HETEG, O6-HETEG, Bis-G, and N3HETEA, respectively. The sensitivity was higher than that of the previous method developed in our group18 and at the same level with the quantification method of similar DNA

Table 2. Intra- and Interday Precisions and Recoveries of Four SM-DNA Adducts precision (RSD) adduct

concentration (ng/mL)

intraday (n = 7)

interday (n = 6)

recovery n=3

0.009 9.90 180 0.002 2.93 133 0.16 17.1 1318 0.09 9.53 238

8.8% 2.1% 1.4% 5.6% 3.6% 2.2% 4.9% 2.8% 2.2% 6.2% 0.9% 0.4%

9.5% 2.0% 0.9% 4.4% 3.0% 2.4% 5.2% 2.7% 4.2% 5.9% 1.7% 1.0%

118 ± 0.6% 107 ± 0.2% 100 ± 2% 103 ± 2% 76 ± 3% 86 ± 2% 81 ± 12% 129 ± 2% 123 ± 4% 109 ± 3% 112 ± 1% 93 ± 2%

N7-HETEG

O6-HETEG

Bis-G

N3-HETEA

Table 3. Determination Results of Four SM-DNA Adducts ex Vivo (n = 3) N7-HETEG

N3-HETEA

O6-HETEG

Bis-G

exposure concentration (mM)

ng/g

%

ng/g

%

ng/g

%

ng/g

%

0.024 4.8 48

6.9 ± 0.8 170 ± 40 6570 ± 820

51.6 63.8 83.2

0.3 ± 0.2 8.8 ± 2.8 158 ± 16

2.5 3.3 2.0

0.08 ± 0.01 0.41 ± 0.05 9.8 ± 6.9

0.56 0.15 0.12

6.1 ± 1.4 89 ± 42 1160 ± 123

45.3 32.7 14.7

496



Article

to be 76% to 129% (n = 3), and both the interday (n = 6) and intraday (n = 7) precisions were less than 10% (Table 2). Analysis of SM-DNA Adducts in Rat Derma ex Vivo. In derma samples exposed to low, medium, and high SM concentrations ex vivo, all four SM-DNA adducts were positively detected in exposed derma samples (Table 3). The results showed that the N7 position of guanine was the most reactive nucleophilic site and can react with SM readily. The next reactive position was the N3 position of adenine, and the O6 of guanine was the least reactive. Only 0.01% to 0.1% (w/w) of administrated SM reacted with DNA by an alkylating reaction, and the amounts of total SM-DNA adducts were about 0.002%, 0.360%, and 10.8% (w/w) of total DNA (calculated according to the extracted amount of DNA). The content of adduct was also transformed to the molar concentration to investigate the molar contribution of each SMDNA adduct. The simultaneous quantification results showed that the molar percentages of N7-HETEG, Bis-G, N3-HETEA, and O6-HETEG were 61−88%, 10−35%, 2−4%, and 0.1−0.7%, respectively. The molar percentages of Bis-G and N3-HETEA were significantly different from previous published results,13 i.e., Bis-G has much higher abundance, 10−35% versus 17%,13 and N3-HETEA has much lower abundance, 2−4% versus 16%. Meanwhile, our results were well supported by an in vitro

Figure 4. Amounts of four SM-DNA adducts in different derma exposed concentrations ex vivo (n = 3). The value beyond the column represents the mean value of each amount for SM-DNA adduct. The error bar represents the standard deviation of results for three parallel derma samples ex vivo.

adducts. 21−24 Three levels of low, medium, and high concentration were tested, method accuracy was determined

Figure 5. Dose and time dependence profiles of four SM-DNA adducts in the derma of Sprague−Dawley rats (n = 5). Levels of the four adducts’ response positively with the exposed dosages and negatively with the elapsed time after dosing. Inserts show the time dependence profiles of four SM-DNA adducts at the dosage of 45 mg/kg. Regarding the ultrahigh dose of 45.0 mg/kg group, the last data were collected on the 5th day postexposure due to high mortality. The error bar represents the detection standard deviation of five derma samples of rats in the same group. 497



Article

Figure 6. Linearity between four SM-DNA adducts and derma exposed dosages in vivo (n = 5). The error bar represents the detection standard deviation of five derma samples of rats in the same group.

obviously observed. All adducts decreased with time after exposure and positively correlated with the exposure dosages. The mean maximum contents of all four SM-DNA adducts were achieved on day 1 postexposure, i.e., 112, 346, 628, and 1910 ng/g for dose 5.6, 11.2, 22.5, and 45.0 mg·kg−1, respectively, and the contents of adducts in derma decreased rapidly during postdosing days 1 to 4. The results indicate a dose dependent relationship between content of adducts and exposure dose. Regression analysis of mean maximum concentration of adducts versus dose showed an r2 beyond 0.964 (Figure 6). The total amount of four adducts was ca. from 0.001% to 0.01% (w/w) of the applied dose of SM. The abundance of four adducts in the different exposure dosage levels is shown in Figure 7. The molar percentages of N7-HETEG, Bis-G, N3-HETEA, and O6-HETEG were ca. 64 to 81%, 18 to 42%, 1.3 to 4.6%, and 0.04 to 0.62%, respectively, which correlates with the results ex vivo very well. SM penetrated the rat derma from epidermis in vivo, while it reacted directly with the derma homogenate in the ex vivo experiment, and the molar percentage results in vivo and ex vivo are quite similar. Therefore, it suggests that the formation rate and ratio of four SM-DNA adducts were not significantly influenced by skin structure, such as the epidermis and dermis. Here, for the first time, the real status of four DNA adducts presented in vivo was revealed, and a much higher abundance of Bis-G and a lower abundance of N3-HETEA were found compared with the previous results.13 It is important to note that the higher abundance of Bis-G in all SM-DNA adducts indicates a more serious DNA lesion by SM alkylation because this interstrand cross-linked adduct is believed to stall DNA replication and finally induce a double-strand break.13

Figure 7. Amounts of four SM-DNA adducts in different derma exposed dosages in vivo (n = 5) at the first day postdosing. The value beyond the column represents the mean value of each amount for the SM-DNA adduct. The error bar represents the standard deviation of results from five derma samples of rats in the same group.

investigation report published in 2013, in which N7-HETEG, Bis-G, and N3-HETEA in cellular DNA were isolated and determined by HPLC-MS/MS. Bis-G has the second abundance, existing as less than half of the content of N7HETEG.30 All trends in the molar percentage values of four SM-DNA adducts were the same at three different exposure concentrations (Figure 4). It is indicated that SM primarily reacts with the most reactive site, N7 of guanine, and then with other but less reactive positions, N3 of adenine, and O6 of guanine ex vivo. Analysis of SM-DNA Adducts in Rat Derma in Vivo. In derma samples exposed to low, medium, high, and ultrahigh SM dosage levels in vivo, the dose and time dependence profiles of four SM-DNA adducts are shown in Figure 5. Regarding the dose of the 45.0 mg/kg group, the last data were collected on the fifth day postexposure due to high mortality. The DNA adduct was not detected in rat derma samples collected before dosing or from the control group. For all dosage levels, both time and dose dependent trends of four SM-DNA adducts were

■

CONCLUSIONS

To develop a highly sensitive and selective method, four SMDNA adducts and their stable isotope-labeled ISs were synthesized. The stable ID-UPLC-MS/MS method for simultaneous determination of four SM-DNA adducts was developed, and high sensitivity, precision, and accuracy were achieved. This method was successfully applied to determine four SM-DNA adducts in the SM-exposed Sprague−Dawley rat derma ex vivo and in vivo, and a dose or time dependent 498



Article

relationship was found. It is the first time that the real presented status of four DNA adducts was revealed, in which we found a much higher abundance for the cross-linked adduct Bis-G and much less for N3-HETEA than previous published results. We hope that our experimental proof provides a quantitative view for further understanding DNA lesions which are formed on exposure to SM.

■

high performance liquid chromatography-tandem mass spectrometry. J. Chromatogr., B 917−918, 100−107. (3) Payne, O. J., Graham, S. J., Dalton, C. H., Spencer, P. M., Mansson, R., Jenner, J., Azeke, J., and Braue, E. (2013) The effects of sulfur mustard exposure and freezing on transdermal penetration of tritiated water through ex vivo pig skin. Toxico. in Vitro 27, 79−83. (4) Marrs, T. C., Maynard, R. L., and Sidell MD, F. R. (2007) Chemical Warfare Agents Toxicology and Treatment, 2nd ed., John Wiley & Sons Ltd, New York. (5) Smith, J. R., Capacio, B. R., and Korte, W. D. (2008) Analysis for plasma protein biomarkers following an accidental human exposure to sulfur mustard. J. Anal. Toxicol. 32, 17−24. (6) Noort, D., Fidder, A., Degenhardt-Langelaan, C. E. A. M., and Hulst, A. G. (2008) Retrospective detection of sulfur mustard exposure by mass spectrometric analysis of adducts to albumin and hemoglobin: an in vivo study. J. Anal. Toxicol. 32, 25−30. (7) Benschop, H. P., van der Schans, G. P., and Noort, D. (1997) Verification of exposure to sulfur mustard in two casualities of the Iran-Iraq Conflict. J. Anal. Toxicol. 21, 249−251. (8) Lawley, P. D., and Brookes, P. (1967) Interstrand cross-linking of DNA by difunctional alkylating agents. J. Mol. Biol. 25, 143−160. (9) Fidder, A., Moes, G. W., Schefer, A. G., van der Schans, G. P., Baan, R. A., de Jong, L. P., and Benschop, H. P. (1994) Synthesis, characterization, and quantitation of the major adducts formed between sulfur mustard and DNA of calf thymus and human blood. Chem. Res. Toxicol. 7, 199−204. (10) Tarantini, A., Maitreb, A., Lefebvre, E., Marques, M., Marie, C., Ravanat, J. L., and Douki, T. (2009) Relative contribution of DNA strand breaks and DNA adducts to the genotoxicity of benzo[a]pyrene as a pure compound and in complex mixtures. Mutat. Res. 671, 67−75. (11) Martinez, G. R., Loureiro, A. P., Marques, S. A., Miyamoto, S., Yamaguchi, L. F., Onuki, J., Almeida, E. A., Garcia, C. C., Barbosa, L. F., Medeiros, M. H., and Di Mascio, P. (2003) Oxidative and alkylating damage in DNA. Mutat. Res. 544, 115−127. (12) Garner, R. C. (1998) The role of DNA adducts in chemical carcinogenesis. Mutat. Res. 402, 67−75. (13) Kehe, K., Balszuweit, F., Steinritz, D., and Thiermann, H. (2009) Molecular toxicology of sulfur mustard-induced cutaneous inflammation and blistering. Toxicology 263, 12−19. (14) Boysen, G., Pachkowski, B. F., Nakamura, J., and Swenberg, J. A. (2009) The formation and biological significance of N7-guanine adducts. Mutat. Res. 678, 76−94. (15) Ludlum, D. B., Kent, S., and Mehta, J. R. (1986) Formation of O6-ethylthioethylguanine in DNA by reaction with the sulfur mustard, chloroethyl sulfide, and its apparent lack of repair by O6-alkylguanineDNA alkyltransferase. Carcinogene 7, 1203−1206. (16) Zhang, S., Balbo, S., Wang, M., and Hecht, S. S. (2011) Analysis of acrolein-derived 1, N2-propanodeoxyguanosine adducts in human leukocyte DNA from smokers and nonsmokers. Chem. Res. Toxicol. 24, 119−124. (17) Marie-Desvergne, C., Maître, A., Bouchard, M., Ravanat, J. L., and Viau, C. (2010) Evaluation of DNA adducts, DNA and RNA oxidative lesions, and 3-hydroxybenzo(a)pyrene as biomarkers of DNA damage in lung following intravenous injection of the parent compound in rats. Chem. Res. Toxicol. 23, 1207−1214. (18) Wei, Y., Yue, L., Liu, Q., Chen, J., and Xie, J. (2011) A sensitive high performance liquid chromatography-positive electrospray tandem mass spectrometry method for N7-[2-[(2-hydroxyethyl)thio]-ethyl]guanine determination. J. Chromatogr., B 879, 1707−1712. (19) Malayappan, B., Johnson, L., Nie, B., Panchal, D., Matter, B., Jacobson, P., and Tretyakova, N. (2010) Quantitative high-performance liquid chromatography-electrospray ionization tandem mass spectrometry analysis of Bis-N7-guanine DNA-DNA cross-links in white blood cells of cancer patients receiving cyclophosphamide therapy. Anal. Chem. 82, 3650−3658. (20) Goggin, M., Loeber, R., Park, S., Walker, V., Wickliffe, J., and Tretyakova, N. (2007) HPLC-ESI+-MS/MS analysis of N7-guanineN7-guanine DNA cross-links in tissues of mice exposed to 1,3butadiene. Chem. Res. Toxicol. 20, 839−847.

ASSOCIATED CONTENT

S Supporting Information *

Synthesis route of four SM-DNA adducts and deuterated analogues; 1H NMR spectra of N7-HETEG, N3-HETEA, O6HETEG, Bis-G, N7-HETEG-d4, N3-HETEA-d4, O6-HETEG-d4, and Bis-G-d4; and the determined median lethal dose of Sprague−Dawley rats with dermal exposure to SM. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*(L.G.) Tel: +86 10 66930621. Fax: +86 10 68225893. E-mail: [email protected]. * (J.X.) Tel/Fax: +86 10 68225893. E-mail: [email protected]. Present Address

§ (L.Y.) College of Life and Environmental Sciences, Minzu University of China, 27 South Zhongguancun Avenue, Haidian District 100081, Beijing, China.

Funding

This work was supported by National Science and Technology Major Project of the Ministry of Science and Technology of China (Grant No. 2012ZX09301003-001-010), and China Postdoctoral Science Foundation (No. 20090450198). Notes

The authors declare no competing financial interest.

■

ABBREVIATIONS SM, sulfur mustard; N7-HETEG, N7-[2-[(2-hydroxyethyl)thio]ethyl]guanine; O6-HETEG, O6-[2-[(2-hydroxyethyl)thio]ethyl]guanine; N3-HETEA, N3-[2-[(2-hydroxyethyl)thio]ethyl]adenine; Bis-G, bis[2-(guanin-7-yl)ethyl]sulfide; IDUPLC-MS/MS, stable isotope dilution−ultrahigh performance liquid chromatography−tandem mass spectrometry; MRM, multiple reaction monitoring; LOD, limit of detection; LLOQ, lower limit of quantification; S/N ratio, the ratio of signal-tonoise; IS, internal standard; GC-MS, gas chromatography− mass spectrometry; GMP, guanosine-5′-monophosphate disodium; TDG, thiodiglycol; NMR, nuclear magnetic resonance; DMSO, dimethyl sulfoxide; UV−vis, ultraviolet visible; HPLC, high performance liquid chromatography; MS, mass spectrometry; SPF, specific pathogen free; AAALAC, Association for Assessment and Accreditation of Laboratory Animal Care International; LD50, median lethal dose; QqQtrap MS, triple quadrupole-linear ion trap mass spectrometry

■

REFERENCES

(1) Organisation for the Prohibition of Chemical Weapons (2005), Convention on the Prohibition of the Development, Production, Stockpiling and Use of Chemical Weapons and on Their Destruction, 3rd ed., the Technical Secretariat of the Organisation for the Prohibition of Chemical Weapons, Hague. (2) Li, C., Chen, J., Liu, Q., Xie, J., and Li, H. (2013) Simultaneous quantification of seven plasma metabolites of sulfur mustard by ultra 499



Article

(21) Chen, H-J. C., Wang, Y.-C., and Lin, W.-P. (2012) Analysis of ethylated thymidine adducts in human leukocyte DNA by stable isotope dilution nanoflow liquid chromatography-nanospray ionization tandem mass spectrometry. Anal. Chem. 84, 2521−2527. (22) Chen, H-J. C., Wang, Y.-C., and Lin, W.-P. (2011) Quantitative analysis of multiple exocyclic DNA adducts in human salivary DNA by stable isotope dilution nanoflow liquid chromatography-nanospray ionization tandem mass spectrometry. Anal. Chem. 83, 8543−8551. (23) Chen, H-J. C., Lin, G.-J., and Lin, W.-P. (2010) Simultaneous quantification of three lipid peroxidation-derived etheno adducts in human DNA by stable isotope dilution nanoflow liquid chromatography nanospray ionization tandem mass spectrometry. Anal. Chem. 82, 4486−4493. (24) Sangaraju, D., Goggin, M., Walker, V., Swenberg, J., and Tretyakova, N. (2012) NanoHPLC-nanoESI+-MS/MS quantitation of Bis-N7-guanine DNA-DNA cross-links in tissues of B6C3F1 mice exposed to subppm levels of 1,3-butadiene. Anal. Chem. 84, 1732− 1739. (25) Tretyakova, N., Goggin, M.., Sangaraju, D., and Janis, G. (2012) Quantitation of DNA adducts by stable isotope dilution mass spectrometry. Chem. Res. Toxicol. 25, 2007−2035. (26) Rao, M. K., Sharma, M., Raza, S. K., and Jaiswal, D. K. (2003) Synthesis, characterization and mass spectrometric analysis of cysteine and valine adducts of sulphur mustard. Phosphorus, Sulfur Silicon Relat. Elem. 178, 559−566. (27) Nie, Z., Liu, Q., and Xie, J. (2011) Improvements in monitoring the N-terminal valine adduct in human globin after exposure to sulfur mustard and synthesis of reference chemicals. Talanta 85, 1154−1159. (28) Fujii, T., Walker, G. C., and Leonard, N. J. (1979) 3-substituted adenines. In vitro enzyme inhibition and antiviral activity. J. Med. Chem. 22, 125−129. (29) Gundersen, L. L., Benneche, T., and Rise, F. (1992) Regiochemistry and stereochemistry in Pd(0)-catalyzed allylic alkylation of nucleoside bases. Acta Chem. Scand. 46, 761−771. (30) Batal, M., Boudry, I., Clery-Barraud, C., Mouret, S., and Douki, T. (2013) Relative yields of monomeric and dimeric adducts induced by sulphur mustard in isolated and cellular DNA as determined by HPLC/tandem mass spectrometry. Toxicol. Environ. Chem. 95, 260− 276.

500


Abundance of Four Sulfur Mustard-DNA Adducts ex Vivo and in Vivo

Recommend Documents