Signal Amplification of Graphene Oxide Combining with Restriction

Aug 10, 2012 - Endonuclease for Site-Specific Determination of DNA Methylation ... determine DNA methylation at the site of CpG and has an ability to ...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/ac

Signal Amplification of Graphene Oxide Combining with Restriction Endonuclease for Site-Specific Determination of DNA Methylation and Assay of Methyltransferase Activity Wen Li, Ping Wu,* Hui Zhang, and Chenxin Cai* Jiangsu Key Laboratory of New Power Batteries, College of Chemistry and Materials Science, Nanjing Normal University, Nanjing 210097, P. R. China S Supporting Information *

ABSTRACT: Site-specific identification of DNA methylation and assay of MTase activity are important in determining specific cancer types, providing insights into the mechanism of gene repression, and developing novel drugs to treat methylation-related diseases. This work reports an electrochemical method for gene-specific methylation detection and MTase activity assay using HpaII endonuclease to improve selectivity and employing signal amplification of graphene oxide (GO) to enhance the assay sensitivity. The method was developed by designing a probe DNA, which was immobilized on electrode surface, to hybridize with target DNA (one 137 mer DNA from exon 8 promoter region of the Homo sapiens p53 gene, was extracted from HCT116 cells). The assay is based on the electrochemical responses of the reporter (thionine), which was conjugated to 3′-terminus of the probe DNA via GO, after the DNA hybrid was methylated (under catalysis of M.SssI MTase) and cleaved by HpaII endonuclease (a site-specific endonuclease recognizing the duplex symmetrical sequence of 5′-CCGG-3′ and catalyzing cleavage between the cytosines). This model can determine DNA methylation at the site of CpG and has an ability to discriminate the target DNA sequence from even single-base mismatched sequence. The electrochemical signal has a linear relationship with M.SssI activities ranging from 0.1 to 450 U/mL with a detection limit of ∼(0.05 ± 0.02) U/mL at a signal/noise of 3. The advantages of this assay are ease of performance having a good specificity and selectivity. In addition, we also demonstrate the method can be used for rapid evaluation and screening of the inhibitors of MTase and has a potential application in discovery of new anticancer drugs.

G

specific cancer types, providing insights into the mechanism of gene repression, and developing novel drugs to treat methylCpG-related cancer.4,16 Current DNA methylation assays involve technologies for distinguishing both methyl-cytosine and cytosine in DNA, including bisulfites,17,18 restriction enzymes,19−21 molecules with biological/chemical affinity,22,23 electrochemical methods,24−27 various PCR (polymerase chain reaction)-based techniques,28 fluorescence methods,29−31 colorimetric approaches,32−34 HPLC (high-performance liquid chromatography),35,36 etc. Two widely used assay methods are the bisulfitebased assay and restriction enzyme-based assay. In the bisulfite assay, cytosine is deaminated and then converted to uracil under the treatment of bisulfite, whereas methyl-cytosine remains almost unchanged, owing to the extremely low reactivity. The information about methyl-cytosine in DNA sequence is obtained by PCR after bisulfite treatment. Although this method is currently considered the gold standard assay technique, it still has various disadvantages, including that primers for specific positions are always needed, advanced

enomic DNA methylation, which refers to methyltransferase (MTase)-catalyzed covalent addition of a methyl group to adenine or cytosine residues in the specific DNA sequence,1,2 is one of the most important epigenetic events that plays a critical role in the regulation of gene transcription, eukaryotes development, and cellular differentiation as well as the pathogenesis of various human diseases such as cancers.3−5 Methylation in higher eukaryotic cells has frequently been observed at carbon 5 position of cytosine (C) in the 5′-CG-3′ sequence (CpG).6,7 In normal cells, most CpG islands (several tens to hundreds of CpG repetitions) spanning the promoter regions are unmethylated, and their downstream genes are transcriptionally active. In contrast, when promoter CpG islands in cancer cells are methylated, their downstream genes such as tumor suppressor genes are consistently silenced.3,8 Aberrant methylation of CpG islands in promoter regions of genes can change normal cellular functions and phenotypes9−11 and has been recognized as a potential biomarker of large scale epigenetic events, particularly useful for disease diagnosis and prognostics as well as to study the response of various organisms to environmental conditions.12−15 Thus, an analysis of DNA methylation is important with respect to early diagnosis of genetic diseases. In particular, the identification of site-specific CpG methylation can be used for determining © 2012 American Chemical Society

Received: July 15, 2012 Accepted: August 10, 2012 Published: August 10, 2012 7583

dx.doi.org/10.1021/ac301990f | Anal. Chem. 2012, 84, 7583−7590

Analytical Chemistry

Article

Figure 1. Schematic illustration of the procedures of the gene-specific DNA methylation detection and MTase activity assay based on the electrochemical signal amplification of GO and restriction endonuclease.

gene-specific methylation and assay of MTase activity. In this approach, we used HpaII endonuclease to improve selectivity and employed signal amplification of graphene oxide (GO) to enhance the assay sensitivity. The assay is based on the voltammetric response of an electrochemical reporter (thionine), which was conjugated to the 3′-terminus of the probe DNA via GO, after the DNA hybrid was methylated by methyltransferase (M.SssI) and further cleaved by HpaII endonuclease (Figure 1). The HpaII endonuclease, which is a widely used restriction enzyme in the gene-specific methylation assay, recognizes the duplex symmetrical sequence 5′-CCGG-3′ and catalyzes double-stranded cleavage between the cytosines.46 After being digested by HpaII, the DNA hybrid was cleaved at a specific site and electrochemical signals of the reporter were decreased or disappeared. However, the cleavage of the endonuclease is blocked by CpG methylation. Therefore, the voltammetric signal after cleavage is related to the methylation status and MTase activity, which forms the basis of MTase activity assay and site-specific methylation determination. We also demonstrated the application of the developed method for a rapid evaluation and screening of the inhibitors of MTase, which may be helpful for the discovery of anticancer drugs. Since aberrant DNA methylation has been detected in a wide range of cancers and can be used as a biomarker of cancer cells, the developed method can be used for the cancer risk assessment through human biomonitoring of the genomic DNA methylation level.

equipments are required, the work is laborious and timeconsuming, specificity still remains a challenge and, most importantly, the extreme duration of the reaction always causes intense DNA degradation.2,37 In the restriction enzyme assay, the restriction enzymes can sensitively recognize the methylation position and catalyze the cleavage of the specific base sequence. When the cytosine in the specific sequence is methylated, the cleavage reaction is inhibited. Although this method sometimes produces false positives due to incomplete cleavage of the specific DNA sequence by restriction enzymes, it can still provide an economical way to elucidate the methylation position of a sequence without the need to know the primary sequence.37 However, these methods were not designed to identify the DNA methylation and analyze the MTase activity simultaneously. Evaluating MTase activities is important in the fields of drug discovery and clinical diagnostics because MTases also play a key role in cellular differentiation and development, gene suppression, tumorigenesis, and genetic diseases.38,39 MTase has become a novel family of pharmacological targets for the treatment of tumors.40 Therefore, development of sensitive, selective, simple, and economical methods for DNA methylation determination and MTases activity assays is highly required. Electrochemical techniques have been widely used for DNA analysis.41−44 These techniques are simple and inexpensive and are expected to have promising ability to study DNA methylation.24−26 Several researchers have focused on electrochemical determination of DNA methylation by direct electrochemical oxidation of both methyl-cytosine and cytosine in oligonucleotides using an electron cyclotron resonance nanocarbon film electrode25,26 and a multiwalled carbon nanotube film electrode,45 respectively. These works are all global assays that only provide information of DNA methylation. The great challenges for electrochemical analysis of gene-specific methylation are the identification of the localization of methylated cytosines in the DNA sequence. Here, we report an electrochemical approach for detection of

■

EXPERIMENTAL SECTION Chemicals. All chemicals and solvents were of reagent grade or better. Hydrogen tetrachloroaurate trihydrate (HAuCl4·3H2O), graphite powder (99.998%, 325 mesh, Alfa Aesar), tris(hydroxymethyl)aminomethane (Tris), tris(2carboxyethyl)phosphine hydrochloride (TCEP, 98%), N-(3dimethylaminopropyl)-N′-ethylcarbodiimide (EDC), N-hydroxysuccinimide (NHS, 98%), dithiothreitol (DTT), bovine serum albumin (BSA), 5-azacytidine (5-Aza), and 5-aza-2′7584

dx.doi.org/10.1021/ac301990f | Anal. Chem. 2012, 84, 7583−7590

Analytical Chemistry

Article

thoroughly being rinsed with double distilled water, the AuNPscovered GC electrode was ready to be used for assembly of probe DNA (S1). The produced AuNPs have an average size of ∼25 nm (Figure S2, Supporting Information). The real surface area was estimated to be (0.34 ± 0.03) cm2 based on the amount of charge consumed during the reduction of the Au surface oxide monolayer in 1 M H2SO4 (Figure S3, Supporting Information) and a reported value of ∼400 μC/cm2 for Au.50 The large surface area of the AuNPs in comparison with the bare Au electrode will significantly enhance the voltammetric signals. Assembly of S1 on AuNP-Covered Electrode and Conjugation of Thionine/GO. The probe was assembled on the electrode surface by forming the Au−S bond between the −SH moiety at 5′-terminus of S1 and AuNPs. First, a mixture of 10 μL of S1 solution (10 μM) with 10 μL of TCEP solution (10 mM) was incubated for 1 h to reduce the disulfide bond at the 5′-terminus of S1 and generate a free thiol group for surface immobilization, followed by diluting the mixture to 100 μL with 10 mM Tris-HCl buffer (pH 7.4). Then, the AuNP-covered GC electrode was incubated in diluted S1 solution for 24 h (step (b) in Figure 1), resulting to the assembly of S1 on the electrode surface. Finally, the electrode was thoroughly rinsed with Tris-HCl buffer and water, in turn, and stored in buffer. Thionine/GO composites were conjugated to the electrode surface by formation of the amide between the group of −COOH at GO and −NH2 moiety at the 3′-terminus of the assembled S1 via EDC-NHS coupling. The thionine/GO was first activated in the EDC-NHS mixture (5 mM EDC, 10 mM NHS in Tris-HCl buffer, pH 7.4) for 30 min. After that, the S1assembled electrode was incubated into the mixture (containing 0.5 mg/mL thionine/GO) for 3 h (step (c) in Figure 1). The −NH2 group at S1 would react with the activated −COOH moiety, leading to covalently link thionine/GO at the 3′terminus of S1. Hybridization of S1 with Target DNA. Hybridization was conducted at 37 °C by incubating the thionine/GO-S1assembled electrode in 2 mL of 10 mM Tris-HCl buffer (pH 7.4) containing S2 (0.1 μM) or one-base mismatched DNA (S3, 0.1 μM) for 2 h (step (d) in Figure 1). Our results indicated that 2 h was long enough to ensure the complete hybridization. After hybridization, the electrode was thoroughly rinsed with water. Methylation of CpG and Inhibition. The methylation of S1/S2 hybrid was performed at 37 °C for 2 h in 10 mM TrisHCl buffer (pH 7.9) containing 160 μM SAM, 50 mM NaCl, 10 mM MgCl2, 1 mM dithiothreitol (DTT), and various concentration of M.SssI (from 0 to 550 U/mL) (step (e) in Figure 1). To study the inhibition effects of two representative anticancer drugs, 5-Aza and 5-Aza-dC, on the M.SssI activity, the methylation of S1/S2 hybrid was performed at 37 °C in 10 mM Tris-HCl buffer (pH 7.9) containing 160 μM SAM, 50 mM NaCl, 10 mM MgCl2, 1 mM dithiothreitol (DTT), 500 U/ mL M.SssI, and various concentration of the inhibitors (from 0 to 5 μM). Cleavage of HpaII endonuclease. HpaII cleavage was performed at 37 °C in 10 mM Tris-HCl buffer (pH 7.4) containing 20 U/mL HpaII, 50 mM NaCl, 0.1 mM EDTA, 1 mM dithiothreitol (DTT), 200 μg/mL bovine serum albumin (BSA), and 50% glycerol for 2 h (step (f) in Figure 1). After cleavage, the electrode was thoroughly washed and transferred

deoxycytidine (5-Aza-dC) were from Sigma-Aldrich and used as received. Thionine from Sigma-Aldrich was recrystallized thrice using water before use. S-Adenosylmethionine (SAM), E. coli CpG methyltransferase M.SssI, and E. coli restriction endonuclease HpaII were supplied by New England BioLabs (Ipswich, MA). All solutions were prepared with doubly distilled water. The synthetic thiol-capped single-stranded 32 mer probe DNA (S1, 5′-SH-(CH2)6-TTC TCT TCC TCT GTG CGC CGG TCT CTC CCA GG-(CH2)6-NH2-3′) and 137 mer modified Homo sapiens p53 gene containing one mismatch base (S3, 5′-CC TGG GAG AGA CCT GCG CAC AGA GGA AGA GAA-3′, the total DNA sequence is presented in Supporting Information) were purchased from BioSune Biotechnology (Shanghai, China). The probe DNA was designed to hybridize to target DNA from the promoter region of the Homo sapiens p53 gene (S1, 137 mer DNA, 5′-CC TGG GAG AGA CCG GCG CAC AGA GGA AGA GAA-3′, the total DNA sequence is presented in Supporting Information). The target DNA (S1) was extracted from HCT116 cells using a Genomic DNA Purification Kit (Promega) according to the manufacturer’s instruction and amplified by PCR. DNA concentration and quality were estimated by measuring the absorbance at 260 and 280 nm. The cells were cultured in RPMI 1640 medium containing 10% fetal calf serum, 100 U/mL penicillin, and 100 μg/mL streptomycin in 5% CO2 at 37 °C. Synthesis of Graphene Oxide (GO) Sheets. GO was synthesized by a modified Hummers method,47 starting from graphite powder (99.998%, 325 mesh, Alfa Aesar). The detailed procedures have been reported in our previous work.48,49 The synthesized GO sheets are rippled and resemble crumpled silk veil waves as viewed from transmission electron microscopic (TEM) images (Figure S1A, Supporting Information). The thickness of the GO sheet is ∼0.8 nm as indicated by atomic force microscopic (AFM) images (Figure S1B, Supporting Information). Before assembly of thionine and covalently conjugating on the electrode surface, GO was cut into small pieces with a size of ∼5 × 5 nm by ultrasonication (Figure S1C, Supporting Information). Assembly of Thionine on GO Surface. For assembly of thionine on the GO surface, 0.25 mg of GO was dispersed in 1 mL of thionine (1 mM) aqueous solution. After being stirred for 20 min at ambient temperature, the dispersion was filtered through a 0.22 μm Nylon filter, then thoroughly washed to remove the unadsorbed or loosely adsorbed thionine, and finally dried under vacuum at ambient temperature overnight to obtain thionine-assembled GO composites (denoted as thionine/GO hereafter). Deposition of Gold Nanoparticles (AuNPs) on GC Electrode Surface. For molecular diagnostics of human diseases, a short DNA probe is usually terminally anchored or chemically grafted to a solid surface. In this work, the probe DNA was immobilized on the glassy carbon (GC) electrode surface (3-mm in diameter, CH Instruments). Prior to immobilization, a layer of AuNPs was produced on the electrode surface for the aim of signal amplification. The AuNPs was electrodeposited on the electrode surface from an aqueous solution of 1 M H2SO4 containing 1 mM AuCl4− at a constant potential of −0.4 V (versus saturated calomel electrode, SCE) for 10 min (step (a) in Figure 1). Before deposition, the solution was deaerated by purging high-purity nitrogen, and the nitrogen environment was then kept over the solution to prevent oxygen from reaching the solution. After 7585

dx.doi.org/10.1021/ac301990f | Anal. Chem. 2012, 84, 7583−7590

Analytical Chemistry

Article

into acetate buffer (0.1 M, pH 5) to record the electrochemical response (step (g) in Figure 1). Electrochemical Measurements. Cyclic voltammetry (CV) and differential pulse voltammetry (DPV) experiments were performed with an Autolab PGSTAT302N electrochemical station (Metrohm). A two-compartment threeelectrode cell with a sample volume of 5 mL was employed. A coiled Pt wire and a SCE were used as the counter and the reference electrodes, respectively. The buffer (10 mM phosphate buffer, PBS, pH 7.4) was purged with high-purity nitrogen for at least 30 min prior to each electrochemical measurement, and the nitrogen environment was then kept over the solution to prevent oxygen from reaching the solution. DPV signals were measured using a potential step of 5 mV, pulse width of 25 ms, pulse period of 100 ms, and pulse amplitude of 50 mV.

Figure 2. Voltammetric responses of the thionine/GO-S1-assembled (a), S1-assembled (b), and GO-S1-assembled GC electrode (c) in 10 mM PBS (pH 7.4) at a scan rate of a 50 mV/s.

■

RESULTS AND DISCUSSION Figure 1 illustrates the procedures of site-specific DNA methylation detection and MTase activity assay based on the signal amplification of GO and specific endonuclease. One 137 mer sequence (S2) in the promoter region of the Homo sapiens p53 gene and one CpG site (5′-CCGG-3′) in this sequence were of interest in this work since p53 has proved to be a tumor-suppressor gene. The methylation of the promoter region of p53 plays a key role in the development of at least 60% of tumors of the colon, breast, lung, ovaries, cervix, adrenal cortex, bone, and bladder and at least 30% of brain tumors.51 To fabricate the sensing platform, one synthetic DNA probe (S1), which was designed to hybridize to target DNA from promoter region of Homo sapiens p53 gene, was immobilized on the AuNP-covered GC electrode surface. An electrochemical reporter (thionine) was subsequently conjugated to it via GO. The aim of covering the GC electrode surface with AuNPs (∼25 nm in diameter) and conjugating thionine to probe DNA via GO is to amplify the voltammetric signals and to enhance the assay sensitivity. We used the cyclic voltammetry (CV) to characterize the conjugation of thionine with probe DNA since the anodic and cathodic peaks (including peak potentials and currents) can be studied simultaneously and the reversibility of the redox reaction can be easily analyzed with the use of this technique. CV response of the thionine/GO-S1-assembled electrode in 10 mM PBS (pH 7.4) exhibits a pair of well-defined redox peaks with anodic and cathodic peak potential of approximately −225 and −265 mV (vs SCE, 50 mV/s, pH 7.4, Figure 2a), respectively, which are the characteristic redox features of thionine.52,53 However, the S1-assembled (Figure 2b) and GOS1-assembled electrodes (without assemble of thionine on the GO surface, Figure 2c) do not show such the characteristic redox response, verifying the assembly of thionine on the electrode surface via GO. Moreover, the assembled thionine can undergo quasi-reversible electrochemical reaction because the potential separation of the redox peaks is small (∼40 mV) and the height of the anodic and cathodic peaks is almost the same. These results ensure that the methylation level of CpG in S2 and M.SssI activity can be directly indicated by the voltammetric signal of thionine after cleavage with HpaII endonuclease. Integration of either anodic or cathodic peak, corrected from the background current, gives the Faradaic charge required for the full oxidation (or full reduction) of the conjugated thionine. The total amount of thionine on the electrode surface (Γ, in

mol/cm2) was evaluated to be ∼(8.5 ± 1.0) × 10−10 mol/cm2, which was deduced using Γ = Q/nFA, where n is the number of electrons transferred (n = 2 for thionine in this work), F is the Faraday constant, Q is the voltammetric peak area in coulombs, and A is the effective surface area of AuNP-covered electrode (in cm2). This value is equivalent to surface density of (5.1 ± 0.6) × 1014 thionine molecules/cm2 (an average value of five independent measurements). Such a surface density suggests that every S1 on the electrode surface has been conjugated with ∼10−25 thionine molecules on average because the surface density of the self-assembled monolayer of single-stranded oligonucleotides (20−30 mer) via sulfur−gold affinity is ∼(2− 5) × 1013 molecules/cm2.54,55 This feature suggests that the conjugation of thionine with S1 via GO is effective and has high efficiency. By this way, GO can amplify the measured electrochemical signal and thus enhance the assay sensitivity. We used DPV to sense the hybridization of the probe DNA with target DNA and cleavage efficiency of the endonuclease since the DPV technique has much higher sensitivity than conventional sweep techniques when detecting very low concentrations of redox probe. DPV response of the thionine/GO-S1 shows an anodic peak at approximately −235 mV (Figure 3a), which is similar to that obtained by cyclic voltammetry. After being hybridized with S2, the position and height of this DPV peak (Figure 3b) are almost identical with those shown in Figure 3a, suggesting hybridization of S1 with S2 does not affect DPV features of the conjugated thionine. After the S1/S2 hybrid (unmethylated) was treated with HpaII endonuclease for 10 min at 37 °C and then rinsed with copious amounts of water, the DPV response of thionine decreased significantly (Figure 3c), suggesting that some thionine molecules have been removed from the electrode surface due to the cleavage of the endonuclease. The signal decreased further with prolonging the cleavage time (Figure 3d) and completely disappeared after a 2 h cleavage (Figure 3e). However, if the S1/S2 hybrid was methylated at the site of CG by M.SssI, the treatment of HpaII endonuclease does not alter the DPV characteristics of thionine (Figure 3f) because the cleavage of the endonuclease is blocked by CpG methylation. To further verify that the disappearance of DPV signal of thionine in Figure 3e is directly related to the specific cleavage of the HpaII endonuclease, a 137-mer modified Homo sapiens p53 gene containing one-mismatch base (S3, 5′-CCT GGG 7586

dx.doi.org/10.1021/ac301990f | Anal. Chem. 2012, 84, 7583−7590

Analytical Chemistry

Article

Figure 3. DPV responses of thionine/GO-S1 (a) and thionine/GO-S1/S2 (b) in 10 mM PBS (pH 7.4). Curves (c) and (e) are the DPV response of thionine/GO-S1/S2 hybrid after 10 min and 2 h, respectively, HpaII cleavage at 37 °C in 10 mM Tris-HCl buffer (pH 7.4). Curve (d) shows the dependence of DPV signal of thionine/GO-S1/S2 hybrid on the cleavage time with HpaII. Error bars are based on three experiments. Curves (f) and (g) are DPV responses of thionine/GO-S1/S2 (f) and thionine/GO-S1/S3 (g), respectively, after the hybrids were methylated by M.SssI for 2 h at 37 °C in 10 mM Tris-HCl buffer (pH 7.9) and then cleaved by HpaII for 2 h.

Figure 4. (A) Effects of methylation time on DPV responses of thionine/GO-S1/S2 hybrid in 10 mM PBS (pH 7.4). The responses were recorded after the hybrid was cleaved by HpaII for 2 h. The methylated time is (a) 0, (b) 5, (c) 20, (d) 60, (e) 80, (f) 100, and (g) 120 min, respectively. (B) Dependence of DPV signals of thionine on methylation time. Every point is an average value of three measurements.

AGA GAC CTG CGC ACA GAG GAA GAG AA-3′, the total DNA sequence is presented in Supporting Information) was synthesized and hybridized with thionine/GO-S1. After the S1/ S3 hybrid was treated with the endonuclease, DPV signal was recorded (Figure 3g). It is obvious that the signal of thionine is still observed (Figure 3g). Moreover, the height remains almost invariant in comparison with those in Figure 1a,b,f, suggesting that HpaII has no effect on the S1/S3 hybrid because the hybrid does not contain a specific recognition sequence for the endonuclease. The results demonstrate that the interaction model of M.SssI-HpaII endonuclease can be used for highly selective and rapid determination of DNA methylation at the site of CpG and has an ability to discriminate the target DNA sequence from even a single-base mismatched sequence.

We also studied the dependence of the DPV signal on the concentrations of S2. After thionine/GO-S1 was hybridized with different concentrations of S2, methylated by M.SssI, and cleaved by HpaII, the DPV signal was recorded. The results showed that the method could show a response to an S2 concentration as low as ∼0.1 pM, indicative of the high sensitivity of the method. In response to these previous reported methods for methylation detection, the proposed method, which operates via directly detecting the electrochemical signal, provides a rapid simple site-specific methylation assay because it does not require bisulfite conversion, separation such as chromatography, and electrophoresis techniques. It also does not need the addition of reagents or radio-labeled substrates, capable of 7587

dx.doi.org/10.1021/ac301990f | Anal. Chem. 2012, 84, 7583−7590

Analytical Chemistry

Article

Figure 5. Effect of M.SssI concentration on the DPV response of the thionine/GO-S1/S2 hybrids. The concentration of M.SssI is (a) 0, (b) 0.5, (c) 5.0, (d) 25, (e) 50, (f) 100, (g) 150, (h) 200, (i) 250, (j) 300, (k) 350, (l) 400, (m) 450, and (n) 550 U/mL, respectively. Before DPV was recorded, the hybrids were methylated by M.SssI (2 h) and then cleaved by HpaII for 2 h. (B) The plot of DPV signal vs M.SssI concentration. Every point is an average value of three measurements.

functioning in complex samples. Furthermore, our method using GO to amplify the signal might overcome some of the problems associated with the thermal and environmental instability inherent in biological materials such as enzymes, which are usually used in redox amplification. We further used this method to monitor the process of methylation and to quantify the methylation level. To evaluate the effects of methylation time on the methylation process, the thionine/GO-S1/S2 hybrid was methylated with a various time under catalysis of M.SssI, and then, DPV signal was recorded after the methylated hybrid was cleaved by HpaII for 2 h (Figure 4A). With an increase of methylation time, DPV signal increases rapidly (the slope is ∼0.086 ± 0.012 μA/min, stage I in Figure 4B). However, when the methylation time prolongs, the increasing rate of DPV signal gradually decreases (the slope is ∼0.042 ± 0.002 μA/min, stage II in Figure 4B), suggesting the drop in rate of the M.SssI-catalyzed methylation reaction. This feature is probably caused by the consumption of the substrates. These values of slopes can be used to evaluate the methylation rate of CpG site catalyzed by M.SssI at different stages. The methylation will be completed within 2 h since the height of DPV peak reaches a plateau at that time (stage III in Figure 4B). We applied the method to quantitatively determine the M.SssI activity. To assess the analytical performance of the model, the thionine/GO-S1/S2 hybrids were methylated with various concentrations of M.SssI (0−550 U/mL) for 2 h and then cleaved by HpaII for 2 h and thoroughly washed with water, and finally, DPV response of thionine was recorded in 10 mM PBS (pH 7.4). DPV peak current increases with the concentration of M.SssI (Figure 5A). This is in accordance with the fact that at higher concentration of M.SssI, more S1/S2 hybrids are methylated. The signal increases linearly with M.SssI concentration (in U/mL) from 0.1 to 450 U/mL with a detection limit of ∼(0.05 ± 0.02) U/mL at a signal/noise of 3 and then levels off at higher concentration (Figure 5B). This is because almost all of the S1/S2 hybrids are methylated at a high concentration of M.SssI and subsequently cannot be cleaved by HpaII. DPV signal remains practically constant even though the concentration of M.SssI is further increased. The detection range of the method is much wider than that of previously reported MTase assays based on colorimetric approaches. The detection limit is also much lower than that obtained using cross-linking AuNPs aggregation (2.5 U/mL)32 and methylation-responsive DNAzyme methods (6 U/mL).34

The assay repeatability of the method was estimated from the DPV response of 100 U/mL of M.SssI with five different and freshly fabricated detection systems. The RSD (relative standard deviation) is ∼3.1%. The assay precision was examined from the slopes of calibration plots obtained from five independent detection systems. The RSD of these slopes is ∼3.4%. These results suggest the method can be used for quantitatively analyzing MTase activity conveniently and efficiently. We also demonstrated the validity of our method in evaluating and screening the inhibitors of M.SssI with use of 5-azacytidine (5-Aza) and 5-aza-2′-deoxycytidine (5-Aza-dC) as model inhibitors. Both compounds are representative anticancer drugs and have been used in the majority of methylation inhibition experiments and also in a large number of clinical trials.56 Before studying the inhibited effects of the anticancer drugs M.SssI, we first evaluated the influence of the inhibitors on HpaII activity. The experiments were performed by cleaving the thionine/GO-S1/S2 hybrid with the endonuclease for 2 h in the presence of different concentrations of the inhibitors (5Aza or 5-Aza-dC); DPV signal was then recorded and compared with that obtained for cleavage in the absence of the inhibitors. The results indicated that the two drugs have no influence on the activity of HpaII when the concentration of 5Aza and 5-Aza-dC is not more than 10 and 2 μM, respectively (not shown here). After that, we evaluated the inhibited effects of the two drugs on M.SssI activity by comparing DPV signal under various concentrations of the drug. The activity of the M.SssI decreases with the increasing concentration of 5-Aza and 5-Aza-dC (Figure 6), showing significant dose-dependent inhibition of the methylation of genomic DNA by 5-Aza and 5-Aza-dC. The inhibition of DNA methylation by M.SssI (about 10%) can be observed even when treated with trace 5-Aza-dC as low as 0.01 μM. The IC50 value, the inhibitor concentration required to reduce enzyme activity by 50%, is acquired from the plots of relative activity of M.SssI versus the inhibitors concentration and is found to be ∼0.4 and 3.5 μM for 5-Aza and 5-Aza-dC, respectively. These results indicate that the developed method has a potential application in studying the inhibited effects of anticancer drugs on MTase and for screening MTase inhibitors. It is noteworthy that such kind of study on the pharmacological inhibition of DNA MTase can provide a broad spectrum of therapeutic applications such as antibiotics and anticancer therapeutics, because DNA methylation plays an important role in both prokaryotes and eukaryotes. 7588

dx.doi.org/10.1021/ac301990f | Anal. Chem. 2012, 84, 7583−7590

Analytical Chemistry

Article

Program for Outstanding Innovation Research Team of Universities in Jiangsu Province, and the Priority Academic Program Development of Jiangsu Higher Education Institutions.

■

(1) Lee, G. E.; Kim, J. H.; Taylor, M.; Muller, M. T. J. Biol. Chem. 2010, 285, 37630−37640. (2) Liu, S.; Tu, Y.; Li, W.; Wu, P.; Zhang, H.; Cai, C. X. Chin. J. Anal. Chem. 2011, 39, 1451−1458. (3) Frigola, J.; Song, J.; Stirzaker, C.; Hinshelwood, R. A.; Peinado, M. A.; Clark, S. J. Nat. Genet. 2006, 38, 540−549. (4) Choy, J. S.; Wei, S.; Lee, J. Y.; Tan, S.; Chu, S.; Lee, T. H. J. Am. Chem. Soc. 2010, 132, 1782−1783. (5) Scholten, S. Epigenetics 2010, 5, 455−459. (6) Stains, C. I.; Furman, J. L.; Segal, D. J.; Ghosh, I. J. Am. Chem. Soc. 2006, 128, 9761−9765. (7) Branciamore, S.; Chen, Z. X.; Riggs, A. D.; Rodin, S. N. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 15485−15490. (8) Zhang, Z. J.; Chen, C. Q.; Manev, H. Anal. Chem. 2004, 76, 6829−6832. (9) Heard, E.; Clerc, P.; Avner, P. Annu. Rev. Genet. 1997, 31, 571− 610. (10) Baylin, S. B.; Herman, J. G. Trends Genet. 2000, 16, 168−174. (11) Chen, R. Z.; Pettersson, U.; Beard, C.; Jackson-Grusby, L.; Jaenisch, R. Nature 1998, 395, 89−93. (12) Compare, D.; Rocco, A.; Liguori, E.; D’Armiento, F. P.; Persico, G.; Masone, S.; Coppola-Bottazzi, E.; Suriani, R.; Romano, M.; Nardone, G. J. Clin. Pathol. 2011, 64, 677−682. (13) Terry, M. B.; Delgado-Cruzata, L.; Vin-Raviv, N.; Wu, H. C.; Santella, R. M. Epigenetics 2011, 6, 828−837. (14) Kim, M.; Long, T. I.; Arakawa, K.; Wang, R.; Yu, M. C.; Laird, P. W. PLoS One 2010, 5, e9692. (15) Wright, R. O.; Schwartz, J.; Wright, R. J.; Bollati, V.; Tarantini, L.; Park, S. K.; Hu, H.; Sparrow, D.; Vokonas, P.; Baccarelli, A. Environ. Health Perspect. 2010, 118, 790−795. (16) Esteller, M.; Corn, P. G.; Baylin, S. B.; Herman, J. G. Cancer Res. 2001, 61, 3225−3229. (17) Docherty, S. J.; Davis, O. S.; Haworth, C. M.; Plomin, R.; Mill, J. Methods 2010, 52, 255−258. (18) Aichinger, E.; Kohler, C. Methods Mol. Biol. 2010, 655, 433− 443. (19) Oakes, C. C.; La Salle, S.; Trasler, J. M.; Robaire, B. Methods Mol. Biol. 2009, 507, 271−280. (20) Bruce, S.; Hannula-Jouppi, K.; Lindgren, C. M.; LipsanenNyman, M.; Kere, J. Clin. Chem. 2008, 54, 491−499. (21) Blair, S.; Yu, Y. N.; Gillespie, D.; Jensen, R.; Myszka, D.; Badran, A. H.; Ghosh, I.; Chagovetz, A. Anal. Chem. 2010, 82, 5012−5019. (22) Craighead, H. G.; Cipriany, B. R.; Zhao, R. Q.; Murphy, P. J.; Levy, S. L.; Tan, C. P.; Soloway, P. D. Anal. Chem. 2010, 82, 2480− 2487. (23) Tanaka, K.; Tainaka, K.; Umemoto, T.; Nomura, A.; Okamoto, A. J. Am. Chem. Soc. 2007, 129, 14511−14517. (24) Dai, Z.; Hu, X.; Wu, H.; Zou, X. Chem. Commun. 2012, 48, 1769−1771. (25) Kato, D.; Sekioka, N.; Ueda, A.; Kurita, R.; Hirono, S.; Suzuki, K.; Niwa, O. J. Am. Chem. Soc. 2008, 130, 3716−3717. (26) Kato, D.; Goto, K.; Fujii, S.-I.; Takatsu, A.; Hirono, S.; Niwa, O. Anal. Chem. 2011, 83, 7595−7599. (27) Liu, S.; Wu, P.; Li, W.; Zhang, H.; Cai, C. X. Chem. Commun. 2011, 47, 2844−2846. (28) Lyko, F.; Ramsahoye, B. H.; Jaenisch, R. Nature 2000, 408, 538−540. (29) Duan, X.; Liu, L.; Feng, F.; Wang, S. Acc. Chem. Res. 2010, 43, 260−270. (30) Wang, X.; Song, Y.; Song, M.; Wang, Z.; Li, T.; Wang, H. Anal. Chem. 2009, 81, 7885−7891.

Figure 6. Dose-dependent inhibition of M.SssI activity by two typical anticancer drugs 5-Aza and 5-Aza-dC. The relative activity of M.SssI is represented as the ratio of DPV signal of thionine conjugated to S1/S2 hybrids, which are methylated by M.SssI under various concentrations of inhibitors, to that obtained in the absence of inhibitors. Before DPV signal was recorded, the thionine/GO-S1/S2 hybrids were methylated for 2 h by M.SssI at different concentrations of 5-Aza or 5-Aza-dC and then cleaved by HpaII for 2 h. Every point is an average value of three measurements.

■

CONCLUSIONS In summary, we have described a new method for site-specific methylation detection and MTase activity assay using HpaII endonuclease to improve selectivity and employing signal amplification of GO to enhance the assay sensitivity. The detection is based on the voltammetric signal of electrochemical reporter conjugated to probe DNA after the DNA hybrid was methylated by M.SssI and cleaved by HpaII endonuclease. The assay is easily performed, requiring no bisulfite conversion, PCR amplification, or separation; has a good specificity and selectivity; and has the ability to perform real-time monitoring. In addition, we also demonstrated the application of the developed method for the rapid and sensitive assay of the activity of MTase and evaluation and screening of the inhibitors of MTase, which may be helpful for the discovery of anticancer drugs. With its high sensitivity and high speed, the developed method has a potential application in cancer risk assessment through human biomonitoring of the DNA methylation level and in analysis of environmental exposurerelated DNA methylation change.

■

ASSOCIATED CONTENT

S Supporting Information *

Total sequence of S1 and S3; TEM and AFM images of GO; SEM image and voltammetric response of the AuNPs. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (P.W.); [email protected] (C.C.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by the National Natural Science Foundation of China (21175067), the Research Fund for the Doctoral Program of Higher Education of China (20103207110004), the Natural Science Foundation of Jiangsu Province (BK2011779), the Foundation of the Jiangsu Education Committee (09KJA150001 and 10KJB150009), the 7589

dx.doi.org/10.1021/ac301990f | Anal. Chem. 2012, 84, 7583−7590

Analytical Chemistry

Article

(31) Zhao, C.; Qu, K.; Song, Y.; Ren, J.; Qu, X. Adv. Funct. Mater. 2011, 21, 583−590. (32) Song, G.; Chen, C.; Ren, J.; Qu, X. ACS Nano 2009, 3, 1183− 1189. (33) Liu, T.; Zhao, J.; Zhang, D.; Li, G. Anal. Chem. 2010, 82, 229− 233. (34) Li, W.; Liu, Z.; Lim, H.; Nie, Z.; Chen, J.; Xu, X.; Yao, S. Anal. Chem. 2010, 82, 1935−1941. (35) Friso, S.; Choi, S. W.; Dolnikowski, G. G.; Selhub, J. Anal. Chem. 2002, 74, 4526−4531. (36) Torres, A. L.; Barrientos, E. Y.; Wrobel, K.; Wrobel, K. Anal. Chem. 2011, 83, 7999−8005. (37) Tian, T.; Wang, S. R.; Wu, J. G.; Zhou, X. Sci. China 2011, 54, 1233−1243. (38) Sahin, E.; Gumuslu, S.; Erdogan, A.; Gultekin, M. Cancer Metastasis Rev. 2010, 29, 655−676. (39) Lee, J.; Kim, Y.-K.; Min, D.-H. Anal. Chem. 2011, 83, 8906− 8912. (40) Esteller, M. Oncogene 2002, 21, 5427−5440. (41) Lubin, A. A.; Plaxco, K. W. Acc. Chem. Res. 2010, 43, 496−505. (42) Liu, S.; Hu, Y.; Jin, J.; Zhang, H.; Cai, C. X. Chem. Commun. 2009, 1635−1637. (43) Liu, S.; Wu, P.; Li, W.; Zhang, H.; Cai, C. X. Anal. Chem. 2011, 83, 4752−4758. (44) Patolsky, F.; Katz, E.; Willner, I. Angew. Chem., Int. Ed. 2002, 41, 3398−3402. (45) Wang, P.; Dai, Z.; Dai, Z.; Zou, X. Chem. Commun. 2010, 46, 7781−7783. (46) Quint, A.; Cedar, H. Nucleic Acids Res. 1981, 9, 633−646. (47) Kovtyukhova, N. I.; Ollivier, P. J.; Martin, B. R.; Mallouk, T. E.; Chizhik, S. A.; Buzaneva, E. V.; Gorchinskiy, A. D. Chem. Mater. 1999, 11, 771. (48) Hu, Y. J.; Wu, P.; Yin, Y. J.; Zhang, H.; Cai, C. X. Appl. Catal., B: Environ. 2012, 111−112, 208−217. (49) Wu, X.; Hu, Y.; Jin, J.; Zhou, N.; Wu, P.; Zhang, H.; Cai, C. X. Anal. Chem. 2010, 82, 3588−3596. (50) Szamocki, R.; Velichko, A.; Holzapfel, C.; Mucklich, F.; Ravaine, S.; Garrigue, P.; Sojic, N; Hempelmann, R.; Kuhn, A. Anal. Chem. 2007, 79, 533−539. (51) Nigro, J. M.; Baker, S. J.; Preisinger, A. C.; Jessup, J. M.; Hostetter, R.; Cleary, K.; Bigner, S. H.; Davidson, N.; Baylin, S.; Devilee, P.; Glover, T.; Collins, F. S.; Weston, A.; Modali, R.; Harris, C. C.; Vogelstein, B. Nature 1989, 342, 705−707. (52) Chen, D. X.; Wang, Q.; Jin, J.; Wu, P.; Wang, H.; Yu, S.; Zhang, H.; Cai, C. X. Anal. Chem. 2010, 82, 2448−2455. (53) Meng, L.; Wu, P.; Chen, G. X.; Cai, C. X.; Sun, Y. M.; Yuan, Z. H. Biosens. Bioelectron. 2009, 24, 1751−1756. (54) Fan, C.; Plaxco, L. W.; Heeger, A. J. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 9134−9137. (55) Zhang, Y.; Wang, Y.; Wang, H.; Jiang, J.-H.; Shen, G.-L.; Yu, R.Q.; Li, J.-H. Anal. Chem. 2009, 81, 1982−1987. (56) Christman, J. K. Oncogene 2002, 21, 5483−5495.

7590

dx.doi.org/10.1021/ac301990f | Anal. Chem. 2012, 84, 7583−7590