Determination of methylation specificity of sequence-specific DNA methyltransferases using matrix assisted laser desorption/ionization time-of-flight mass spectrometry
Determination of methylation specificity of sequence-specific DNA methyltransferases using matrix assisted laser desorption/ionization time-of-flight mass spectrometryTakashi Tamura, Yoshinori Araki, Seiji Yamaoka, Kenji Inagaki and Hidehiko Tanaka*
Department of Bioresources Chemistry, Faculty of Agriculture, Okayama University, 1-1-1 Tsushimanaka, Okayama 700, Japan
Received July 16, 1997;Revised and Accepted September 2, 1997
ABSTRACT
We describe here a sensitive and straightforward method for characterizing the methylation specificity of type II DNA methyltransferase (MTase) using matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry. DNA substrate, prepared by ligation of a commercially available oligonucleotide, was modified by the subject MTase, and was derivatized to a mixture of single-stranded oligonucleotides through endonuclease treatment, heat-denaturation and limited digestion by 3'-terminus-specific phosphodiesterase I. MALDI-TOF mass spectrometry was used to determine the mass differences between the digestion products, and the methylated nucleotide was explicitly identified by the mass increase of 14 Da due to the base modification. The method was applicable to the three representative MTases M.EcoRI, M.BamHI and M.HaeIII.
Mass spectrometry is an intrinsically attractive approach for sequencing modified oligonucleotides because the structural elements are represented by differences in mass. Pieles et al. first invented a method for sequencing single-stranded oligonucleotide by time-dependent phosphodiesterase digestion of an oligonucleotide coupled with matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (1 ). They used termini-specific exonucleases to sequentially digest a single-stranded oligonucleotide, and a portion of the digestion sample was analyzed by the MALDI mass spectrometry to determine the mass differences between the digestion products, which provide the sequence information. We here demonstrate that the combination of phosphodiesterase I digestion and MALDI-TOF mass spectrometry enabled a rapid, sensitive and straightforward method for characterizing the methylation specificity of various type II DNA MTases. Type II MTases have been characterized to modify bases of either C or A symmetrically positioned in the complementary strands of a palindromic recognition sequence (2 ), and the methylation can be at N6 of adenine, or at N4 or C5 of cytosine. Our method is applicable to a type II MTase whose recognition sequence is already known. The method does not require handling radioactive materials nor any sulfur analogs of nucleotides, but it uses a commercially available oligonucleotide which contains the recognition sequence. Our method is described in the following examples in which methylation specificity was determined for M.EcoRI, M.BamHI and M.HaeIII.
A commercially available 12mer pEcoRI linker DNA, 5'-pCCGGAATTCCGG-3' (5000 pmol; TaKaRa, Japan), was dissolved in 43 µl of deionized water and incubated on GeneAmp PCR System 2400 (Perkin Elmer) for 10 min at 95, 85, 75, 65, 55, 45oC and left at 37oC for several hours. To the resulting solution were added 2 µl of T4 ligase (1000 U/µl; Nippon Gene, Japan) and 5 µl of ligation buffer (*10; 500 mM Tris-HCl pH 7.9, 100 mM MgCl2, 200 mM DTT, 10 mM ATP), and incubated at 16oC for 24 h. Then, 4 µl from the resulting solution was mixed with 4 µl of M.EcoRI solution (40 U/µl; TaKaRa), 2 µl of 800 µM S-adenosyl-l-methionine and 10 µl of methylation buffer (*2; 200 mM Tris-HCl pH 8.0, 4 mM DTT, 20 mM EDTA). The mixture was incubated at 37oC for 2 h, and the volume was increased to 50 µl by adding 30 µl of deionized sterile water. The modified polynucleotide thus obtained was separated from salts and buffer components using Bio-Spin Chromatography Column 6 (BioRad). The eluate from the column was mixed with 5 µl of 100 mM ammonium acetate buffer pH 7.5, 5 µl of 100 mM MgCl2 and 3 µl of R.HaeIII solution (10 U/µl 50% glycerol; TaKaRa), and incubated at 37oC for 2 h. Then, the sample was heated at 96oC for 10 min, and cooled on ice. After the solution was evaporated on SpeedVac concentrator (Servant) and dissolved in 20 µl of 5 mM ammonium acetate buffer pH 7.5, snake venom phosphodiesterase I (EC 3.1.15.1, Boehringer Mannheim, Germany; 100 mU) was added and incubated at 37oC. A portion of 4 µl was taken from the mixture every 5 min and boiled for 10 min to terminate the digestion. Then, each sample solution was mixed with 10-20 mg of ion exchange resin (Dowex 50W-X8, 50-100 mesh, ammonium form; Dow Chemical Company), and vigorously vortexed for 2 min (3 ). Fresh solutions of 0.5 M 2,4,6-trihydroxyacetophenone (THAP) in methanol and 0.4 M diammonium hydrogen citrate in 50% aqueous acetonitrile were daily prepared before the analysis. MALDI samples were prepared by mixing 0.5 µl of the ammonium citrate solution and 1.0 µl of DNA solution on a solid sample probe tip, and allowed the drop to almost dry. Then, 0.5 µl of the THAP solution was loaded on the sample spot, and allowed to dry. Mass spectra were obtained in the negative ion mode on a ThermoBioanalysis VISION 2000 (Hemel Hempstead, UK) instrument equipped with a 337 nm emission nitrogen laser (Laser Science Inc., Newton, MA, USA) equipped with an ion reflector. Ions are accelerated to an energy of 5 keV in the ion source. Before hitting the Secondary electron multiplier (SEM) detector (Hamamatsu R2362), ions are postaccelerated by a conversion dynode to a final impact energy of 25 kV. The spectrometer was calibrated externally with the mass peaks of THAP, flavin adenine dinucleotide, 8mer EcoRI linker and 12mer pEcoRI linker. Figure 1 shows the MALDI-TOF mass spectra of the 5, 10 and 20 min digestion, and each peak observed is labeled with `exo' to distinguish it from the molecular ion of the undigested strand. Peaks corresponding to the oligonucleotide after the cleavage of G (exo1), then G (exo2), C (exo3), C (exo4), T (exo5), T (exo6) and mA (exo7) were observed. Each molecular ion was assigned to the digestion products derived from the modified oligonucleotide (Table 1 ). The mass difference between exo6 and exo7 was 326.5 Da, which is in good agreement with the molecular mass of N6-methyl-adenylate of 327.2 Da. The smaller fragment peaks on the low m/z side of the exo peaks occur fortuitously at positions where the mass differences are much less than a single nucleotide loss and therefore cannot be misinterpreted as products of the phosphodiesterase activity. Although quantitative information obtained from relative intensity of mass spectral peaks is not very reliable in MALDI, the relative intensity of molecular ion peaks changed as the function of digestion time, indicating the change in the population of oligonucleotides during the phosphodiesterase digestion. Interestingly, exo6 showed highest peak intensity among the others, indicating the oligonucleotide bearing a N6-methyl-adenylate at the 3'-end can retard the enzymatic digestion.
