Label-Free and Selective Photoelectrochemical Detection of Chemical

Jun 18, 2013 - Label-Free and Selective Photoelectrochemical Detection of Chemical DNA Methylation Damage Using DNA Repair Enzymes. Yiping Wu ...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/ac

Label-Free and Selective Photoelectrochemical Detection of Chemical DNA Methylation Damage Using DNA Repair Enzymes Yiping Wu, Bintian Zhang, and Liang-Hong Guo* State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-environmental Sciences, Chinese Academy of Sciences, Beijing 100085, China ABSTRACT: Exogenous chemicals may produce DNA methylation that is potentially toxic to living systems. Methylated DNA bases are difficult to detect with biosensors because the methyl group is small and chemically inert. In this report, a label-free photoelectrochemical sensor was developed for the selective detection of chemically methylated bases in DNA films. The sensor employed two DNA repair enzymes, human alkyladenine DNA glycosylase and human apurinic/ apyrimidinic endonuclease, to convert DNA methylation sites in DNA films on indium tin oxide electrodes into strand breaks. A DNA intercalator, Ru(bpy)2(dppz)2+ (bpy=2,2′bipyridine, dppz = dipyrido[3,2-a:2′,3′-c]phenazine) was then used as the photoelectrochemical signal indicator to detect the DNA strand breaks. Its photocurrent signal was found to correlate inversely with the amount of 3-methyladenines (metAde) produced with a methylating agent, methylmethane sulfonate (MMS). The sensor detected the methylated bases produced with as low as 1 mM MMS, at which concentration the amount of metAde on the sensor surface was estimated to be 0.5 pg, or 1 metAde in 1.6 × 105 normal bases. Other DNA base modification products, such as 5-methylcytosine and DNA adducts with ethyl and styrene groups did not attenuate the photocurrent, demonstrating good selectivity of the sensor. This strategy can be utilized to develop sensors for the detection of other modified DNA bases with specific DNA repair enzymes.

T

methyl adducts indicate that metAde blocks DNA replication by impeding polymerases.5,6 Although metGuo does not block replication or miscode, its depurination produces an abasic site which is known to be mutagenic.7 MMS is a suspected carcinogen.8 Living organisms have developed a number of DNA base repairing strategies to protect against genetic abnormalities. The main strategy for correcting DNA base lesion is base excision repair (BER). Alkyladenine DNA glycosylase (AAG) is a highly efficient DNA repair enzyme which rapidly excises the alkylated adenine to generate an apurinic site. The resulting apurinic site would be further converted to a single nucleotide gap by an apurinic/ apyrimidinic endonuclease (APE) through cleaving the 3′and 5′- phosphodiester bond. A series of repairing steps would then proceed with a DNA polymerase and a DNA ligase.9 Although metGuo is poorly repaired, the metGuo deoxyribose bond is unstable. The state of the base modification and loss would achieve steady within a few days.10 Considering the importance of DNA methylation in cell function and genetic disease diagnosis, a variety of chemical and biochemical assays have been developed for the detection of methylated bases. Conventional techniques include antibodybased immunological assay, restriction enzyme based cleavage

he sequence of bases in the interior of DNA double helix constitutes the genetic code that is critically important for all organisms. Exogenous or endogenous sources may destroy the integrity of cellular DNA at any time. Numerous DNA lesions have been identified, including strand breaks, covalent modification of nucleobases with chemicals, depuration of nucleobases, oxidative damage, etc.1 Among them, DNA modification by alkylating chemical agents has attracted great attention recently because it would trigger serious consequences if it persists.2 In mammals, normal DNA methylation involves the covalent attachment of a methyl group to the C5 position of cytosine catalyzed by DNA methyltransferase with S-adenosyl-methionine as the methyl donor. It frequently occurs in the 5′-CG-3′ repeat sequence, known as the CpG islands. In normal cells, most of the CpG islands in the gene promoter regions are unmethylated, and their downstream genes are therefore transcriptionally active.3 Apart from enzymatic DNA methylation described above, DNA chemical methylation occurs directly with some alkylating chemical agents such as methyl methanesulfonate (MMS), without the participation of DNA methyltransferases.4 The reaction sites of this nonenzymatic DNA methylation are the ring nitrogens of purine bases, which are poor acceptor sites for enzymatic DNA methylation. 7-Methylguanine (metGuo) and metAde are the two major adducts of DNA methylation with MMS, each accounting for 82% and 11% of the total DNA damage.3,4 Studies on the biological consequences of the © 2013 American Chemical Society

Received: May 4, 2013 Accepted: June 18, 2013 Published: June 18, 2013 6908

dx.doi.org/10.1021/ac401346x | Anal. Chem. 2013, 85, 6908−6914

Analytical Chemistry

■

assay, and bisulfite based sequencing assay.11,12 With the advancement in analytical instrumentation, high-performance liquid chromatography (HPLC) has become a powerful technique for the quantitative determination of the global levels of DNA methylation.13 More recent studies have focused on simple and rapid identification of DNA methylation products with biosensors and bioassays.14 Among them, electrochemical sensors offer some distinctive advantages in terms of simplicity, low cost, fast response time and miniaturization.15 Niwa et al. reported several studies on direct electrochemical detection of methylated cytosine by exploiting the differential oxidation potentials between cytosine and methylated cytosine on a nanocarbon film electrode.16−18 Using carbon nanotube modified electrodes, Wang et al also observed differential responses between unmethylated and methylated cytosine. Furthermore, the interference on methylated cytosine by direct oxidation of thymine was eliminated.19 Osmium bipyridine derivatives were attached specifically and covalently to methylcytosine in DNA as electrochemical signal reporters.20 Gene-specific methylation analysis was also achieved by taking advantage of the discriminative ability of some restriction enzymes again methylated cytosine bases in specific DNA sequences immobilized on electrode surface.21,22 When enzyme-catalyzed DNA methylation was carried out in situ on the electrode surface, methyltransferase activity could also be evaluated.21 Electrochemistry was also employed as a detection method by coupling with some biochemical reactions. For example, the methylation of human p16 gene CpG islands was detected by linker PCR coupled with electrochemistry, using Co(phen)32+ as a redox indicator.23 A high-affinity DNA binder, ferrocene naphthalene diimide, was synthesized and employed as a redox indicator in the electrochemical detection following bisulfite treatment.24,25 All the studies described above target methylcytosine as the enzymatic product. By comparison, reports on the electrochemical detection of chemical DNA methylation products (methylpurines) are very scarce. Detection of chemical DNA methylation is highly desired, since it provides a theoretic basis for the toxicity testing of chemicals, as well as a research tool for the investigation of toxicity mechanisms. Rusling et al. developed a voltammetric sensor for the rapid screening of chemical toxicity. The sensor incorporated a redox polymer film on electrode as a catalyst for the electrochemical oxidation of guanine bases in DNA. Upon exposure to a methylating agent such as MMS and dimethyl sulfate, guanine oxidation current increased as a result of purine methylation, DNA partial unwinding, and enhanced catalytic reaction rate. It was estimated that approximately 1 damaged base/800 from 2 mM MMS was detected.26,27 We have been developing photoelectrochemical biosensors for different applications.28,29 One of the attractive aspects of photoelectrochemical detection technique is its potentially high sensitivity owing to the total separation of the excitation source (light) and the detection signal (current).30−32 Using this technique, a photoelectrochemical DNA sensor for the screening of DNA-damaging chemical agents was constructed successfully.33 A photoelectrochemical sensor for the quantitative determination of 8-oxodGuo in DNA films was also developed.34 In this paper, we report our study on the development of a label-free photoelectrochemical sensor for the detection of methylated purines generated by reacting DNA films with MMS.

Article

EXPERIMENTAL SECTION Chemicals and Materials. Calf thymus double-stranded DNA (CT-DNA, 13K base pairs), bovine serum albumin (BSA), poly-(diallydimethyl ammonium chloride) (PDDA), methyl methanesulfonate (MMS), ethyl methanesulfonate (EMS), and styrene oxide were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). Oligonucleotides with 25 bases containing one 8-oxodGuo (Oligo-OH) were purchased from Integrated DNA Technologies (Coralville, IA). Oligonucleotides with 25 bases containing one 5-mdC (Oligo-mC), and the oligonucleotides complementary to Oligo-OH and Oligo-mC were supplied by Sangon Corp. (Shanghai, P. R. China). The sequences of these oligonucleotides are illustrated below. Oligonucleotide hybridization was performed in 2 × SSC buffer (0.3 M sodium chloride, 30 mM sodium citrate, pH 7.0) on a Biometra T-Gradient thermocycler (Goettingen, Germany), and then slowly cooled to room temperature. DNA concentration was determined by its absorbance at 260 nm (Agilent Technologies, Santa Clara, CA). Human alkyladenine DNA glycosylase (hAAG), human apurinic/apyrimidinic endonuclease (APE1) and 1 × NEB buffer 4 (50 mM potassium acetate, 20 mM Tris-acetate, 10 mM magnesium acetate, 1 mM DTT, pH 7.9) were from New England Biolabs (Ipswich, MA). MetGuo and metAde were obtained from Fluka (St. Louis, MO, U.S.A.). Fifteen percent tin(IV) oxide, as a colloidal dispersion of 15 nm particles in water, was obtained from Alfa Aesar (Ward Hill, MA). Ru(bpy)2(dppz)2+ was synthesized according to a published procedure.35 All other chemicals and solvents were analytical grade reagents. Solutions were prepared in high-purity water from a Millipore Milli-Q (Biocel, Merck, U.S.A.) water purification system. Indium−tin oxide (ITO) conductive glass was supplied by Weiguang Corp (Shenzhen, P. R. China).

DNA Film Assembly. SnO2 nanoparticle modified ITO electrodes were prepared as previously described.29 CT-DNA or the hybridized oligonucleotide duplex (Duplex-OH, DuplexmC) was assembled on SnO2 nanoparticle electrode surface by layer-by-layer electrostatic adsorption, with PDDA as the first layer and the nucleic acid as the second layer. The concentration of PDDA and nucleic acid used for film deposition was 2.0 mg/mL and 0.5 mg/mL respectively. The DNA film covered electrodes are denoted as SnO2/PDDA/CTDNA, SnO2/PDDA/Duplex-mC, and SnO2/PDDA/DuplexOH, respectively. DNA Methylation and Excision. The SnO2/PDDA/CTDNA electrodes described above were incubated in a chemical (in 20 mM PB, pH 6.5) at 37 °C with vortexing (200 rpm) for the required time, rinsed with high-purity water, and blown dry with nitrogen. Deposit 10 μL solution containing 1 U hAAG and 3 U APE1 on the DNA film and keep the electrodes in an incubator at 37 °C for 1 h. Finally, the electrodes were rinsed with 50 μM EDTA (eliminate the interference of magnesium ions in the buffer), blown dry with nitrogen, and ready for photocurrent measurement. Photoelectrochemical Measurement. Photocurrent was measured on a CHI 630A electrochemical analyzer (CH Instruments, Austin, TX) by placing the electrode in 30 mM 6909

dx.doi.org/10.1021/ac401346x | Anal. Chem. 2013, 85, 6908−6914

Analytical Chemistry

Article

oxalate buffer (pH 5.8), using a Pt flag counter electrode, a Ag/ AgCl (3 M KCl) reference electrode, and a bias voltage of +0.1 V. The area of the working electrode in contact with the electrolyte was 0.25 cm2. The light source for photocurrent measurement was a 473 nm blue laser with 1.5 mW/cm2 power and an illumination area of 0.18 cm2. Agarose Gel Electrophoresis. DNA sample for agarose gel electrophoresis was prepared by reacting 0.01 mg/mL CTDNA with 50 mM MMS for 1 h at 37 °C with vortexing (200 rpm). One μL of the exposed DNA sample and 1 μL of 1 U/μL hAAG and 1 μL of 1 U/μL APE 1 were added into 7 μL 1 × NEB buffer 4, reacted at 37 °C for 1 h. The sample was electrophoresed on a 0.6% agarose gel in 1 × TBE buffer (45 mM Tris, 45 mM boric acid, 1 mM EDTA, pH 8.0) for 120 min at 30 V. Ethidium bromide (0.5 μg/mL) was added in the agarose gel as the DNA stain. High Performance Liquid Chromatography Tandem Mass Spectrometry. Methylated CT-DNA samples for neutral thermal hydrolysis were prepared by reacting 200 μL of 0.5 mg/mL CT-DNA with 300 μL MMS for 1 h at 37 °C with vortexing (200 rpm). DNA was precipitated out with ethanol. Cleavage of the N-glycosidic bond of the methylated DNA bases (mainly metGuo and metAde) was achieved by heating the ethanol-precipitated DNA at 100 °C for 30 min in 5 mM phosphate (pH 7.5). The solution was then filtered by centrifugation (13000g, 30 min) using an ultrafiltration tube (cutoff mass 3000 Da) (Millipore, Biclerica, MA). The filtrate was analyzed on Agilent 1290 HPLC/6410B TripleQuad system (Agilent Technologies, Palo Alto, CA). The analytical column was Venusil MP C18 (5 μm, 100 Å, 4.6 × 250 mm) (Agela Technologies, Tianjin, China), maintained at 30 °C. A binary gradient consisting of 90% 5 mM ammonium acetate, pH 5.0, and 10% methanol was used for HPLC analysis at a rate of 0.8 mL/min. Injection volume was 20 μL. A microsplitter valve (Agilent Technologies, Palo Alto, CA) delivered 50% of the flow to the mass spectrometer. Capillary voltage was set at 3500 V. High purity nitrogen (99.999%) was used as collision gas. Ion source temperature was 300 °C, with nitrogen for nebulization and desolvation. Chromatograms were obtained in the positive ion and multiple reaction monitoring mode (ESI+MRM). The MRM mode was set as follow: m/z 166→149 for metGuo (collision energy, 5 eV); m/z 150→133 for metAde (5 eV). Quantification of the methylated bases was done by integrating the corresponding chromatographic peaks.

Figure 1. Illustration of the process of converting DNA methylation sites into DNA strand breaks for photoelectrochemical detection using DNA repair enzymes.

Figure 2. Photocurrent of Ru(bpy)2(dppz)2+ intercalated into the SnO2/PDDA/CT-DNA film after its incubation with (a) 20 mM phosphate buffer, (b) 50 mM MMS, (c) 20 mM phosphate then a mixture of 4 U hAAG and 4 U APE 1, (d) 50 mM MMS then a mixture of 4 U hAAG and 4 U APE 1. The photocurrent measurement was performed in 30 mM oxalate, pH 5.8. Arrows indicate light on and off.

■

RESULTS AND DISCUSSION We have previously developed a photoelectrochemical sensor for the detection of DNA damages using a photoactive DNA intercalator as the signal reporter.33 The intercalator responded to the drastic change of DNA structure associated with the formation of DNA strand breaks after the DNA was exposed to oxidizing chemicals. Detection of methylated DNA poses a great challenge. Previous study on DNA alkylation has suggested that MMS reacts with nucleobases preferentially on the N7 position of guanine and the N3 of adenine through the SN2 mechanism.2 Since the methyl group of metAde projects into the minor groove and metGuo projects into the major groove, neither of these methyl groups would interrupt hydrogen bonding interactions with its base pairing partner in ds-DNA.36 Thus, unlike DNA strand break, DNA methylation induces minimal change to the double helical structure of DNA. As will be shown later, the intercalator did not respond directly to DNA methylation. An additional difficulty is that the methyl

group is chemically inert, thus excludes itself from being labeled by chemical reactions. This is in contrast to 8-oxodGuo lesions, which can be chemically activated and then labeled with a biotin-containing conjugate.34,37 As mentioned in the Introduction, living organisms have developed a number of DNA base repairing strategies such as base excision repair (BER) to protect against genetic abnormalities. BER removes the damaged base before the site is fully repaired. Inspired by the living systems, we proposed to utilize the DNA repair enzymes to convert methylated DNA sites into strand breaks in order for the signaling molecule to respond (Figure 1). After DNA methylation, human alkyladenine DNA glycosylase (hAAG) would be used to remove the methylated adenine bases, producing abasic sites on DNA. Human apurinic/apyrimidinic endonuclease (APE 1) would then be employed to break up the DNA strand at the abasic sites and produce DNA fragments of various chain 6910

dx.doi.org/10.1021/ac401346x | Anal. Chem. 2013, 85, 6908−6914

Analytical Chemistry

Article

Figure 2 illustrates the results of photoelectrochemical measurement. On a SnO2/PDDA/CT-DNA electrode, the signal reporter Ru(bpy)2(dppz)2+ bound to the ds-DNA by intercalation and generated a photocurrent of about 550 nA. The signal did not change appreciably after the electrode was exposed to the methylating agent MMS, reflecting the minimal disruption of DNA helical structure after base methylation. Contact of the electrode with a mixture of hAAG and APE 1 in the absence of MMS did not attenuate the signal either. However, after the film was reacted with MMS and subsequently incubated with the two repair enzymes, photocurrent dropped by as much as 25%, presumably because of the conversion of the methylated DNA into strand breaks after enzyme excision. In our previous work, stable DNA films were fabricated successfully on SnO2 electrodes with short synthetic oligonucleotides (24 base pairs).34 Therefore, desorption of fragmented calf thymus DNA from the electrode surface after enzyme repair can be ruled out. To verify that the methylated CT-DNA was broke down to shorter fragments under the action of DNA repair enzymes, agarose gel electrophoresis of various DNA samples was carried out. CT-DNA is a heterogeneous polymer of various chain lengths with an average length of 13000 base pairs. In the electrophoresis image (Figure 3), the intact DNA sample displayed a major band with a length of a little shorter than 15K base pairs and many less intense bands with shorter lengths

Figure 3. Agarose gel electrophoresis images of various DNA samples. (1) DNA markers, Others: CT-DNA exposed to (2) 20 mM PB, (3) 50 mM MMS, (4) 50 mM MMS, then 1 U hAAG, (5) 50 mM MMS then 1 U APE 1, (6) 50 mM MMS then a mixture of 1 U hAAG and 1 U APE 1, and (7) 20 mM PB then a mixture of 1 U hAAG and 1 U APE 1.

lengths. Compared with native DNA, the signal reporter would bind much less to the DNA fragments, leading to a significant reduction in photoelectrochemical signal.

Figure 4. HPLC-MS chromatograms of CT-DNA after neutral thermal hydrolysis. (A) 100 nM metAde standard; (B) DNA incubated with 50 mM MMS, pH 6.5, 37 °C,for 1 h; (C) control; (D) 100 nM metGuo standard; (E) DNA incubated with 50 mM MMS, pH 6.5, 37 °C, for 1 h; (F) control. The retention times of metAde and metGuo are 6.8 and 16.2 min, respectively. 6911

dx.doi.org/10.1021/ac401346x | Anal. Chem. 2013, 85, 6908−6914

Analytical Chemistry

Article

Figure 6. (A) Relative photocurrent of Ru(bpy)2(dppz)2+ intercalated into the SnO2/PDDA/Duplex film after it was reacted with 1 U hAAG and 3 U APE1. Duplex-OH contains one 8-oxodGuo lesion in 25bp, and Duplex-mC contains one 5-mC in 25bp. (B) Relative photocurrent of Ru(bpy)2(dppz)2+ intercalated into the SnO2/PDDA/CTDNA film after it was incubated with 50 mM MMS, 50 mM EMS, or 2% styrene oxide, and then reacted with 1 U hAAG and 3 U APE1.

Figure 5. (A) Photocurrent of Ru(bpy)2(dppz)2+ intercalated into the SnO2/PDDA/CT-DNA film after it was incubated with MMS and then reacted with 1 U hAAG and 3 U APE 1. From a to f: 0, 1, 5, 10, 25, 50 mM MMS. Inset is the relative photocurrent as a function of MMS concentration. The error bar represents the standard deviation of 3 replicate electrodes. (B) The relative photocurrent as a function of the peak height of methylated adenine in the HPLC-MS chromatograms. Arrows indicate light on and off.

SnO2/PDDA/CT-DNA electrode after methylation and enzyme conversion decreased progressively with MMS concentration (Figure 5A). DNA methylation in the sensor film was detected with as low as 1 mM MMS. For each concentration, 3 replicate electrodes were measured, with the relative standard deviation ranging from 1.8% to 4.4%. The inset of Figure 5A is a plot of the relative photocurrent as a function of MMS concentration, which is quite linear in the range from 0 to 50 mM (R2 = 0.962). The linearity indicates that the sensor was responding to DNA methylation induced by MMS. Figure 5B is a plot of the relative photocurrent versus the peak height of methylated adenine (substrate of hAAG) in the HPLC-MS chromatograms. A reasonably good correlation was observed (R2 = 0.943). Taking together, the above results demonstrate that the developed photoelectrochemical sensor can detect DNA methylation induced by MMS, and the signal is dependent on the amount of methylated adenines in the sensor film. In our previous work, the amount of DNA immobilized on the electrode surface was measured using a quartz crystal microbalance.38 On the basis of the DNA concentration and electrode size used in the current work, the amount of CTDNA on the electrode should be less than 80 ng. From the HPLC-MS measurement, it can be estimated that the yield of methylated adenines by 1 mM MMS is 6.4 ng/mg CT-DNA. Assuming this solution phase yield holds true for the surface reaction, less than 0.5 pg methylated adenines on the electrode was detected by our sensor. This is equivalent to a detection limit of 1 metAde out of 1.6 × 105 normal bases. As mentioned in the Introduction, some chemical agents can produce more than one type of DNA damage. To correctly identify the DNA damage product after chemical exposure,

(lane 2). Reaction with MMS did not change the pattern of DNA electrophoresis image, which suggests there is no DNA strand breaks in the reaction products (lane 3). However, for the DNA sample reacted with both MMS and the mixture of hAAG and APE 1, the major band disappeared in the image. Instead, many new bands with much shorter lengths emerged, representing DNA strand breaks (lane 6). The electrophoresis results confirmed that hAAG and APE 1 acted together and converted the methylated DNA base into strand breaks. HPLC-MS measurement was then performed to identify the methylation products of CT-DNA with MMS. After reacting with MMS at 37 °C for 1 h,CT-DNA was neutral thermal hydrolyzed to release the methylated bases. Figure 4 summarizes the HPLC-MS analyses results of the products. Collision induced dissociation (CID) of metAde and metGuo standards displayed major fragment of m/z 149 and m/z 133, respectively, both corresponding to the loss of NH3. The MRM transition m/z 150 (metAde) →133 (Ade) and m/z 166 (metGuo)→149 (Guo) for the DNA hydrolysate produced peaks at 6.8 and 16.2 min. These two peaks correspond to the methylated DNA on guanine and adenine after reaction with MMS, which did not show in the native DNA. Compared to metGuo, the amount of metAde is much lower. This is consistent with the result reported in a previous study that approximately 82% of the MMS methylation product is metGuo.2 Under the optimized conditions for the DNA repair enzymes (1 U hAAG, 3 U APE 1, 1 h reaction time), the photocurrent of 6912

dx.doi.org/10.1021/ac401346x | Anal. Chem. 2013, 85, 6908−6914

Analytical Chemistry good selectivity is required for the sensor. The selectivity of the current sensor was evaluated in the following two experiments. In the first experiment, two types of oligonucleotide duplexes were synthesized and assembled on the SnO2 electrodes by the same protocol as CT-DNA. Duplex-OH and Duplex-mC are both 25 base pair long, and each respectively contains one 8oxodGuo lesion and one 5-methylcytosine base, as illustrated above. The duplex-covered film was incubated with hAAG and APE1 before its photocurrent was measured. As shown in Figure 6A, compared to the blank controls, the photocurrent of Duplex-OH decreased slightly, while that of Duplex-mC had no significant change. In organisms, DNA methyltransferases including Dnmt1, −3a and −3b that are responsible for establishing and maintaining the methylation state of 5methylcytosine across the genome.39 N-Methylpurine DNA glycosylase is not expected to repair 5-methylcytosine. Our photocurrent result suggests that hAAG may excise 8-oxodGuo in the duplex and convert the DNA into strand breaks. This is consistent with the report by Bessho et al. that N-methylpurine DNA glycosylase can indeed remove the oxidized bases.40 In another experiment of evaluating the sensor selectivity, the SnO2/PDDA/CT-DNA electrode was exposed separately to two other well-known DNA damaging agents, ethyl methanesulfonate (EMS) and styrene oxide (SO). These two chemicals are known to add an ethyl or styrene group to the ring nitrogen of purines, producing the analogues of 3methyladenine.41 As depicted in Figure 6B, the photocurrent of the film exposed to either 50 mM EMS or 2% SO did not change substantially relative to the blank, demonstrating good selectivity of the sensor toward methylated adenines. To our knowledge, this is the first time that chemically methylated DNA bases were detected selectively by electrochemical sensors.

ACKNOWLEDGMENTS

■

REFERENCES

This work was supported by the Ministry of Science and Technology of China (2012AA062804) and the National Natural Science Foundation of China (20825519). We thank Prof. Hailin Wang and Dr. Ruichuan Yin for help with HPLC experiments.

(1) Gates, K. S. Chem. Res. Toxicol. 2009, 22, 1747−1760. (2) Wyatt, M. D.; Pittman, D. L. Chem. Res. Toxicol. 2006, 19, 1580− 1594. (3) Beranek, D. T.; Weis, C. C.; Swenson, D. H. Carcinogenesis 1980, 1, 595−606. (4) Singer, B. Molecular Biology of Mutagens and Carcinogens; Plenum Press: New York, 1983. (5) Chaudhuri, I.; Essigmann, J. M. Carcinogenesis 1991, 12, 2283− 2289. (6) Lindahl, T.; Sedgwick, B.; Sekiguchi, M.; Nakabeppu, Y. Annu. Rev. Biochem. 1988, 57, 133−157. (7) Ezaz-Nikpay, K.; Verdine, G. L. Chem. Biol. 1994, 1, 235−40. (8) Lewis, R. J. Carcinogenically Active Chemicals: A Reference Guide; Van Nostrand Reinhold: New York, 1990. (9) Lindahl, T.; Karran, P.; Wood, R. D. Curr. Opin. Genet. Dev. 1997, 7, 158−169. (10) Lindahl, T. Nature 1993, 362, 709−715. (11) Brena, R. M.; Huang, T. H.-M.; Plass, C. J. Mol. Med. 2006, 84, 365−377. (12) Gupta, R.; Nagarajan, A.; Wajapeyee, N. Biotechniques 2010, 49, iii−xi. (13) Fraga, M. F.; Uriol, E.; Diego, L. B.; Berdasco, M.; Esteller, M.; Canal, M. J.; Rodriguez, R. Electrophoresis 2002, 23, 1677−1681. (14) Acevedo, L. G.; Sanz, A.; Jelinek, M. A. Epigenomics 2011, 3, 93−101. (15) Paleček, E.; Bartošík, M. Chem. Rev. 2012, 112, 3427−3481. (16) Kato, D.; Sekioka, N.; Ueda, A.; Kurita, R.; Hirono, S.; Suzuki, K.; Niwa, O. J. Am. Chem. Soc. 2008, 130, 3716−3717. (17) Goto, K.; Kato, D.; Sekioka, N.; Ueda, A.; Hirono, S.; Niwa, O. Anal. Biochem. 2010, 405, 59−66. (18) Kato, D.; Goto, K.; Fujii, S.-i.; Takatsu, A.; Hirono, S.; Niwa, O. Anal. Chem. 2011, 83, 7595−7599. (19) Wang, P.; Mai, Z.; Dai, Z.; Zou, X. Chem. Commun. 2010, 46, 7781−7783. (20) Tanaka, K.; Tainaka, K.; Kamei, T.; Okamoto, A. J. Am. Chem. Soc. 2007, 129, 5612−5620. (21) Liu, S.; Wu, P.; Li, W.; Zhang, H.; Cai, C. Chem. Commun. 2011, 47, 2844−2846. (22) Dai, Z.; Hu, X.; Wu, H.; Zou, X. Chem. Commun. 2012, 48, 1769−1771. (23) Hou, P.; Ji, M.; Ge, C.; Shen, J.; Li, S.; He, N.; Lu, Z. Nucleic Acids Res. 2003, 31, e92. (24) Sato, S.; Hokazono, K.; Irie, T.; Ueki, T.; Waki, M.; Nojima, T.; Kondo, H.; Takenaka, S. Anal. Chim. Acta 2006, 578, 82−87. (25) Sato, S.; Tsueda, M.; Takenaka, S. J. Organomet. Chem. 2010, 695, 1858−1862. (26) Wang, B.; Rusling, J. F. Anal. Chem. 2003, 75, 4229−4235. (27) Yang, J.; Wang, B.; Rusling, J. F. Mol. Biosyst. 2005, 1, 251−259. (28) Dong, D.; Zheng, D.; Wang, F.-Q.; Yang, X.-Q.; Wang, N.; Li, Y.-G.; Guo, L.-H.; Cheng, J. Anal. Chem. 2003, 76, 499−501. (29) Liang, M.; Liu, S.; Wei, M.; Guo, L.-H. Anal. Chem. 2006, 78, 621−623. (30) Wang, G.; Xu, J.; Chen, H.-Y. Sci. China, Ser. B:Chem. 2009, 39, 1336−1347. (31) Zhao, X.; Zhou, S.; Shen, Q.; Jiang, L.-P.; Zhu, J.-J. Analyst 2012, 137, 3697−3703. (32) Zhang, X.; Guo, Y.; Liu, M.; Zhang, S. RSC Adv. 2013, 3, 2846− 2857. (33) Liang, M.; Guo, L.-H. Environ. Sci. Technol. 2007, 41, 658−664.

■

CONCLUSIONS To conclude, a new photoelectrochemical sensor has been developed for the selective detection of methylated bases in DNA films. The sensor employed specific DNA repair enzymes to convert DNA methylation sites into strand breaks that a DNA intercalating signal reporter was able to recognize and respond to. The photocurrent signal was related to the amount of 3-methyladenine produced with a methylating agent, MMS. The sensor detected the methylated bases produced with as low as 1 mM MMS, at which concentration the amount of 3methyladenine on the sensor surface was estimated to be 42.5 fmol. The sensor also demonstrated good selectivity, as it discriminated against other base modifications such as 5methylcytosine and DNA adducts with ethyl and styrene groups. In principle, this strategy can be used to design sensors for other DNA base modifications as long as DNA repair enzymes specific for these modified bases are available. All the sensors can then be integrated into an array format that would provide a high throughput, high content tool for the rapid screening of DNA-damaging chemicals.

■

■

Article

AUTHOR INFORMATION

Corresponding Author

*Telephone: +86-10-62849685. E-mail: [email protected]. Notes

The authors declare no competing financial interest. 6913

dx.doi.org/10.1021/ac401346x | Anal. Chem. 2013, 85, 6908−6914

Analytical Chemistry

Article

(34) Zhang, B.; Guo, L.-H.; Greenberg, M. M. Anal. Chem. 2012, 84, 6048−6053. (35) Musumeci, S.; Rizzarelli, E.; Fragalà, I.; Sammartano, S.; Bonomo, R. P. Inorg. Chim. Acta 1973, 7, 660−664. (36) Abner, C. W. J. Biol. Chem. 2001, 276, 13379−13387. (37) Xue, L.; Greenberg, M. M. J. Am. Chem. Soc. 2007, 129, 7010− 7011. (38) Wei, M.-Y.; Guo, L.-H.; Chen, H. Microchim. Acta 2006, 155, 409−414. (39) Miranda, T. B.; Jones, P. A. J. Cell. Physiol. 2007, 213, 384−390. (40) Bessho, T.; Roy, R.; Yamamoto, K.; Kasai, H.; Nishimura, S.; Tano, K.; Mitra, S. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 8901−8904. (41) Setayesh, F. R.; DeCorte, B. L.; Horton, P.; Harris, C. M.; Harris, T. M.; Stone, M. P. Chem. Res. Toxicol. 1998, 11, 766−777.

6914

dx.doi.org/10.1021/ac401346x | Anal. Chem. 2013, 85, 6908−6914