Analytica Chimica Acta 804 (2013) 252–257
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Target-mediated consecutive endonuclease reactions for specific and sensitive homogeneous fluorescence assay of O6 -methylguanine-DNA methyltransferase Dinh-Vu Le a,b , Dian-Ming Zhou a , Li-Juan Tang a,∗ , Jian-Hui Jiang a , Ru-Qin Yu a , Yu-Zhi Wang a,∗ a State Key Laboratory for Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082, PR China b Faculty of Chemical Engineering, Industrial University of Hochiminh City, Viet Nam
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• A homogeneous fluorescence strategy for O6 -methylguanine-DNA methyltransferase activity assay. • Based on target-mediated consecutive endonuclease reactions. • Simple, robust, highly sensitive and selective. • A potential convenient platform for alkyltransferase activity assay and related studies.
a r t i c l e
i n f o
Article history: Received 23 April 2013 Received in revised form 11 October 2013 Accepted 16 October 2013 Available online 22 October 2013 Keywords: O6 -Methylguanine DNA methyltransferase DNA-repair enzyme Alkyltransferase Restriction endonuclease Fluorescence
a b s t r a c t O6 -Methylguanine-DNA methyltransferase (MGMT) is one of the most important DNA-repair enzymes. Herein, a simple, sensitive and selective homogeneous fluorescence assay strategy is developed for the detection of MGMT on the basis of target-mediated two consecutive endonuclease reactions. The activity assay of MGMT is firstly accomplished using a hairpin-structured DNA substrate to offer a specific recognition site on the substrate DNA for restriction endonuclease PvuII, and thus to initiate the first endonuclease reaction. The product which activates the second endonuclease reaction allows an efficient amplification approach to create an abundance of fluorescence signal reporters. The first endonuclease reaction offers the method high specificity and the second one furnishes the assay improved sensitivity. The results reveal that the MGMT assay strategy shows dynamic responses in the concentration range from 1 to 120 ng mL−1 with a detection limit of 0.5 ng mL−1 . By simply altering the alkylated bases, this strategy can also be extended for the detection of other alkyltransferases. Therefore, the developed strategy might provide an intrinsically convenient, sensitive and specific platform for alkyltransferase activate assay and related biochemical studies due to its label-free, homogeneous, and fluorescence-based detection format. © 2013 Elsevier B.V. All rights reserved.
1. Introduction
∗ Corresponding authors. Tel.: +86 731 88821903; fax: +86 731 88821848. E-mail addresses: [email protected] (L.-J. Tang), [email protected] (Y.-Z. Wang). 0003-2670/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.aca.2013.10.036
Electrophilic alkylation caused by mutagens in the environment, in food, or endogenous metabolic products is often highly toxic, due to its ability to alkylate the bases of DNA and result in miscoding of DNA strand. Compared with other DNA methyl adducts, O6 -methylguanine (O6 MeG) is the most critical lesion
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257
for its pre-mutagenic and pre-toxic [1,2]. O6 -Methylguanine DNA methyltransferase (MGMT) is a DNA-repair enzyme which repairs O6 -alkylation adducts in a one-step alkyl transfer reaction by removing alkyl group from O6 -guanine in DNA [3–5]. The activity screening of MGMT has emerged as an essential step in investigating O6 MeG induced cell death, chromosomal aberrations, mutations and cancer [6–8]. Radioactive methods play an important role in MGMT assay by using 3 H labeled alkyl adduct in DNA as the substrate [9,10]. However, extensive multiple-step sample treatments (e.g. hydrolysis, extractions, concentration, etc.), which take tens of minutes or even hours, are normally required by these methods [11–14]. Timeconsuming separations are also needed to clean up the matrixes by using separation techniques such as liquid chromatography and capillary electrophoresis [15,16]. Although detection of MGMT can also be carried out by polymerase chain reaction (PCR) [17,18], mass spectrometry (MS) [19,20], fluorescence [21], or enzymelinked immunosorbent assay (ELISA) [22], these methods may still suffer from some drawbacks such as limited sensitivity, timeconsuming analysis, requirements for expensive equipments, or multistep washing and separation. In this context, the development of sensitive and selective methods for convenient MGMT activity monitoring is still a topic of intensive interest in bioanalytical chemistry. In the present study, a homogeneous fluorescence assay strategy is developed for activity screening of MGMT on the basis of targetmediated consecutive endonuclease reactions. In this strategy, active MGMT can offer a specific recognition site on the substrate DNA for restriction endonuclease PvuII [23] and thus initiates the first endonuclease reaction. The product of PvuII-catalyzed reaction then activates the second endonuclease reaction of Nt.BstNBI nickase [24] in the presence of a fluorescence-quenched DNA probe, which can produce a great amount of signal reporters and enable fluorescence based sensitive quantification of MGMT activity. In this strategy, fluorescence-based readouts permit the use of simple instrumentation, and target-mediated two endonuclease reactions can furnish the method desirable specific and sensitivity, respectively. Moreover, the developed technique can be implemented in homogeneous format with no need of washing and separation steps, thus affording improved assay robustness, simplicity and throughput in MGMT detection.
2. Experimental 2.1. Materials and reagents O6 -Methylguanine-DNA methyltransferase (human recombinant) was purchased from Cayman Chemical (Ann Arbor, MI, USA). Restriction endonucleases PvuII and Nt.BstNBI nickase, 10× NEBuffer 3 (1× NEBuffer 3 containing 50 mM Tris–HCl, 100 mM NaCl, 10 mM MgCl2 , 1 mM dithiothreitol, pH 7.9 at 25 ◦ C) were purchased from New England BioLabs (Ipswich, MA, USA). Bovine serum albumin (BSA) was provided by National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). All other chemicals were of analytical grade and obtained from Sinopharm Chemical Reagent Co. Ltd. (Beijing, China). All other solutions were prepared using ultrapure water, which was obtained through a Millipore Milli-Q water purification system (Billerica, MA, USA) and had an electric resistance >18.3 M. All oligonucleotides used in the experiments were synthesized from Takara Biotechnology Co. Ltd. (Dalian, China). Thermodynamic parameter and secondary structure of all oligonucleotides were calculated using bioinformatics software (http://bioin-fo.rpi.edu/application/). The sequences of the synthesized oligonucleotides are given in Table 1.
253
Table 1 Synthesized oligonucleotides (5 → 3 ) used in the experiments.a
Sequence 1
CGA CAC GAC ACC AXC TGC TTA GGT ACA GAC TCA AGC AGC TGG TGT CGT GTC G (X = O6-me-dG, O6-methylguanine)
Probe 2
CTT GAG TCT GT(-(CH2)6-FAM)A CCT(-(CH2)6-BHQ1)
Sequence 3
CGA CAC GAC ACC AGC TGC TTA GGT ACA GAC TCA AGC AGC TGG TGT CGT GTC G
a Sequence 1 is used as the substrate of MGMT with an O6 -methylated guanine at the recognition site of restriction endonuclease PvuII. Fluorescencequenched probe 2 is synthesized with black hole quencher-1 (BHQ1) quencher and 5-carboxyfluorescein (FAM) fluorophore. Sequence 3 used in the control experiments has the same sequence as Sequence 1 but it is not methylated. The italic typed parts of sequence 1 and 3 are both complementary to probe 2. The recognition sequence for PvuII is highlighted in red, and the recognition sequence for Nt.BstNBI nickase is highlighted in blue. Sequences in a single strand are complementary to form hairpin structure.
2.2. Activity assay of MGMT using the developed strategy The MGMT-mediated endonuclease reaction of PvuII was performed at 37 ◦ C for 1.5 h in 50 L reaction buffer (1× NEBuffer 3, BSA 0.01% (w v−1 )) containing 20 nM Sequence 1, 40 U mL−1 PvuII, and MGMT of a given concentration. Then, the reaction was terminated through heating at 65 ◦ C for 10 min. After that, 1 L Probe 2 (10 M) was added to the reaction mixture and heated to 95 ◦ C for 10 min followed by slowly cooling down to room temperature, to allow the hybridization of Probe 2 and the single-strand DNA product of PvuII-catalyzed endonuclease reaction. The solution was incubated with 10 L Nt.BstNBI nickase (200 U mL−1 ) at 55 ◦ C for 1 h after the final volume diluted into 100 L to initiate the second endonuclease reaction. The resulting solution was immediately subjected to fluorescence measurements. The fluorescence spectra were recorded at room temperature in a quartz cuvette on an F-7000 spectrofluorometer (Hitachi, Japan). The excitation wavelength was 494 nm and the emission wavelengths were in the range from 505 to 600 nm with both excitation and emission slits of 5 nm. 2.3. Gel electrophoresis analysis of MGMT To verify MGMT-mediated endonuclease reaction, samples were prepared by incubating 1 L 10 M hairpin-structured DNA Sequence 1 or Sequence 3 with 19 L reaction buffer which contained 200 U mL−1 PvuII in the absence or presence of 1 g mL−1 MGMT at 37 ◦ C for 90 min, then the reaction was terminated through heating at 65 ◦ C for 10 min. The resulting mixture was analyzed using gel electrophoresis in 4% (w w−1 ) agarose containing 1% (w w−1 ) ethidium bromide with voltage of 140 and a running time of 50 min. The gel was visualized using a ChemiDocTM XRS+ imaging system with Image LabTM (Bio-Rad, USA). 3. Results and discussion 3.1. Probe design and analytical principle The MGMT activity assay strategy relies on the target-mediated consecutive endonuclease reactions. As illustrated in Scheme 1, in the stem of a hairpin-structured DNA Sequence 1, it is designed to have a recognition site of the restriction endonuclease PvuII with a methyl adduct at the O6 -position of guanine (5 -CAO6-me G::CTG-3 ). In the loop of sequence 1, there is a half recognition site (5 GACTC-3 ) for Nt.BstNBI nickase, and the sequence of the loop is perfect complementary to an engineered DNA probe 2 which is labeled with 5-carboxy fluorescein (FAM) fluorophore and a black hole quencher-1 (BHQ1) quencher in adjacent nucleotides. In intact probe 2, the FAM fluorescence is quenched. In MGMT activity probing, MGMT reacts with the methylated substrate DNA sequence
254
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257
Scheme 1. Illustration of the MGMT activity assay strategy based on target-mediated consecutive endonuclease reactions. The behavior of MGMT removing methyl group from sequence 1 can activate the first, PvuII-catalyzed endonuclease reaction. This reaction releases a single DNA sequence from sequence 1 for fluorescence-quenched probe 2, and thus initiates another endonuclease reaction to activate an amplification approach for fluorescence signal.
1 and removes the methyl group from this substrate DNA. This reaction thus provides a specific recognition site for the restriction endonuclease PvuII (5 -CAGCTG-3 ), then, activating the first endonuclease reaction in the presence of PvuII and releasing the loop sequence from this substrate DNA. The perfect complement of the released loop sequence and the fluorescence-quenched DNA probe suddenly offers a full recognition site for Nt.BstNBI nickase to activate the second endonuclease reaction. Since the nicking site just locates between the FAM fluorophore and the BHQ1 quencher labeled on the DNA probe 2, probe 2 is split by Nt.BstNBI nickase into two short fragments that can be readily released from the loop sequence. This separates the fluorophore from the quencher, thus activates the fluorescence signal, and re-exposes the loop sequence to the remaining intact probes. In this way, the second endonuclease reaction can perform a cyclical DNA-cleaving operation via the repetition of annealing, cleaving, and releasing of the fluorescence reporters, therefore, furnishing an amplification approach for MGMT-mediated endonuclease reaction of PvuII. In contrast, when MGMT is absent, the methyl adduct at the recognition site of PvuII in sequence 1 will prevent PvuII from scissoring sequence 1 because of the high specificity of PvuII. This prevents the release of the loop sequence, thus inhibiting the second endonuclease reaction and the activation of the fluorescence signal. Because the fluorescence activation is achieved only with MGMT-mediated two consecutive endonuclease reactions, the proposed strategy readily provides a homogeneous fluorescence technique for activity screening of MGMT. Fig. 1 depicts typical fluorescence spectra of the developed strategy in the assay of MGMT. In the mixture, where 10 nM DNA substrate sequence 1 was incubated with 20 U mL−1 Nt.BstNBI nickase in the presence of 100 nM fluorescence-quenched DNA probe 2, a relatively weak fluorescence signal with a maximum intensity of ∼2300 at 522 nm was observed (curve a). For the mixture, of which sequence 1 was first incubated with 20 U mL−1 PvuII then with 20 U mL−1 Nt.BstNBI nickase in the presence of probe 2, no obvious enhancement of the fluorescence signal was obtained (curve b). In contrast, when the sequence 1 was first reacted with 80 ng mL−1 MGMT followed by the incubation of PvuII and Nt.BstNBI nickase, the fluorescence signal was increased with a maximum readout of ∼7300 (curve c). Such a greatly enhanced signal suggested MGMT was essential for fluorescence activation and it can effectively mediate the two consecutive endonuclease reactions to release activated fluorescence reporters. Furthermore, a control experiment was performed using non O6 -methylguanine labeled sequence 3 as the substrate. Because no methyl adduct at the recognition site of PvuII, the PvuII-catalyzed endonuclease
reaction can arise in the absence of MGMT, thus, activating the nickase-catalyzed endonuclease reaction. A similar strong fluorescence signal comparable with that obtained using the proposed strategy was observed in the control experiment (curve d). As the difference between the two mixtures (of which the signals were displayed in curve b and curve d) was only the DNA substrates, it revealed the methyl adduct in the DNA can immediately prevent PvuII-catalyzed endonuclease reaction. In other words, for methylated DNA substrate sequence 1 the two endonuclease reactions can only be specifically activated by MGMT. 3.2. Gel electrophoresis analysis of MGMT To further verify the mechanism of the proposed MGMT activity assay strategy, gel electrophoresis analysis was performed. Because PvuII can hydrolyze the DNA product of MGMT-catalyzed reaction into small segments, gel electrophoresis analysis could provide a straightforward evidence for the activation of PvuII and thus furnish an indirect evidence for MGMT-catalyzed transfer of methyl
Fig. 1. Typical fluorescence spectra obtained in MGMT activity assays: sequence 1, +Nt.BstNBI nickase, +probe 2 (a); sequence 1, +PvuII, +Nt.BstNBI nickase, +probe 2 (b); sequence 1, +MGMT, +PvuII, +Nt.BstNBI nickase, +probe 2 (c); sequence 3, +PvuII, +Nt.BstNBI nickase, +probe 2 (d). The concentrations of sequence 1, probe 2, sequence 3, PvuII, Nt.BstNBI nickase and MGMT are 10 nM, 100 nM, 10 nM, 20 U mL−1 , 20 U mL−1 , and 80 ng mL−1 , respectively.
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257
Fig. 2. Gel electrophoresis image obtained in MGMT analysis: lanes 1 and 6, DNA size marker; lane 2, sequence 1; lane 3, sequence 1, +PvuII; lane 4, sequence 1, +PvuII, +MGMT; lane 5, sequence 3, +PvuII. The concentrations of sequence 1, sequence 3, PvuII, and MGMT are 500 nM, 500 nM, 200 U mL−1 , and 1 g mL−1 , respectively.
group. Fig. 2 displays typical gel electrophoresis image of these DNA chimeras before and after degradation by restriction endonuclease PvuII. As shown in Fig. 2, it was observed that, after the methylated DNA substrate was treated with PvuII in the absence of MGMT, no obvious change can be detected for the substrate (lane 3) compared with the intact DNA substrate (lane 2). While treating the DNA substrate with both of MGMT and PvuII, a bright band with obviously changed migration shift was obtained on the gel (lane 4), which contrasted clearly with the intact DNA substrate (lane 2) and the substrate treated with only PvuII (lane 3). Compared with the bright band observed in the control experiment of which PvuII was treated with the non O6 -methylguanine labeled DNA (lane 5), it can be inferred that the bright band described above was actually arising from the product of the PvuII-catalyzed endonuclease reaction. Note that for the experiment of lane 3, the dissimilarity in the experiment of lane 4 was the absence of MGMT, and the dissimilarity in the experiment of lane 5 was the use of non-methylated DNA as the substrate. These findings revealed that with a methyl group added in the guanine of the recognition site for PvuII, the PvuII-catalyzed endonuclease reaction can only be activated by active MGMT via moving the methyl adduct from the DNA substrate. Therefore, the developed strategy exhibited a great potential in providing a highly specific platform for MGMT activity assay and related researches. 3.3. Time optimization of MGMT-mediated endonuclease reaction of PvuII MGMT-mediated endonuclease reaction of PvuII was the ratelimiting step of the developed strategy in MGMT assay. The fluorescence response of the MGMT assay may be greatly dependent upon the reaction time of the methylated DNA substrate with MGMT and PvuII. Fig. 3 depicts the dependency of fluorescence signal of the MGMT assay on the reaction time. With prolonged reaction time, the observed fluorescence intensity gradually increased and became almost leveled off at 1.5 h. Thus, the reaction time was set to 1.5 h throughout subsequent experiments. Generally, the proposed strategy showed a greatly improved analysis time in MGTMT activity assay compared with most of the classic methods as summarized in Table 2. Generally, these methods required a time more than 4 h.
255
Fig. 3. Dependency of peak fluorescence intensities on enzyme-catalyzed reaction time. The concentrations of sequence 1, probe 2, PvuII, Nt.BstNBI nickase and MGMT are 10 nM, 100 nM, 20 U mL−1 , 20 U mL−1 , and 80 ng mL−1 , respectively. Error bars are standard deviations across three repetitive experiments.
3.4. Quantitative nature of the developed strategy in MGMT activity assay The performance of the developed strategy for quantitative activity screening of MGMT was further investigated. Fig. 4A shows the typical fluorescence spectra of the developed strategy in response to MGMT at different concentrations under optimal reaction time. The fluorescence intensities increased gradually with increasing concentration of MGMT with a quite wide dynamic range from 1 to 120 ng mL−1 . A plot of the fluorescence peak intensities at 522 nm versus the MGMT concentrations revealed a linear correlation between the peak intensities and the MGMT concentrations in the range from 1 ng mL−1 to 25 ng mL−1 (Fig. 4B). In terms of the rule of three times deviation over the blank response, the detection limit was estimated to be as low as 0.5 ng mL−1 . Such a low detection limit was much better than most existing strategies for MGMT assay [10–12,22], suggesting a desirable high sensitivity of the developed strategy which might be furnished by the signal amplifying strategy of the second endonuclease reaction. Moreover, the strategy exhibited excellent reproducibility due to its homogeneous assay format with simple operations. Relative standard deviations (RSDs) of fluorescence intensity responses at 522 nm were 2.4%, 1.7%, 1.6%, and 2.1% in three repetitive assays of 1, 3.5, 10, and 25 ng mL−1 MGMT. Therefore, we might conclude that the developed fluorescence biosensor holds great potential for quantitative activity assay of MGMT with desirable sensitivity and reproducibility. Additionally, further experiments using fetal bovine serum as the reaction substrate were performed to demonstrate the specificity of the developed strategy under complex experiment conditions. Serum usually contains many biomolecules and some
Table 2 Analysis time summarized from the different existing methods. Method
Time of analysis
Radioactive methods [10–12,14] Liquid chromatography [16] Mass spectrometry [20] Fluorescence [21] Enzyme-linked immunosorbent assay [22]
>4 h >12 h >12 h 5–6 h ∼5 h
256
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257
small molecules, such as H2 O2 , dopamine, ethanol and CO, which may interfere with the activity assay of MGMT. As shown in Fig. 5, it could be found that when the same concentration of MGMT added, there was no significant difference between the fluorescent intensities of the samples containing fetal bovine serum as the substrate and that of the samples with NEBuffer 3 as the substrate. These results exhibited that even under a complex experiment conditions the developed sensing strategy still had desirable selectivity for active MGMT. 4. Conclusion A novel homogeneous fluorescence strategy was proposed for sensitive and specific activity screening of MGMT based on target-mediated consecutive endonuclease reactions. The first endonuclease reaction of PvuII furnished this technique with desirable specificity in MGMT assay, and the second endonuclease reaction of Nt.BstNBI nickase allowed efficient amplification to create an abundance of fluorescence signal reporters. The developed strategy was demonstrated to display desirable selectivity in MGMT activity assay and a low detection limit of 0.5 ng mL−1 in quantitative assay. Moreover, the homogeneous assay format using simplified instrumentation for fluorescence signal readouts made this strategy easily automated for quantitative analysis with desirable robustness. This strategy also can be used for sensitively and specifically activity screening of other alkyltransferases by simply altering the related alkylated bases. Therefore, it can be further extended for high-throughput alkyltransferases activity assays via using microplate-based assays and multicolor. In view of these advantages, the developed strategy was expected to provide an intrinsically robust, sensitive, and specific assay platform for convenient alkyltransferase activity analysis and related studies. Acknowledgements
Fig. 4. (A) Typical fluorescence spectral responses of the developed strategy to MGMT of varying concentrations. (B) Corresponding peak fluorescence intensity readings versus MGMT concentrations. Inset: Linear relationship between fluorescence intensity and MGMT concentration. Error bars are standard deviations across three repetitive experiments.
4000
Fluorescense (a.u.)
3800 3600 3400 3200 3000 2800 2600
a
b
c
Fig. 5. Bar plot of corresponding fluorescence responses at 522 nm of MGMT activity analysis in 10-fold-diluted fetal bovine serum (blue) and 1× NEBuffer 3 (red): containing 10 nM sequence 1, 100 nM probe 2, 20 U mL−1 PvuII, U mL−1 Nt.BstNBI nickase in the absence (a) or presence of 2 (b) or 5 (c) ng mL−1 MGMT. Error bars are standard deviations across three repetitive experiments. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
This work was supported by NSFC (21175040, 21025521 and 21205034), National Key Basic Research Program (2011CB911000), Doctoral Fund of Ministry of Education of China (New Teachers, 20120161120032), Hunan Provincial NSFC (13JJ4031). References [1] E.L. Loechler, C.L. Green, J.M. Essigmann, Proc. Natl. Acad. Sci. 81 (1984) 6271–6275. [2] R. Saffhill, G.P. Margison, P.J. O’Connor, Biochim. Biophys. Acta 823 (1985) 111–145. [3] A.E. Pegg, M.E. Dolan, R.C. Moschel, Prog. Nucleic Acids Res. Mol. Biol. 51 (1995) 167–223. [4] M. Xu-Welliver, A.E. Pegg, Carcinogenesis 23 (2002) 823–830. [5] S. Mitra, B. Kaina, Prog. Nucleic Acid Res. Mol. Biol. 44 (1993) 109–142. [6] G.P. Margison, M.F. Santibanez-Koref, Bioessays 24 (2002) 255–266. [7] A.E. Pegg, Mutat. Res. 462 (2000) 83–100. [8] S.L. Gerson, Nat. Rev. Cancer 4 (2004) 296–307. [9] M. D’Incalci, L. Citti, P. Taverna, C.V. Catapano, Cancer Treat. Rev. 15 (1988) 279–292. [10] K. Ishiguro, K. Shyam, P.G. Penketh, A.C. Sartorelli, Anal. Biochem. 383 (2008) 44–51. [11] K. Lim, Y.J. Kim, S. Kim, W.K. Paik, Anal. Biochem. 236 (1996) 284–289. [12] G.N. Major, E.J. Gardner, P.D. Lawley, Biochem. J. 277 (1991) 89–96. [13] B. Myrnes, K. Norstrand, K.E. Giercksky, C. Sjunneskog, H. Krokan, Carcinogenesis 5 (1984) 1061–1064. [14] E.A. Waldstein, E.H. Cao, R.B. Setlow, Anal. Biochem. 126 (1982) 268–272. [15] S.L. Gerson, K. Miller, N.A. Berger, J. Clin. Invest. 76 (1985) 2106–2114. [16] M. Mellai, V. Caldera, L. Annovazzi, A. Chiò, M. Lanotte, P. Cassoni, G. Finocchiaro, D. Schiffer, Cancer Genomics Proteomics 6 (2009) 219–228. [17] P. Parrella, A. la Torre, M. Copetti, V.M. Valori, R. Barbano, A. Notarangelo, M. Bisceglia, A.P. Gallo, T. Balsamo, M.L. Poeta, M. Carella, D. Catapano, S. Parisi, B. Dallapiccola, E. Maiello, V. D’Angelo, V.M. Fazio, J. Biomed. Biotechnol. 2009 (2009) 531692. [18] A.B. Håvik, P. Brandal, H. Honne, H.S. Dahlback, D. Scheie, M. Hektoen, T.R. Meling, E. Helseth, S. Heim, R.A. Lothe, G.E. Lind, J. Transl. Med. 10 (2012) 36.
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257 [19] D. Kotandeniya, D. Murphy, U. Seneviratne, R. Guza, A. Pegg, S. Kanugula, N. Tretyakova, Chem. Res. Toxicol. 24 (2011) 1966–1975. [20] S.K. Niturea, C.E. Doneanub, C.S. Velua, N.I. Baileya, K.S. Srivenugopal, Biochem. Biophys. Res. Commun. 337 (2005) 1176–1184. [21] E.L. Kreklau, M. Limp-Foster, N. Liu, Y. Xu, M.R. Kelley, L.C. Erickson, Nucleic Acids Res. 29 (2001) 2558–2566.
257
[22] P. Georgiadis, S. Kaila, P. Makedonopoulou, E. Fthenou, L. Chatzi, V. Pletsa, S.A. Kyrtopoulos, Cancer Epidemiol. Biomarkers Prev. 20 (2011) 82–90. [23] M.R. Rice, M.D. Koons, R.M. Blumenthal, Nucleic Acids Res. 27 (1999) 1032–1038. [24] R.D. Morgan, C. Calvet, M. Demeter, R. Agra, H. Kong, Biol. Chem. 381 (2000) 1123–1125.
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Target-mediated consecutive endonuclease reactions for specific and sensitive homogeneous fluorescence assay of O6 -methylguanine-DNA methyltransferase Dinh-Vu Le a,b , Dian-Ming Zhou a , Li-Juan Tang a,∗ , Jian-Hui Jiang a , Ru-Qin Yu a , Yu-Zhi Wang a,∗ a State Key Laboratory for Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082, PR China b Faculty of Chemical Engineering, Industrial University of Hochiminh City, Viet Nam
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• A homogeneous fluorescence strategy for O6 -methylguanine-DNA methyltransferase activity assay. • Based on target-mediated consecutive endonuclease reactions. • Simple, robust, highly sensitive and selective. • A potential convenient platform for alkyltransferase activity assay and related studies.
a r t i c l e
i n f o
Article history: Received 23 April 2013 Received in revised form 11 October 2013 Accepted 16 October 2013 Available online 22 October 2013 Keywords: O6 -Methylguanine DNA methyltransferase DNA-repair enzyme Alkyltransferase Restriction endonuclease Fluorescence
a b s t r a c t O6 -Methylguanine-DNA methyltransferase (MGMT) is one of the most important DNA-repair enzymes. Herein, a simple, sensitive and selective homogeneous fluorescence assay strategy is developed for the detection of MGMT on the basis of target-mediated two consecutive endonuclease reactions. The activity assay of MGMT is firstly accomplished using a hairpin-structured DNA substrate to offer a specific recognition site on the substrate DNA for restriction endonuclease PvuII, and thus to initiate the first endonuclease reaction. The product which activates the second endonuclease reaction allows an efficient amplification approach to create an abundance of fluorescence signal reporters. The first endonuclease reaction offers the method high specificity and the second one furnishes the assay improved sensitivity. The results reveal that the MGMT assay strategy shows dynamic responses in the concentration range from 1 to 120 ng mL−1 with a detection limit of 0.5 ng mL−1 . By simply altering the alkylated bases, this strategy can also be extended for the detection of other alkyltransferases. Therefore, the developed strategy might provide an intrinsically convenient, sensitive and specific platform for alkyltransferase activate assay and related biochemical studies due to its label-free, homogeneous, and fluorescence-based detection format. © 2013 Elsevier B.V. All rights reserved.
1. Introduction
∗ Corresponding authors. Tel.: +86 731 88821903; fax: +86 731 88821848. E-mail addresses: [email protected] (L.-J. Tang), [email protected] (Y.-Z. Wang). 0003-2670/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.aca.2013.10.036
Electrophilic alkylation caused by mutagens in the environment, in food, or endogenous metabolic products is often highly toxic, due to its ability to alkylate the bases of DNA and result in miscoding of DNA strand. Compared with other DNA methyl adducts, O6 -methylguanine (O6 MeG) is the most critical lesion
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257
for its pre-mutagenic and pre-toxic [1,2]. O6 -Methylguanine DNA methyltransferase (MGMT) is a DNA-repair enzyme which repairs O6 -alkylation adducts in a one-step alkyl transfer reaction by removing alkyl group from O6 -guanine in DNA [3–5]. The activity screening of MGMT has emerged as an essential step in investigating O6 MeG induced cell death, chromosomal aberrations, mutations and cancer [6–8]. Radioactive methods play an important role in MGMT assay by using 3 H labeled alkyl adduct in DNA as the substrate [9,10]. However, extensive multiple-step sample treatments (e.g. hydrolysis, extractions, concentration, etc.), which take tens of minutes or even hours, are normally required by these methods [11–14]. Timeconsuming separations are also needed to clean up the matrixes by using separation techniques such as liquid chromatography and capillary electrophoresis [15,16]. Although detection of MGMT can also be carried out by polymerase chain reaction (PCR) [17,18], mass spectrometry (MS) [19,20], fluorescence [21], or enzymelinked immunosorbent assay (ELISA) [22], these methods may still suffer from some drawbacks such as limited sensitivity, timeconsuming analysis, requirements for expensive equipments, or multistep washing and separation. In this context, the development of sensitive and selective methods for convenient MGMT activity monitoring is still a topic of intensive interest in bioanalytical chemistry. In the present study, a homogeneous fluorescence assay strategy is developed for activity screening of MGMT on the basis of targetmediated consecutive endonuclease reactions. In this strategy, active MGMT can offer a specific recognition site on the substrate DNA for restriction endonuclease PvuII [23] and thus initiates the first endonuclease reaction. The product of PvuII-catalyzed reaction then activates the second endonuclease reaction of Nt.BstNBI nickase [24] in the presence of a fluorescence-quenched DNA probe, which can produce a great amount of signal reporters and enable fluorescence based sensitive quantification of MGMT activity. In this strategy, fluorescence-based readouts permit the use of simple instrumentation, and target-mediated two endonuclease reactions can furnish the method desirable specific and sensitivity, respectively. Moreover, the developed technique can be implemented in homogeneous format with no need of washing and separation steps, thus affording improved assay robustness, simplicity and throughput in MGMT detection.
2. Experimental 2.1. Materials and reagents O6 -Methylguanine-DNA methyltransferase (human recombinant) was purchased from Cayman Chemical (Ann Arbor, MI, USA). Restriction endonucleases PvuII and Nt.BstNBI nickase, 10× NEBuffer 3 (1× NEBuffer 3 containing 50 mM Tris–HCl, 100 mM NaCl, 10 mM MgCl2 , 1 mM dithiothreitol, pH 7.9 at 25 ◦ C) were purchased from New England BioLabs (Ipswich, MA, USA). Bovine serum albumin (BSA) was provided by National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). All other chemicals were of analytical grade and obtained from Sinopharm Chemical Reagent Co. Ltd. (Beijing, China). All other solutions were prepared using ultrapure water, which was obtained through a Millipore Milli-Q water purification system (Billerica, MA, USA) and had an electric resistance >18.3 M. All oligonucleotides used in the experiments were synthesized from Takara Biotechnology Co. Ltd. (Dalian, China). Thermodynamic parameter and secondary structure of all oligonucleotides were calculated using bioinformatics software (http://bioin-fo.rpi.edu/application/). The sequences of the synthesized oligonucleotides are given in Table 1.
253
Table 1 Synthesized oligonucleotides (5 → 3 ) used in the experiments.a
Sequence 1
CGA CAC GAC ACC AXC TGC TTA GGT ACA GAC TCA AGC AGC TGG TGT CGT GTC G (X = O6-me-dG, O6-methylguanine)
Probe 2
CTT GAG TCT GT(-(CH2)6-FAM)A CCT(-(CH2)6-BHQ1)
Sequence 3
CGA CAC GAC ACC AGC TGC TTA GGT ACA GAC TCA AGC AGC TGG TGT CGT GTC G
a Sequence 1 is used as the substrate of MGMT with an O6 -methylated guanine at the recognition site of restriction endonuclease PvuII. Fluorescencequenched probe 2 is synthesized with black hole quencher-1 (BHQ1) quencher and 5-carboxyfluorescein (FAM) fluorophore. Sequence 3 used in the control experiments has the same sequence as Sequence 1 but it is not methylated. The italic typed parts of sequence 1 and 3 are both complementary to probe 2. The recognition sequence for PvuII is highlighted in red, and the recognition sequence for Nt.BstNBI nickase is highlighted in blue. Sequences in a single strand are complementary to form hairpin structure.
2.2. Activity assay of MGMT using the developed strategy The MGMT-mediated endonuclease reaction of PvuII was performed at 37 ◦ C for 1.5 h in 50 L reaction buffer (1× NEBuffer 3, BSA 0.01% (w v−1 )) containing 20 nM Sequence 1, 40 U mL−1 PvuII, and MGMT of a given concentration. Then, the reaction was terminated through heating at 65 ◦ C for 10 min. After that, 1 L Probe 2 (10 M) was added to the reaction mixture and heated to 95 ◦ C for 10 min followed by slowly cooling down to room temperature, to allow the hybridization of Probe 2 and the single-strand DNA product of PvuII-catalyzed endonuclease reaction. The solution was incubated with 10 L Nt.BstNBI nickase (200 U mL−1 ) at 55 ◦ C for 1 h after the final volume diluted into 100 L to initiate the second endonuclease reaction. The resulting solution was immediately subjected to fluorescence measurements. The fluorescence spectra were recorded at room temperature in a quartz cuvette on an F-7000 spectrofluorometer (Hitachi, Japan). The excitation wavelength was 494 nm and the emission wavelengths were in the range from 505 to 600 nm with both excitation and emission slits of 5 nm. 2.3. Gel electrophoresis analysis of MGMT To verify MGMT-mediated endonuclease reaction, samples were prepared by incubating 1 L 10 M hairpin-structured DNA Sequence 1 or Sequence 3 with 19 L reaction buffer which contained 200 U mL−1 PvuII in the absence or presence of 1 g mL−1 MGMT at 37 ◦ C for 90 min, then the reaction was terminated through heating at 65 ◦ C for 10 min. The resulting mixture was analyzed using gel electrophoresis in 4% (w w−1 ) agarose containing 1% (w w−1 ) ethidium bromide with voltage of 140 and a running time of 50 min. The gel was visualized using a ChemiDocTM XRS+ imaging system with Image LabTM (Bio-Rad, USA). 3. Results and discussion 3.1. Probe design and analytical principle The MGMT activity assay strategy relies on the target-mediated consecutive endonuclease reactions. As illustrated in Scheme 1, in the stem of a hairpin-structured DNA Sequence 1, it is designed to have a recognition site of the restriction endonuclease PvuII with a methyl adduct at the O6 -position of guanine (5 -CAO6-me G::CTG-3 ). In the loop of sequence 1, there is a half recognition site (5 GACTC-3 ) for Nt.BstNBI nickase, and the sequence of the loop is perfect complementary to an engineered DNA probe 2 which is labeled with 5-carboxy fluorescein (FAM) fluorophore and a black hole quencher-1 (BHQ1) quencher in adjacent nucleotides. In intact probe 2, the FAM fluorescence is quenched. In MGMT activity probing, MGMT reacts with the methylated substrate DNA sequence
254
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257
Scheme 1. Illustration of the MGMT activity assay strategy based on target-mediated consecutive endonuclease reactions. The behavior of MGMT removing methyl group from sequence 1 can activate the first, PvuII-catalyzed endonuclease reaction. This reaction releases a single DNA sequence from sequence 1 for fluorescence-quenched probe 2, and thus initiates another endonuclease reaction to activate an amplification approach for fluorescence signal.
1 and removes the methyl group from this substrate DNA. This reaction thus provides a specific recognition site for the restriction endonuclease PvuII (5 -CAGCTG-3 ), then, activating the first endonuclease reaction in the presence of PvuII and releasing the loop sequence from this substrate DNA. The perfect complement of the released loop sequence and the fluorescence-quenched DNA probe suddenly offers a full recognition site for Nt.BstNBI nickase to activate the second endonuclease reaction. Since the nicking site just locates between the FAM fluorophore and the BHQ1 quencher labeled on the DNA probe 2, probe 2 is split by Nt.BstNBI nickase into two short fragments that can be readily released from the loop sequence. This separates the fluorophore from the quencher, thus activates the fluorescence signal, and re-exposes the loop sequence to the remaining intact probes. In this way, the second endonuclease reaction can perform a cyclical DNA-cleaving operation via the repetition of annealing, cleaving, and releasing of the fluorescence reporters, therefore, furnishing an amplification approach for MGMT-mediated endonuclease reaction of PvuII. In contrast, when MGMT is absent, the methyl adduct at the recognition site of PvuII in sequence 1 will prevent PvuII from scissoring sequence 1 because of the high specificity of PvuII. This prevents the release of the loop sequence, thus inhibiting the second endonuclease reaction and the activation of the fluorescence signal. Because the fluorescence activation is achieved only with MGMT-mediated two consecutive endonuclease reactions, the proposed strategy readily provides a homogeneous fluorescence technique for activity screening of MGMT. Fig. 1 depicts typical fluorescence spectra of the developed strategy in the assay of MGMT. In the mixture, where 10 nM DNA substrate sequence 1 was incubated with 20 U mL−1 Nt.BstNBI nickase in the presence of 100 nM fluorescence-quenched DNA probe 2, a relatively weak fluorescence signal with a maximum intensity of ∼2300 at 522 nm was observed (curve a). For the mixture, of which sequence 1 was first incubated with 20 U mL−1 PvuII then with 20 U mL−1 Nt.BstNBI nickase in the presence of probe 2, no obvious enhancement of the fluorescence signal was obtained (curve b). In contrast, when the sequence 1 was first reacted with 80 ng mL−1 MGMT followed by the incubation of PvuII and Nt.BstNBI nickase, the fluorescence signal was increased with a maximum readout of ∼7300 (curve c). Such a greatly enhanced signal suggested MGMT was essential for fluorescence activation and it can effectively mediate the two consecutive endonuclease reactions to release activated fluorescence reporters. Furthermore, a control experiment was performed using non O6 -methylguanine labeled sequence 3 as the substrate. Because no methyl adduct at the recognition site of PvuII, the PvuII-catalyzed endonuclease
reaction can arise in the absence of MGMT, thus, activating the nickase-catalyzed endonuclease reaction. A similar strong fluorescence signal comparable with that obtained using the proposed strategy was observed in the control experiment (curve d). As the difference between the two mixtures (of which the signals were displayed in curve b and curve d) was only the DNA substrates, it revealed the methyl adduct in the DNA can immediately prevent PvuII-catalyzed endonuclease reaction. In other words, for methylated DNA substrate sequence 1 the two endonuclease reactions can only be specifically activated by MGMT. 3.2. Gel electrophoresis analysis of MGMT To further verify the mechanism of the proposed MGMT activity assay strategy, gel electrophoresis analysis was performed. Because PvuII can hydrolyze the DNA product of MGMT-catalyzed reaction into small segments, gel electrophoresis analysis could provide a straightforward evidence for the activation of PvuII and thus furnish an indirect evidence for MGMT-catalyzed transfer of methyl
Fig. 1. Typical fluorescence spectra obtained in MGMT activity assays: sequence 1, +Nt.BstNBI nickase, +probe 2 (a); sequence 1, +PvuII, +Nt.BstNBI nickase, +probe 2 (b); sequence 1, +MGMT, +PvuII, +Nt.BstNBI nickase, +probe 2 (c); sequence 3, +PvuII, +Nt.BstNBI nickase, +probe 2 (d). The concentrations of sequence 1, probe 2, sequence 3, PvuII, Nt.BstNBI nickase and MGMT are 10 nM, 100 nM, 10 nM, 20 U mL−1 , 20 U mL−1 , and 80 ng mL−1 , respectively.
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257
Fig. 2. Gel electrophoresis image obtained in MGMT analysis: lanes 1 and 6, DNA size marker; lane 2, sequence 1; lane 3, sequence 1, +PvuII; lane 4, sequence 1, +PvuII, +MGMT; lane 5, sequence 3, +PvuII. The concentrations of sequence 1, sequence 3, PvuII, and MGMT are 500 nM, 500 nM, 200 U mL−1 , and 1 g mL−1 , respectively.
group. Fig. 2 displays typical gel electrophoresis image of these DNA chimeras before and after degradation by restriction endonuclease PvuII. As shown in Fig. 2, it was observed that, after the methylated DNA substrate was treated with PvuII in the absence of MGMT, no obvious change can be detected for the substrate (lane 3) compared with the intact DNA substrate (lane 2). While treating the DNA substrate with both of MGMT and PvuII, a bright band with obviously changed migration shift was obtained on the gel (lane 4), which contrasted clearly with the intact DNA substrate (lane 2) and the substrate treated with only PvuII (lane 3). Compared with the bright band observed in the control experiment of which PvuII was treated with the non O6 -methylguanine labeled DNA (lane 5), it can be inferred that the bright band described above was actually arising from the product of the PvuII-catalyzed endonuclease reaction. Note that for the experiment of lane 3, the dissimilarity in the experiment of lane 4 was the absence of MGMT, and the dissimilarity in the experiment of lane 5 was the use of non-methylated DNA as the substrate. These findings revealed that with a methyl group added in the guanine of the recognition site for PvuII, the PvuII-catalyzed endonuclease reaction can only be activated by active MGMT via moving the methyl adduct from the DNA substrate. Therefore, the developed strategy exhibited a great potential in providing a highly specific platform for MGMT activity assay and related researches. 3.3. Time optimization of MGMT-mediated endonuclease reaction of PvuII MGMT-mediated endonuclease reaction of PvuII was the ratelimiting step of the developed strategy in MGMT assay. The fluorescence response of the MGMT assay may be greatly dependent upon the reaction time of the methylated DNA substrate with MGMT and PvuII. Fig. 3 depicts the dependency of fluorescence signal of the MGMT assay on the reaction time. With prolonged reaction time, the observed fluorescence intensity gradually increased and became almost leveled off at 1.5 h. Thus, the reaction time was set to 1.5 h throughout subsequent experiments. Generally, the proposed strategy showed a greatly improved analysis time in MGTMT activity assay compared with most of the classic methods as summarized in Table 2. Generally, these methods required a time more than 4 h.
255
Fig. 3. Dependency of peak fluorescence intensities on enzyme-catalyzed reaction time. The concentrations of sequence 1, probe 2, PvuII, Nt.BstNBI nickase and MGMT are 10 nM, 100 nM, 20 U mL−1 , 20 U mL−1 , and 80 ng mL−1 , respectively. Error bars are standard deviations across three repetitive experiments.
3.4. Quantitative nature of the developed strategy in MGMT activity assay The performance of the developed strategy for quantitative activity screening of MGMT was further investigated. Fig. 4A shows the typical fluorescence spectra of the developed strategy in response to MGMT at different concentrations under optimal reaction time. The fluorescence intensities increased gradually with increasing concentration of MGMT with a quite wide dynamic range from 1 to 120 ng mL−1 . A plot of the fluorescence peak intensities at 522 nm versus the MGMT concentrations revealed a linear correlation between the peak intensities and the MGMT concentrations in the range from 1 ng mL−1 to 25 ng mL−1 (Fig. 4B). In terms of the rule of three times deviation over the blank response, the detection limit was estimated to be as low as 0.5 ng mL−1 . Such a low detection limit was much better than most existing strategies for MGMT assay [10–12,22], suggesting a desirable high sensitivity of the developed strategy which might be furnished by the signal amplifying strategy of the second endonuclease reaction. Moreover, the strategy exhibited excellent reproducibility due to its homogeneous assay format with simple operations. Relative standard deviations (RSDs) of fluorescence intensity responses at 522 nm were 2.4%, 1.7%, 1.6%, and 2.1% in three repetitive assays of 1, 3.5, 10, and 25 ng mL−1 MGMT. Therefore, we might conclude that the developed fluorescence biosensor holds great potential for quantitative activity assay of MGMT with desirable sensitivity and reproducibility. Additionally, further experiments using fetal bovine serum as the reaction substrate were performed to demonstrate the specificity of the developed strategy under complex experiment conditions. Serum usually contains many biomolecules and some
Table 2 Analysis time summarized from the different existing methods. Method
Time of analysis
Radioactive methods [10–12,14] Liquid chromatography [16] Mass spectrometry [20] Fluorescence [21] Enzyme-linked immunosorbent assay [22]
>4 h >12 h >12 h 5–6 h ∼5 h
256
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257
small molecules, such as H2 O2 , dopamine, ethanol and CO, which may interfere with the activity assay of MGMT. As shown in Fig. 5, it could be found that when the same concentration of MGMT added, there was no significant difference between the fluorescent intensities of the samples containing fetal bovine serum as the substrate and that of the samples with NEBuffer 3 as the substrate. These results exhibited that even under a complex experiment conditions the developed sensing strategy still had desirable selectivity for active MGMT. 4. Conclusion A novel homogeneous fluorescence strategy was proposed for sensitive and specific activity screening of MGMT based on target-mediated consecutive endonuclease reactions. The first endonuclease reaction of PvuII furnished this technique with desirable specificity in MGMT assay, and the second endonuclease reaction of Nt.BstNBI nickase allowed efficient amplification to create an abundance of fluorescence signal reporters. The developed strategy was demonstrated to display desirable selectivity in MGMT activity assay and a low detection limit of 0.5 ng mL−1 in quantitative assay. Moreover, the homogeneous assay format using simplified instrumentation for fluorescence signal readouts made this strategy easily automated for quantitative analysis with desirable robustness. This strategy also can be used for sensitively and specifically activity screening of other alkyltransferases by simply altering the related alkylated bases. Therefore, it can be further extended for high-throughput alkyltransferases activity assays via using microplate-based assays and multicolor. In view of these advantages, the developed strategy was expected to provide an intrinsically robust, sensitive, and specific assay platform for convenient alkyltransferase activity analysis and related studies. Acknowledgements
Fig. 4. (A) Typical fluorescence spectral responses of the developed strategy to MGMT of varying concentrations. (B) Corresponding peak fluorescence intensity readings versus MGMT concentrations. Inset: Linear relationship between fluorescence intensity and MGMT concentration. Error bars are standard deviations across three repetitive experiments.
4000
Fluorescense (a.u.)
3800 3600 3400 3200 3000 2800 2600
a
b
c
Fig. 5. Bar plot of corresponding fluorescence responses at 522 nm of MGMT activity analysis in 10-fold-diluted fetal bovine serum (blue) and 1× NEBuffer 3 (red): containing 10 nM sequence 1, 100 nM probe 2, 20 U mL−1 PvuII, U mL−1 Nt.BstNBI nickase in the absence (a) or presence of 2 (b) or 5 (c) ng mL−1 MGMT. Error bars are standard deviations across three repetitive experiments. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
This work was supported by NSFC (21175040, 21025521 and 21205034), National Key Basic Research Program (2011CB911000), Doctoral Fund of Ministry of Education of China (New Teachers, 20120161120032), Hunan Provincial NSFC (13JJ4031). References [1] E.L. Loechler, C.L. Green, J.M. Essigmann, Proc. Natl. Acad. Sci. 81 (1984) 6271–6275. [2] R. Saffhill, G.P. Margison, P.J. O’Connor, Biochim. Biophys. Acta 823 (1985) 111–145. [3] A.E. Pegg, M.E. Dolan, R.C. Moschel, Prog. Nucleic Acids Res. Mol. Biol. 51 (1995) 167–223. [4] M. Xu-Welliver, A.E. Pegg, Carcinogenesis 23 (2002) 823–830. [5] S. Mitra, B. Kaina, Prog. Nucleic Acid Res. Mol. Biol. 44 (1993) 109–142. [6] G.P. Margison, M.F. Santibanez-Koref, Bioessays 24 (2002) 255–266. [7] A.E. Pegg, Mutat. Res. 462 (2000) 83–100. [8] S.L. Gerson, Nat. Rev. Cancer 4 (2004) 296–307. [9] M. D’Incalci, L. Citti, P. Taverna, C.V. Catapano, Cancer Treat. Rev. 15 (1988) 279–292. [10] K. Ishiguro, K. Shyam, P.G. Penketh, A.C. Sartorelli, Anal. Biochem. 383 (2008) 44–51. [11] K. Lim, Y.J. Kim, S. Kim, W.K. Paik, Anal. Biochem. 236 (1996) 284–289. [12] G.N. Major, E.J. Gardner, P.D. Lawley, Biochem. J. 277 (1991) 89–96. [13] B. Myrnes, K. Norstrand, K.E. Giercksky, C. Sjunneskog, H. Krokan, Carcinogenesis 5 (1984) 1061–1064. [14] E.A. Waldstein, E.H. Cao, R.B. Setlow, Anal. Biochem. 126 (1982) 268–272. [15] S.L. Gerson, K. Miller, N.A. Berger, J. Clin. Invest. 76 (1985) 2106–2114. [16] M. Mellai, V. Caldera, L. Annovazzi, A. Chiò, M. Lanotte, P. Cassoni, G. Finocchiaro, D. Schiffer, Cancer Genomics Proteomics 6 (2009) 219–228. [17] P. Parrella, A. la Torre, M. Copetti, V.M. Valori, R. Barbano, A. Notarangelo, M. Bisceglia, A.P. Gallo, T. Balsamo, M.L. Poeta, M. Carella, D. Catapano, S. Parisi, B. Dallapiccola, E. Maiello, V. D’Angelo, V.M. Fazio, J. Biomed. Biotechnol. 2009 (2009) 531692. [18] A.B. Håvik, P. Brandal, H. Honne, H.S. Dahlback, D. Scheie, M. Hektoen, T.R. Meling, E. Helseth, S. Heim, R.A. Lothe, G.E. Lind, J. Transl. Med. 10 (2012) 36.
D.-V. Le et al. / Analytica Chimica Acta 804 (2013) 252–257 [19] D. Kotandeniya, D. Murphy, U. Seneviratne, R. Guza, A. Pegg, S. Kanugula, N. Tretyakova, Chem. Res. Toxicol. 24 (2011) 1966–1975. [20] S.K. Niturea, C.E. Doneanub, C.S. Velua, N.I. Baileya, K.S. Srivenugopal, Biochem. Biophys. Res. Commun. 337 (2005) 1176–1184. [21] E.L. Kreklau, M. Limp-Foster, N. Liu, Y. Xu, M.R. Kelley, L.C. Erickson, Nucleic Acids Res. 29 (2001) 2558–2566.
257
[22] P. Georgiadis, S. Kaila, P. Makedonopoulou, E. Fthenou, L. Chatzi, V. Pletsa, S.A. Kyrtopoulos, Cancer Epidemiol. Biomarkers Prev. 20 (2011) 82–90. [23] M.R. Rice, M.D. Koons, R.M. Blumenthal, Nucleic Acids Res. 27 (1999) 1032–1038. [24] R.D. Morgan, C. Calvet, M. Demeter, R. Agra, H. Kong, Biol. Chem. 381 (2000) 1123–1125.
