Hairpin Fluorescence DNA Probe for Real-Time ... - ACS Publications

Real-time monitoring of DNA methyltransferase activity using a hemimethylated smart probe. Shunxin Jin , Haisheng Liu , Kun Xia , Changbei Ma , Hailun...
0 downloads 0 Views 327KB Size
Anal. Chem. 2007, 79, 1050-1056

Hairpin Fluorescence DNA Probe for Real-Time Monitoring of DNA Methylation Jun Li, Hongfei Yan, Kemin Wang,* Weihong Tan, and Xingwang Zhou

Biomedical Engineering Center, State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry & Chemical Engineering, Hunan University, Changsha 410082, P. R. China

DNA methylation catalyzed by methylase plays an important role in many biological events. However, traditional methods of methylase activity analysis by gel electrophoresis were laborious and discontinuous. In this paper, we report a new strategy to study methylase activity using fluorescent probes coupled with enzyme-linkage reactions. A hairpin DNA probe is prepared with a fluorophore and a quencher linked at the 5′- and 3′-terminus of the probe. A disturbance of the stem sequence by DNA methylation would cause the separation of the fluorophore and the quencher, resulting in the restoration of the fluorescence. We used DNA adenine methylation (Dam) methyltransferase (MTase) and Dpn I endonuclease, both having a 5′-G-A-T-C-3′ recognition sequence. Dam MTase catalyzed the methylation of the sequence of 5′-GATC-3′, and Dpn I cut the sequence of 5′-G-Am-T-C-3′. The fluorescence of the hairpin probe was restored when it was cleaved by Dpn I endonuclease during the course of methylation. Unlike traditional methods, this assay was done in real time and could be used to monitor the dynamic process of methylation. Our method is easy, simple, and nonradioactive, yet as efficient as gel electrophoresis in detecting the activity of methylase. It also had the potential to screen suitable inhibitor drugs for Dam methylase. DNA methylation, a common gene protection approach, plays an important role in both prokaryotes and eukaryotes. This process is carried out by DNA methyltransferases (MTases), which specifically recognize the short palindromic sequences and catalyze the transfer of a methyl group from S-adenosyl-Lmethiolnine (SAM) to the target adenine or cytosine.1-4 The traditional methods for quantifying DNA MTases include the incorporation of radioactivity from [methyl-3H]-SAM to adenine or cytosine in DNA substrate, the digestion of DNA by methylation-sensitive enzymes, or the separation of methylation nucleotides by high-performance liquid chromatography.5-8 Unfortunately, most of these methods are discontinuous, laborious, and require radiolabeled substrates. It is therefore necessary to * To whom correspondence should be addressed. E-mail: [email protected]. Phone and fax: 86-731-8821566. (1) Bestor, T. H.; Verdine, G. L. Curr. Opin. Cell Biol. 1994, 6, 380-389. (2) Robertson, K. D.; Wolffe, A. P. Nat. Rev.: Genet. 2000, 1, 11-19. (3) Cheng, X.; Roberts, R. J. Nucleic Acids Res. 2001, 29, 3784-3795. (4) Jeltsch, A. Chem. Biol. Chem. 2002, 3, 274-293. (5) Messer, W.; Noyer-Weidner, M. Cell 1988, 54, 735-737.

1050 Analytical Chemistry, Vol. 79, No. 3, February 1, 2007

develop continuous, convenient, nonradioactive methods to monitor the process of DNA methylation and to detect the activity of MTase. In the last several years, the hairpin fluorescence molecular beacon (MB)-based DNA probe has been widely used in functional research dealing with enzymes, proteins, and drugs.9,10 Usually, the probe possesses a stem-loop structure, and a fluorophore and a quencher are linked to the two ends of the stem. The stem holds the two moieties in proximity to each other, causing the fluorescence to be quenched by fluorescence resonance energy transfer (FRET). The probes have also been used to provide realtime monitoring for DNA cleavage, ligation, transcription, and phosphorylation in homogeneous solutions.11-14 However, they could not be directly used in methylation because the recognition site of methylase is double-stranded DNA (dsDNA).3 If a complementary DNA probe forms a hybrid with an MB loop, the fluorescence will be restored and no fluorescence change will be observed after DNA methylation. Recently, a novel hairpin DNA probe was developed to investigate the continuous cleavage of DNA with antibiotics.15,16 This probe is similar to an MB, but the loop is so short that it cannot be used to recognize oligos. It acts like a linkage, thereby causing a more efficient quenching of this probe’s fluorophore than is produced when an MB is used. The cleavage site for the antibiotic is designed in the stem portion and will be broken by the cleavage of the nuclease or drugs to yield a new ssDNA and a 5-7 base dsDNA connecting the fluorophore and the quencher.15 The newly formed dsDNA will separate into two short DNA fragments at 37 °C, consequently resulting in spontaneous fluorescence restoration. This is the basis (6) Bergerat, A.; Guschlbauer, W.; Fazakerley, G. V. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 6394-6397. (7) Boye, E.; Marinus, M. G.; Lobner-Olesen, A. J. Bacteriol. 1992, 174, 16821685. (8) Thielking, V.; Du Bois, S.; Eritja, R.; Guschlbauer, W. Biol. Chem. 1997, 378, 407-415. (9) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303-308. (10) Tan, W.; Wang, K.; Drake, T. J. Curr. Opin. Chem. Biol. 2004, 8, 547-553. Fang, X.; Liu, X.; Schuster. S.; Tan, W. J. Am. Chem. Soc. 1999, 121, 29212922. (11) Li, J. J.; Geyer, R.; Tan, W. Nucleic Acids Res. 2000, 28, e52. (12) Tang, Z.; Wang, K.; Tan, W.; Li, J.; Liu, L.; Guo, Q.; Meng, X.; Ma, C.; Huang, S. Nucleic Acids Res. 2003, 31, e148. (13) Marras, S.; Gold, B.; Kramer, F.; Smith, I.; Tyagi, S. Nucleic Acids Res. 2004, 32, e72. (14) Tang, Z.; Wang, K.; Tan, W.; Ma, C.; Li, J.; Liu, L.; Guo, Q.; Meng, X. Nucleic Acids Res. 2005, 33, e97. (15) Biggins, J.; Prudent, J.; Marshall, D.; Ruppen, M.; Thorson, J. S. Proc. Natl Acad. Sci. U.S.A. 2000, 97, 13537-13542. (16) Biggins, J.; Onwueme, K.; Thorson, J. Science 2003, 301, 1537-1541. 10.1021/ac061694i CCC: $37.00

© 2007 American Chemical Society Published on Web 01/06/2007

Figure 1. Schematic diagram of the strategy of real-time monitoring of methylation process by a hairpin probe. (A) The sequence of the designed hairpin probe labeled TAMRA and DABCYL; the probe had low fluoresce in the hairpin conformation. (B) The Dam MTase methylated the hairpin probe at the recognition site, yielding the methylated DNA probe. The methylated probe substrate was recognized and cleaved by Dpn I endonuclease, restoring the fluorescence of TAMRA.

for quantitative analysis of the cleavage reaction. This kind of hairpin DNA probe provides an opportunity to monitor DNA methylation in real time by recording the fluorescence signal, but only if the probe can be methylated and cleaved. DNA adenine methylation (Dam) MTase is an enzyme that methylates the N6-adenine in the symmetric tetranucleotide 5′G-A-T-C-3′ and regulates a number of bacterial functions including DNA replication, segregation of chromosomal DNA, mismatch repair, transposition, and transcription of a certain gene.17,18 It has also been found that Dpn I endonuclease can only cut the sequence of 5′-G-Am-T-C-3′ when the internal adenine is methylated.19 Therefore, it is possible to monitor the adenine methylation process by embedding the 5′-G-A-T-C-3′ in a hairpin probe and designing a suitable enzyme-linkage reaction. In this paper, a 38-base-long hairpin probe was designed to monitor the methylation process of adenine in a dsDNA catalyzed by Dam MTase. The stem of the designed probe contained a sequence of 5′-G-A-T-C-3′, which was specifically recognized by Dam MTase and Dpn I endonuclease. Fluorophore TAMRA and quencher DABCYL were chosen to link at the 5′- and 3′-terminus, respectively. The process was observed by monitoring the fluorescence signal of the hairpin probe. The activity of DAM MTase was measured, and the influence of certain drugs on the activity of Dam MTase was studied. (17) Barras, F.; Marinus, M. G. Trends Genet. 1989, 5, 139-143. (18) Garcia-Del Portillo, F.; Pucciarelli, M. G.; Casadesus, J. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 11578-11583.

EXPERIMENTAL SECTION Reagents and Materials. The hairpin probe 5′-(TAMRA)-CTT AAG GAT CCC GCT TCT TTT GAA GCG GGA TCC TTA AG(DABCYL)-3′, oligo 1 5′-(TAMRA)-CTT AAG GAT CCC GCT TC3′, oligo 2 5′-GAA GCG GGA TCC TTA AG-(DABCYL)-3′ was synthesized by Takara Biotechnology Co. Ltd. (Dalian, China). The Dam MTase (Escherichia coli), Dpn I endonuclease, SAM, and the corresponding buffer solution were purchased from New England Biolabs Inc. Acrylamide and N,N,N′,N′-tetramethylethylenediamine were obtained from Sigma-Aldrich. Co (St. Louis, MO). Other chemicals were of analytical grade and were used without further purification. Deionized water was obtained through the Nanopure Infinity ultrapure water system (Barnstead/Thermolyne Corp., Dubuque, IA). Fluorescence Monitoring Procedure. All fluorescence measurements were carried out on a Perkin-Elmer LS-55 spectrofluorometer with temperature controlled at 37 °C by an aqueous thermostat (Amersham). The fluorescence intensity of all samples were analyzed via a time base scan (λex ) 558 nm, λem ) 580 nm). Slit widths for the excitation and emission were set at 5 and 10 nm, respectively. The spectrum was obtained with a 521-nm excitation. The reaction mixtures (200 µL) consisted of a 200 nM hairpin probe, 1× buffer (10 mM Tis-HCl pH 7.5, 10 mM MgCl2, 50 mM NaCl, 1 mM DTT), 160 µM SAM, 8 units of Dpn I endonuclease, and a different activity of Dam MTase. After adding (19) Geier, G. E.; Modrich, P. J. Biol. Chem. 1979, 254, 1408-1413.

Analytical Chemistry, Vol. 79, No. 3, February 1, 2007

1051

Figure 2. Characterization comparison of the linear probe to the hairpin probe. (A) The sequence of the linear probe, which was composed of two DNA sequences and had the same sequence as the stem of the designed hairpin probe shown in Figure 1 A. (B) The wavelength scan spectra of the linear probe before (line b) and after (line b′) cleavage at the methylation site. (C) The wavelength scan spectra of the hairpin probe before (line c) and after (line c′) cleavage at the methylation site. In order to obtain a symmetric spectrum, 521 nm was chosen as the excitation wavelength.

the buffer, probe substrate, SAM, and Dpn I into the cuvette in order, the samples were equilibrated to the incubation temperature, which was confirmed by a steady background fluorescence emission over 5 min. Subsequently, the Dam MTase was added to initiate methylation reaction, and the fluorescence signal was continuously measured over a time course of 30 min. Methylation Assay by Gel Electrophoresis. The gel electrophoresis assays were performed under the conditions as mentioned above but with the following changes: 5 µM hairpin probe, 8 units of Dam MTase, and 20 units of Dpn I. In the gel electrophoresis assay, the 5-µL samples were removed from the reaction solution after having been incubated at 37 °C for 30 min or 3 h and then heated at 65 °C for 15 min to stop the reaction. Subsequently, the samples were put on a polyacrylamide gel (20% acrylamide, 19:1, acrylamide/bisacrylamide) to separate the cleaved products from the substrate. The electrophoresis was carried in 1× TBE (pH 8.0) at 200 V constant voltage for 3 h. After ethidum bromide staining, the gel was scanned using the Image Master VDS-SL (Amersham). Activity Assay of Dam MTase. A series of standard Dam MTase solutions were prepared from 0.5 to 5 units. To protect the activity of Dam MTase, all of these standard solutions were prepared under 4 °C and stored at -20 °C. The reaction sample for the time scan included a 200 nM hairpin probe, 1× M buffer (10 mM Tis-HCl pH 7.5, 5 mM NaCl 10 mM MgCl2, 1 mM DTT), 160 µM SAM, 8 units of Dpn I endonuclease, and different concentrations of Dam MTase. This allowed the whole methylation reaction to be rapid and agreed with the kinetic principle of the standard couple enzyme-linkage reaction. The initial rate of the 1052

Analytical Chemistry, Vol. 79, No. 3, February 1, 2007

methylation reaction was determined by the linear regression of the initial parts of the fluorescence scan curves after the retention period. Influence of Some Drugs on DNA Methylation. Because there were two enzymes involved in this reaction system, it was necessary to eliminate the influence of drugs on Dpn I. Therefore, an absolutely methylated hairpin probe was prepared by putting DAM MTase, SAM, and the hairpin probe together for 6 h to ensure that all of the probes were methylated. A fluorescence enhancement velocity was obtained by adding 8 units of Dpn I to initiate the cleavage reaction, and the fluorescence signal was continuously measured over a time course of 30 min. As for investigating the influence of drugs on the activity of Dpn I, the procedure was similar to the one described above, except that 1 µM drug was added into the solution. The influence of drugs on the activity of Dam MTase were similar to that noted above except for 1 µM concentrations of different drugs in the samples. After adding the buffer, hairpin probe substrate (100 nM), 160 µM SAM, 8 units of Dpn I, and 1 µM different drugs into the cuvette, the time fluorescence scan started after 8 units of Dam MTase was present. The activity of Dam MTase was also obtained by measuring the initial reaction velocity. To investigate the relationship between the concentration of a specific drug and the inhibition ratio, different concentrations of this drug were added in these samples. RESULTS AND DISCUSSION Design Strategy of Real-Time Monitoring of DNA Methylation. The principle behind monitoring the methylation process was based on the fact that restriction endonuclease quickly cleaves

Figure 3. (A) Time scan spectra of the hairpin probe with and without Dam methylase. Curves 1 and 2 for the prescence and absence of 8 units of Dam methylase. t0 is the time of addition of Dam methylase. (B) The enlarged time scan spectrum in the retention period, from t0 to t1. The dashed straight line was the regression line of the curve. (C, D) The gel electrophoresis images for incubation the sample for 30 min and 3 h. Lanes: (1) hairpin probe; (2) hairpin probe with Dpn I endonucleases; (2) hairpin probe, Dpn I endonucleases and Dam methylase.

the methylated DNA probe after it is produced. Therefore, the DNA probe sequence should be properly designed to perform an enzyme-linkage reaction. Because Dam MTase and Dpn I restriction endonuclease have the same recognition site, a hairpin structure DNA probe was designed as shown in Figure 1 A. The 5′-G-A-T-C-3′ sequence containing 17 base pairs was used in the stem and had a distance of 6 bases to the terminus of the probe. This is because several neighboring bases were needed to accelerate the enzyme reaction. Four T bases were used as the loop of the probe. TAMRA and DABCYL groups were linked at the 5′- and 3′-ends of the probe, respectively. According to the software Primer Premier 5.0, it was shown that the ∆G of this 17-base probe was -30.2 kcal/mol. Reducing the base on the

terminus would increase the ∆G value of the probe, indicating that the intramolecular hybridization decreased. Therefore, in the 17-base pairs probe, TAMRA and DABCYL are close to each other due to the strong intramolecular hybridization. The fluorescence of the fluorophore was significantly quenched by FRET. When both Dam MTase and Dpn I restriction endonuclease are present in the probe solution, the enzyme-linkage reaction will begin, possibly due to the mechanism shown schematically in Figure 1B. First, Dam MTase catalyzed the methylation reaction on the recognition sequence to yield the methylation duplex DNA 5′-G-Am-T-C-3′. Once the methylated probe substrate was present, the cleavage reaction by Dpn I restriction endonuclease was initiated to cleave the methylated probe into two parts. One part was a new ssDNA containing the loop and the sticky terminus. The other was a new hybrid containing a 9-base ssDNA linking TAMRA and a 7-base ssDNA attached by DABCYL. According to the Tm value calculation equation (Tm ) 4 °C × G:C pair + 2 °C × A:T pair), the newly formed hybrid only had a Tm value of 20 °C and would be separated into two short DNA fragments at the reaction temperature (37 °C). Consequently, the spontaneous separation of TAMRA and DABCYL would take place, thus causing the fluorescence restoration of TAMRA. Therefore, the dynamic process of DNA methylation was monitored by recording the fluorescence signal of the sample. The FRET-based probe could also be designed by employing two linear oligos linking a fluorophore and a quencher, respectively.20 In this case, enough base pairs were needed to form a stable hybrid to cause FRET. A linear probe employing two oligos (oligo 1 and oligo 2) containing the methylation sequence was also designed and is shown in Figure 2A. Oligo 1 and oligo 2 had the same sequence and length as the stem of the above hairpin probe, but there is no loop sequence between these two oligos. TAMRA was linked at the 5′-terminus of oligo 1 and DABCYL at the 3′-terminus of oligo 2. When oligo 1 and oligo 2 were mixed together, FRET would take place, and the sample displayed no fluorescence. After being treated with Dam MTase and Dpn I, a fluorescence enhancement was observed, and is illustrated by the spectrum scan shown in Figure 2B. The enhancement was 3.18 (final fluorescence signal to background). For hairpin probes, a more significant fluorescence enhancement was obtained and is shown in Figure 2C. The ratio of signal to background could reach 9.37. It was also found that the fluorescence enhancement velocity was higher for the hairpin probe than for the linear probe. A possible reason for this is that the linear probe had an intermolecular hybridization between two oligos, which was slower than that of the intermolecularly hybridized hairpin probe. The hairpin probe showed a higher sensitivity than the linear probe, and therefore, it was used in the following experiment. Real-Time Monitoring of DNA Methylation Process. For this study, we developed an assay that used a simple hairpin probe to monitor the real-time DNA methylation process by coupling it with an enzyme-linkage reaction. Figure 3A shows the real-time fluorescence scan curves of two samples: curve 1 for the hairpin probe, Dpn I with the addition of Dam Mtase, and curve 2 for the hairpin probe with Dpn I. There was no fluorescence enhancement for curve 2, indicating that the Dpn I endonuclease would not cleave the probe in the absence of Dam MTase. In the case of (20) Marras, S.; Kramer, F.; Tyagi, S. Nucleic Acids Res. 2002, 30, e122.

Analytical Chemistry, Vol. 79, No. 3, February 1, 2007

1053

Figure 4. Effect of species on methylation process. The initial methylation velocity rate for different concentrations of (A) hairpin probe, (B) SAM, (C) Mg2+ range 0, 2, 5, 10, 15, 20, and 25 mM, and (D) Na+ range 0, 50, 100, 150, and 200 mM. All of the initial methylation velocities in (C) and (D) were normalized.

Figure 5. Assay of the activity of Dam methylase. The time curves of methylation catalyzed by various activity of Dam methylase. The curves from the bottom to the top were these obtained with different activities: 0, 0.8, 2.0, 4.0, 8.0, 20, and 40 units/mL. (Inset) The initial methylation velocity rate was plotted as a function of the activity of Dam methylase.

curve 1, a rapid increase of fluorescence was detected after the addition of Dam MTase at t0, which was defined as the time at which Mtase was added. The slope of the curve decreased gradually as time lapsed. After 30 min, the fluorescence signal increased ∼10 times over that of the background. This result clearly shows that fluorescence restoration happened when the methylation reaction took place and the cleavage reaction was initiated. When the reaction time was prolonged, the slope of the curve decreased gradually. This shows that methylation was slowed as the substrate was consumed. Therefore, this method had the ability to monitor real-time DNA methylation and provide information about the dynamics of the methylation process. It was also interesting to note that there was a change in the slope of the curve when Mtase was first added, which is shown by the straight and dash lines in Figure 3B. The fluorescence 1054

Analytical Chemistry, Vol. 79, No. 3, February 1, 2007

enhancement velocity was slow before time t1, which is the connection point of the two dash lines. After t1, the velocity increased and the time gap between t0 and t1 was the so-called retention period of the enzyme-linkage reaction.21 In our report, this period is as short as 1 min when the hairpin probe was methylated to initiate the cleavage reaction catalyzed by Dpn I. After the retention period, the fluorescence increased quickly as shown by the straight dash line. These phenomena agreed with the proposed mechanism of the enzyme-linkage reaction. Our method showed the retention period in real time, thereby enabling us to determine the retention period of the enzyme-linkage reaction. To test the precision of our methods, a gel electrophoresis assay (described in the Experimental Section) was carried out as a comparison. The results are shown in Figure 3C and D. As seen in lanes 1 and 2 in Figure 3C, there was only one band of the original probe when the Dam MTase was absent, indicating that no methylation/cleavage reaction occurred. This was in agreement with the fluorescence time scan results of curve 2 in Figure 2A. The new band appeared in lane 3 when the Dam MTase was added, suggesting that the methylation reaction happened. This result is in accordance with the fluorescence time scan spectrum of curve 1 in Figure 2A. However, the new band is very vague. The methylation time needed to be extended up to 3 h to get a clear band, which is shown by the appearance of a cleavage segment in Figure 3D. These results clearly show that the hairpin probe was as efficient as gel electrophoresis in detecting DNA methylation. It is noteworthy that this assay only took 0.5 h to (21) Lehninger, A. L.; Nelson, D. L.; Cox, M. M. Principle of Biochemistry; Worth Publishers: New York, 1993; pp 51-120.

its maximum when the concentration of Mg2+ was 10 mM, and the fluorescence enhancement velocity decreased when the concentration of Mg2+ was above 10 mM. It is well known that Mg2+ is an essential coenzyme for restriction endonuclease (including Dpn I endonuclease) and will hinder the process of methylation if its concentration is very high.22In the case of Na+, the initial methylation velocity did not change in the range of 0-50 mM Na+, as demonstrated in Figure 4D. However, the initial velocity decreased proportionally when the concentration of Na+ was above 50 mM. The velocity was close to zero when the concentration of Na+ was 200 mM. All of these control experiments confirmed that the initial methylation rate was indeed altered by these external factors. Therefore, the concentration of the probe, Mg2+, and Na+ should be optimized to 100 nM, 10 mM, and 50 mM to obtain a rapid methylation /cleavage reaction. Assay of E. coli Dam Methylase. The difference between the classic couple enzyme-linkage reaction and the common enzyme reaction was that there was a significant retention period.21 The classic couple enzyme reaction could be shown as the following equation: Ex

Ei

A 98 B 98 C

Figure 6. (A) Influence of different drugs on the activity of Dam MTase. (1) No drug, (2) benzylpenicillin, (3) ampicillin sodium, (4) gentamycin, (5) mitomycin, (6) ofloxan, and (7) 5-fluorouracil. (B) The inhibition of different concentrations of gentamycin on the activity of Dam MTase. Curves 1 and 2 were the time scan spectrum with and without gentamycin in the sample. The inset chapter was the inhibition ratio at different concentrations of gentamycin (0, 100, 250, 500, and 1000 nM).

finish, while gel electrophoresis requires up to 3 h to complete, in addition to a preparation time that can last from a few hours up to an entire day. Moreover, this hairpin probe-based assay can be monitored in real time and is homogeneous, which is not the case in traditional gel electrophoresis. Optimization of Substrate Affecting Methylation Process. The methylation process was affected by many molecular species including enzymes, reaction substrate, SAM, and metal ions. To get a rapid methylation reaction and high fluorescence enhancement, the concentrations of hairpin probe, SAM, and several metal ions were optimized to obtain a high level of sensitivity. In Figure 4A, the results show that the initial methylation rate, V0, increased quickly as substrate concentration increased into the range of 0-100 nM. When the probe concentration increased in excess of 100 nM, the methylation rate did not increase significantly. Similarly, the initial methylation/cleavage velocities were also increased by varying the concentration of SAM from 10 to 160 µM (Figure 4B). The fluorescence enhancement velocity increased slowly when the concentration of the probe exceeded 80 µM. It is well known that SAM was not very stable in in vitro experiments, so the SAM concentration used was as high as 160 µM. Figure 4C indicates that the methylation/cleavage reaction also happened, but with a very low velocity if no Mg2+ existed in the reaction mixture. The velocity of methylation/cleavage reached

Ex and Ei represent the function enzyme and indictor enzyme, respectively. At first, only substrate A was present and the velocity of the first reaction, Vx, was fast. On the contrary, the amount of B was small and the velocity of the indictor reaction, Vi, was very slow. During the process of the reaction, the yield of B increased and the velocity of the indictor reaction also rose. The time when Vx was finally equal to Vi was called the retention period. To obtain a precise assay result, the retention period must be reduced to its minimum value. Therefore, the activity of the indictor enzyme should be large enough to keep the yield of B from transferring to C as soon as possible. After the retention period, Vx was always equal to Vi and the activity of Ex was induced from the change in the velocity of C. In our couple enzyme-linkage reaction, A and B are the concentration of hairpin probe and methylation probe, which were the initial reaction products and the second reaction substrates, respectively. C was the concentration of the cleavage probe, which was described by the restored fluorescence signal. From Figure 3, it was shown in Figure 5 that the retention period was only 300 s, and we could calculate the velocity thereafter. It was clearly shown that the initial methylation/cleavage velocity was proportional to the concentration of Dam MTase in the range of 0.8-40 units/mL when the activity of Dpn I endonuclease was equal to 8 units. This constituted the basis for our quantitative assay of Dam MTase. Under the reaction conditions mentioned in the Experimental Section, the detection limit was down to 0.8 units/mL. Similar to other enzyme reactions, a lower detection limit could be reached by optimizing all of the methylation and cleavage conditions, for instance, by increasing the concentration of substrates or prolonging the cleavage time. Investigation of Drugs on the Activity of Methylase. Because Dam MTase exists in many bacteria and pathogens and is essential for bacterial virulence or viability in multiple negative (22) Pingoud, A.; Jeltsch, A. Eur. J. Biochem. 1997, 246, 1-22.

Analytical Chemistry, Vol. 79, No. 3, February 1, 2007

1055

pathogens, it is a promising target for antimicrobial drugs.23,24 Our new approach has the potential for high-throughput screening of inhibitors of Dam MTase. We investigated some direct inhibitors of Dam MTase such as antibiotics and anticancer drugs. Because there were two enzymes involved in this reaction system, it was necessary to eliminate the effect of drugs on Dpn I before studying the effect of drugs on the activity of Dam MTase. The control experiments were performed as described in the Experimental Section, and all of the probes were methylated at first because the cleavage site of Dpn I was the methylation sequence. The results indicated that almost all antibiotic drugs, except for penicillin, had no influence on the activity of Dpn I when the concentrations of the drugs were 100 µM. However, all of the antitumor drugs significantly inhibited the activity of Dpn I when the same concentration was employed. The possible reason was that the antitumor drugs were more toxic than the antibiotics. However, all of the drugs had no influence on the activity of Dpn I endonuclease when the concentration was down to 1 µM. When the concentration of all of the drugs was 1 µM, the influence of these drugs on the activity of Dam MTase was directly measured. The drugs effected Dam methylase differently, as shown in Figure 6A. It was very clear that the benzylpenicillin, gentamycin, and 5-fluorouracil could strongly inhibit the methylase. The inhibition of gentamycin was the most serious one, reaching an inhibition ratio of 60%. 5-Fluorouracil was the second most effective one, much more so than benzylpenicillin. However, the ampicillin sodium, ofloxan, and mitomycin had almost no effect on the methylation. We also investigated how the concentration of gentamycin influenced the activity of methylase. The concentration of genta-

mycin was kept under 1 µM. As shown in Figure 6B, the inhibition ratio was almost a straight line. As the concentration of gentamycin increased, inhibition was enhanced. Therefore, our method has the potential to select the antimicrobial drugs, which could decrease the activity level of Dam MTase in bacteria.

(23) Heithoff, D. M.; Sinsheimer, R. L.; Low, D. A.; Mahan, M. J. Science 1999, 284, 967-970. (24) Low, D. A.; Weyand, N. J.; Mahan, M. J. Infect. Immun. 2001, 69, 71977204.

Received for review November 14, 2006.

1056

Analytical Chemistry, Vol. 79, No. 3, February 1, 2007

CONCLUSIONS DNA methylation is critical in many biological events. We have developed a new method for the real-time monitoring of DNA methylation catalyzed by methylase. Molecular beacon DNA probes were used to monitor DNA methylation. The MBs will have their stem sequences perturbed by DNA methylation, resulting in the restoration of the MB fluorescence. Dam MTase and Dpn I endonuclease with a 5′-G-A-T-C-3′ recognition sequence were used. Unlike traditional gel electrophoresis, this fluorescence assay easily provided us with the ability to detect the activity of methylase in homogeneous solutions in real time. The new assay could be used to monitor the dynamic process of DNA methylation. This method was simple and nonradioactive, yet remained as efficient as gel electrophoresis. We used this new method to screen suitable inhibitor drugs of Dam methylase. ACKNOWLEDGMENT This work was supported by the National Key Basic Research Program (2002CB513100), Major International (Regional) Joint Research Program of Natural Science Foundation of China (20620120107), Natural Science Foundation of China (90606003, 200305006), and Key Project of Hunan Province Technology Plan (03SSY1006).

AC061694I

September

7,

2006.

Accepted