Characterization of an N 6 adenine methyltransferase from Helicobacter pylori strain 26695 which methylates adjacent adenines on the same strand Ritesh Kumar 1 , Asish K. Mukhopadhyay 2 and Desirazu N. Rao 1 1 Department of Biochemistry, Indian Institute of Science, Bangalore, India 2 Division of Bacteriology, National Institute of Cholera and Enteric Disease, Kolkata, India Introduction DNA methylation is one of the most common forms of DNA modification occurring in the prokaryotic genome. This modification does not affect the Wat- son–Crick pairing, but creates a signature motif that can be recognized by the proteins interacting with DNA. It has been shown that DNA methylation can enhance or abrogate the affinity of transcription fac- tors for DNA, thus affecting gene expression and regu- lation. These base modifications thus act as a second line of genetic information [1]. Prokaryotic DNA methyltransferases (MTases) are classified into two major groups – exocyclic amino MTases and endocyclic MTases – based on the position of the methyl group on the bases. The exocy- clic amino MTases methylate adenine at the N 6 position and cytosine at the N 4 position, whereas endocyclic MTases methylate the cytosine at the C 5 position [2,3]. In prokaryotes most of the MTases are associated with a restriction enzyme and form a restriction-modification (R-M) system. R-M systems are involved in the protection of bacteria from bacte- riophage invasion. However, the identification of MTases without any associated restriction enzyme in many bacteria has compelled biologists to explore the functions of MTases beyond the distinction of self and nonself DNA. Extensive work on solitary MTases, Keywords base flipping; DNA methyltransferase; Helicobacter pylori; S-adenosyl- L-methionine; site-directed mutagenesis Correspondence D. N. Rao, Department of Biochemistry, Indian Institute of Science, Bangalore 560 012, India Fax: +91 80 2360 814 Tel: +91 80 2293 2538 E-mail: dnrao@biochem.iisc.ernet.in (Received 18 November 2009, revised 26 December 2009, accepted 25 January 2010) doi:10.1111/j.1742-4658.2010.07593.x Genomic sequences of Helicobacter pylori strains 26695, J99, HPAGI and G27 have revealed an abundance of restriction and modification genes. hp0050, which encodes an N 6 adenine DNA methyltransferase, was cloned, overexpressed and purified to near homogeneity. It recognizes the sequence 5¢-GRRG-3¢ (where R is A or G) and, most intriguingly, methylates both adenines when R is A (5¢-GAAG-3¢). Kinetic analysis suggests a nonpro- cessive (repeated-hit) mechanism of methylation in which HP0050 methyl- transferase methylates one adenine at a time in the sequence 5¢-GAAG-3¢. This is the first report of an N 6 adenine DNA methyltransferase that methylates two adjacent residues on the same strand. Interestingly, HP0050 homologs from two clinical strains of H. pylori (PG227 and 128) methylate only 5¢-GAGG-3¢ compared with 5 ¢-GRRG-3¢ in strain 26695. HP0050 methyltransferase is highly conserved as it is present in more than 90% of H. pylori strains. Inactivation of hp0050 in strain PG227 resulted in poor growth, suggesting its role in the biology of H. pylori. Collectively, these findings provide impetus for exploring the role(s) of this conserved DNA methyltransferase in the cellular processes of H. pylori. Abbreviations 2AP, 2-aminopurine; AdoMet, S-adenosyl- L-methionine; Dam, DNA adenine methylase; DLS, dynamic light-scattering; IPTG, isopropyl thio-b-D- galactoside; K D , dissociation constant; LB, Luria–Bertani; MTase, methyltransferase; R h , hydrodynamic radius; R-M, restriction-modification. 1666 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS such as DNA adenine methylase (Dam) and cell cycle- regulated methylase (CcrM), have indeed shown the role of DNA methylation in regulating cellular events such as bacterial virulence, cell cycle regulation and phase variation [4–6]. The Gram-negative bacterium Helicobacter pylori persistently colonizes the human stomach and is wide- spread throughout the world. It is a major cause of gastritis and peptic ulcer disease, and is an early risk factor for gastric cancer. H. pylori is one of the most genetically diverse species of bacteria, and strain-spe- cific genetic diversity has been proposed to be involved in the organism’s ability to cause different diseases [7,8]. Analysis of genome sequences of H. pylori strains 26695 and J99 revealed the presence of 23 and 22 R-M systems, respectively, far more than the number detected in other bacterial genomes sequenced to date [9–11]. Two more H. pylori strains – HPAG1 (isolated from a patient with chronic atrophic gastritis) and G27 – were sequenced and a similar number of puta- tive R-M systems were identified [12,13]. Comparison of strains 26695 and J99 showed that the two genomes are quite similar, with only 6–7% strain-specific genes. R-M systems are a major source of the strain differ- ences [14]. iceA-hpyIM, which encodes a cognate restriction enzyme and an N 6 adenine methylase has been studied in various H. pylori strains. It was shown that hpyIM expression is growth-phase regulated and required for normal bacterial morphology. Deletion of hpyIM altered the expression of the stress-responsive dnaK operon, suggesting that hpyIM may play a role in H. pylori physiology beyond its R-M function [15]. The Type II MTase, M.HpyAIV, which recognizes the 5¢-GANTC-3¢ site, has been shown to affect the expression of the katA gene encoding the H. pylori catalase [16]. H. pylori 26695 has three DNA MTases that lack cognate restriction enzymes. Vitkute et al. [17] and Lin et al. [18] showed that HP0050, an orphan N 6 adenine MTase from H. pylori 26695 recognizes 5¢-GAGG-3¢ and methylates adenine; these findings were based on the results of a restriction endonuclease assay. The ORF hp0050 has been reported to be part of an R-M system that contains two MTases and an inactive restriction endonuclease. This R-M system was later assigned as HpyAVI, with hp0050 designated as M1.HpyAVI, hp0051 as M2.HpyAVI and hp0052 as HpyAVIP (putative). The hp0050 homolog of H. pylori strain HPAGI (HPAG1_0046) is a chronic atrophic gastritis-associated gene [12]. Strain-specific DNA-modification genes are thought to influence strain-specific phenotypic traits, host speci- ficity, adaptability to changing micro-environmental conditions or virulence [14]. The identification and study of both species-specific and strain-specific MTases of H. pylori could enhance our understanding of the pathogenic mechanisms of this organism. Our findings indicate that hp0050 from strain 26695 has evolved a relaxed specificity as a result of mutations, compared with other strains. These observations fur- ther highlight the capability of this organism to undergo random mutations and evolve proteins with new functions. Results and Discussion HP0050 is an N 6 adenine MTase from H. pylori and belongs to the b subgroup of MTases, based on the linear arrangement of the S-adenosyl-l-methionine (AdoMet)-binding domain (FXGXG), the target rec- ognition domain and the catalytic domain (DPPY). HP0050 MTase is present in all the three sequenced strains of H. pylori. HP0050 MTase is present in all the three strains of H. pylori (26695, J99 and HPAGI) for which genome has been sequenced. The HP0050 proteins from H. pylori J99 and HPAG1 have 91.7% and 90% identities respectively, to the HP0050 protein from H. pylori 26695 [19]. In H. pylori 26695, hp0050 exists as an overlapping ORF with another MTase, hp0051. These MTases are remnant MTases of a defunct R-M system. Both these ORFs have a high similarity with the MnlI DNA MTase that belongs to the Type IIS R-M system [20]. However, in H. pylori the functional MnlI restriction enzyme is absent [21]. Cloning, overexpression and purification of HP0050 protein A 699 bp fragment (Fig. S1A), representing the hp0050 gene from H. pylori 26695, was PCR amplified using primers 1 and 2 (Table 1) and cloned between the BamHI and XhoI sites of the expression vector pET28a (data not shown). A polypeptide of the expected Table 1. Primers used for cloning and mutagenesis. The restriction enzyme site is indicated in bold letters. SN, serial number. SN Primer sequence (5¢-to3¢) Restriction site created(+) ⁄ lost(-) 1 GGATCCATGATACAAATTTATCACGCT BamHI (+) 2 CTCGAGTTAAAACAGATTCAAACG XhoI (+) 3 GGATCCGATCTTAAAAAGCTTAAGAAAATG BamHI (+) 4 CTCGAGATTCAAATAGCGTTTTTA XhoI (+) 5 TAGATCCTTCCATGGGGAGCGGCACCACCGGCT NcoI (+) 6 AACCGAAATGTTTAAAGGAGGGTCCGTGATGAT Psi I (-) R. Kumar et al. N 6 adenine methyltransferase from H. pylori 26695 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS 1667 molecular mass (32 kDa) was expressed at high levels upon induction with 0.5 mm isopropyl thio-b-d-galac- toside (IPTG) (Fig. S1B). HP0050 was expressed as an N-terminal His-tagged protein and was purified. As the purified HP0050 protein has an N-terminal His- tag, western blot analysis was carried out with anti-His IgG and a single band corresponding to HP0050 pro- tein was detected (Fig. S1C). The His-tag was removed using the Thrombin cleancleave TM kit, according to the manufacturer’s instructions (see the Experimental procedures). The protein was purified to > 95% homogeneity, as judged by SDS ⁄ PAGE followed by silver staining (Fig. S1D). Peptide finger mapping of HP0050 A peptide finger map of the HP0050 protein was obtained by digesting purified HP0050 protein with trypsin and subjecting it to MALDI analysis. The finger map thus obtained was then matched with the expected finger map. It was found that eight pep- tide ions matched with the expected ions, as shown by the asterisk in Fig. S2A, suggesting the authenticity of the purified protein. Oligomeric status of HP0050 protein HP0050 protein elutes as a monomer, and the molecular mass was determined to be 28 kDa by analytical gel-fil- tration chromatography (Fig. S2B). Dynamic light-scat- tering (DLS) measurements on HP0050 MTase were performed on a DynaPro DLS instrument using 20 lL of 1.5 mgÆmL )1 of protein with a data-acquisition time of 10 s. Scattering intensities at various time intervals (ls) with the initial (t = 0 s) intensity were compared and a combined correlation function was constructed (inset, Fig. S2C). As seen in Fig. S2C, DLS data, when fitted to the Stokes–Einstein equation, gave a hydrody- namic radius (R h ) of 2.2 nm. The frictional ratio was calculated as 0.89, suggesting that HP0050 is more or less spherical in structure. An ideal spherical protein would give a value of 1.0. Higher values indicate an anisotropic structure. Kinetics of methylation reaction To establish the relationship between the initial veloc- ity of the reaction and the enzyme concentration, the rate of DNA methylation catalysed by HP0050 was determined. pUC19 DNA was used as a substrate. Different concentrations of HP0050 protein (10– 100 nm) were added to the reaction mixture containing DNA (80 nm) and AdoMet (2.0 lm) and incubated at 37 °C. When the initial velocities were plotted against increasing enzyme concentrations, a linear relationship was obtained (Fig. 1A). This indicated that the initial velocity of the reaction was directly proportional to the enzyme concentration. Next, the initial velocities were determined at various concentrations of the sub- strates, [ 3 H]AdoMet and pUC19 DNA. For the deter- mination of K m (DNA) , a series of similar reactions containing HP0050 MTase (100 nm), [ 3 H]AdoMet (2.0 lm) and increasing concentration of pUC19 DNA (10–80 nm) were performed and a conventional hyper- bolic dependence was obtained. Nonlinear regression analysis of initial velocity versus DNA concentration established the K m (DNA) as 19.9 ± 3 nm (Fig. 1B). To determine K m (AdoMet) , a series of reactions containing HP0050 MTase (100 nm), DNA (50 nm) and increasing concentration of [ 3 H]AdoMet (0.3– 12 lm) were performed. Increasing the concentration of AdoMet led to a progressive stimulation in the reaction rate. Whereas the initial portion of the concentration-dependence curve corresponded approx- imately to a conventional hyperbolic dependence, saturation was not achieved (Fig. 1C). Similar obser- vations have been reported for T4 Dam and EcoDam [22,23]. Determination of site of methylation The recognition sequence of HP0050 MTase was previ- ously reported by Vitkute et al. [17], based on restric- tion enzyme digestion, to be 5¢-GAGG-3¢, where A is methylated by HP0050 MTase in the target site. Using different fragments of pUC19 with varying numbers of GAGG sites (fragments 3 to 6), or fragments not con- taining GAGG sites (fragments 1 and 2, Table S1), as a substrates for the methylation reaction by HP0050 MTase, it was observed that besides fragments with GAGG sites, fragment 2 (without GAGG site) was also methylated. It should be noted that fragment 2 has one GAAG site, which could be a recognition site for HP0050 MTase. There are 20 GAAG and 13 GAGG sites per mole- cule of pUC19. To further confirm this observation we used 26 mer duplex substrates (Table 2) with GAGG (duplex 1), GGAG (duplex 2), GAAG (duplex 3), GTGG (duplex 4), GAGA (duplex 5) or GmAmAG (duplex 8) site to determine the specificity of HP0050 MTase. It was found that HP0050 MTase recognized and methylated GAGG, GGAG and GAAG, but did not methylate GTGG, GmAmAG or GAGA (Fig. 2). As HP0050 was able to recognize and methylate both GAGG and GGAG, it was of interest to determine which A was the target for the MTase in the oligonu- N 6 adenine methyltransferase from H. pylori 26695 R. Kumar et al. 1668 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS cleotide with the GAAG sequence. To address this, two 26-mer duplex substrates – one with 5¢- GAmAG- 3¢ (duplex 6) and the other with 5¢-GmAAG-3¢ (duplex 7) (where mA is the methyl-adenine) site – were used individually as a substrate in the methylation assay. It was found that HP0050 MTase was able to methylate duplex 6 and duplex 7, suggesting that both adenine residues were targets for HP0050 MTase (Fig. 2), and Table 2. Duplex DNA used in this study. The underlined region of the oligonucleotide represents the HP0050 MTase recognition sequence, and restriction enzyme sites are shown in bold. 2,2- amino purine; Bt, biotin; mA, methyl adenine. Duplex Sequence (5¢-to3¢) 1 TACAATGTACC GAGGATCTATTGATC ATGTTACATGGCTCCTAGATAACTAG 2 TACAATGTACC GGAGATCTATTGATC ATGTTACATGGCCTCTAGATAACTAG 3 TACAATGTACC GAAGATCTATTGATC ATGTTACATGGCTTCTAGATAACTAG 4 TACAATGTACC GTGGATCTATTGATC ATGTTACATGGCACCTAGATAACTAG 5 TACAATGTACC GAGAATCTATTGATC ATGTTACATGGCTCTTAGATAACTAG 6 TACAATGTACC GAmAGATCTATTGATC ATGTTACATGGCTTCTAGATAACTAG 7 TACAATGTACC GmAAGATCTATTGATC ATGTTACATGGCTTCTAGATAACTAG 8 TACAATGTACC GmAmAGATCTATTGATC ATGTTACATGGCTTCTAGATAACTAG 9 TACAATGTACC G2GGATCTATTGATC ATGTTACATGGCTCCTAGATAACTAG 10 TACAATGTACTC GAAGCTATCTATTGATC ATGTTACATGAGCTTCGATAGATAACTAG 11 TACAATGTATCAT GAAGTACTCTATTGATC ATGTTACATAGTACTTCATGAGATAACTAG 12 TACAATGTATCGC GAAGCGCTCTATTGATC ATGTTACATAGCGCTTCGCGAGATAACTAG 13 TACAATGTACTCGAGCTAGATATCTATTTG GAAGCTGATCGAGTC ATGTTACATGAGCTCGATCTATAGATAAACCTT CGACTAGCTCAG 14 ATACTGTACC GAGGCTGCGATCTAGGTCTGCTGAGG ATGATGTTGT TATGACATGGCTCCGACGCTAGATCCAGACGAC TCCTACTACAACA 15 Bt-TACAATGTACC GAAGATCTATTGATC ATGTTACATGGCTTCTAGATAACTAG 16 Bt-ATACTGTACC GAGGCTGCGATCTAGGTCTGCT GAGGATGATGTTGT TATGACATGGCTCCGACGCTAGATCCAGACGAC TCCTACTACAACA 17 TGCGAGGATGGTCTGTC GAAGCTGATGTT ACGCTCCTACCAGACAGCTTCGACTACAA 18 TACAATGTACC GmAAGCTCTATTGATC ATGTTACATGGCTTCGAGATAACTAG A B C 600 400 200 0 500 400 300 200 100 0 2500 2000 1500 1000 500 Methyl groups transfered (mol·min –1 ) Methyl groups transfered (mol·min –1 ) Methyl groups transfered (mol·min –1 ) 0 0 20406080 0 20 40 60 80 100 HP0050 (n M ) pUC19 DNA (n M ) AdoMET (µ M ) 04812 Fig. 1. Kinetics of methylation. (A) Initial ,velocity versus the con- centration of HP0050 MTase. Increasing concentrations of HP0050 MTase (10–100 n M) were incubated with 80 nM pUC19 and 2.0 lM AdoMet in standard reaction buffer at 37 °C for 15 min, then the reaction was stopped and analyzed as described in the Experimen- tal procedures. (B) Determination of K m (DNA) . Methylation assays were carried out in reactions containing 2.0 l M [ 3 H]AdoMet and increasing concentrations of pUC19 DNA (10–80 n M) in standard reaction buffer at 37 °C for 15 min. HP0050 MTase (100 n M) was added to start the reaction. The data points were analysed using nonlinear regression analysis. (C) Initial velocity versus the concen- tration of AdoMet. Methylation assays were carried out in reactions containing 50 n M pUC19 DNA and increasing concentra- tions of [ 3 H] AdoMet (0.3–12 lM) in standard reaction buffer at 37 °C for 15 min. HP0050 MTase (100 n M) was added to start the reaction. R. Kumar et al. N 6 adenine methyltransferase from H. pylori 26695 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS 1669 this was carried out at different protein concentrations (data not shown). Experiments were then performed to estimate the kinetic constants for these DNA substrates by nonlin- ear regression analysis. The duplex with GAGG was the substrate most preferred, with a K m of 5.2 lm, and DNA with a GAAG site was a preferred substrate over DNA with GGAG, GAmAG or GmAAG sites, with a K m of 13 lm compared with K m values of 17 lm, 27 lm or 29 lm, respectively (Table 3). In addition, the k cat ⁄ K m (specificity constant) was calculated for differ- ent DNA substrates and it was found that the specific- ity constant for duplex 1 was 10 times higher than the specificity constant for duplex 2, suggesting that duplex 1 was a better substrate than duplex 2 (Table 3). The specificity constant for duplex 3 was 2.1-fold higher than the specificity constant for duplex 2, and the K m values were very similar, which again suggests that both the adenines are methylated by HP0050 MTase. Furthermore, to confirm the observation that both adenines in GAAG are methylated by HP0050 MTase, duplex 10 (Table 2) was used as a substrate. Duplex 10 contains an HP0050 MTase recognition sequence (GAAG) with overlapping AluI (AGCT) and TaqI (TCGA) restriction sites. Upon methylation with HP0050 MTase if both adenine bases were modified, the DNA would be resistant to both AluI and TaqI digestion. It is clear from Fig. 3A that the methylated duplex is resistant to restriction with AluI and TaqI, confirming that HP0050 MTase indeed methylates both the adenines in 5¢-GAAG-3¢. Furthermore, duplex 11 (Table 2) was used, which contains an HP0050 MTase site (GAAG) with an overlapping ScaI site (AG- TACT), as a substrate in the methylation assays. Upon methylation, if the second adenine was methylated in GAAG, the DNA would be resistant to ScaI digestion. It was found that upon methylation with HP0050 MTase the duplex DNA was resistant to ScaI digestion (Fig. S3A). It is possible that HP0050 MTase binds strongly to the duplex and thus inhibits the cleavage. To rule out this possibility we used duplex 12, which contains an HP0050 MTase site (GAAG) with an over- lapping AfeI site (AGCGCT), as a substrate in the methylation assays. AfeI is not sensitive to the methyla- tion status of adenine in its cognate sequence. It was found that, upon methylation, the duplex DNA was sensitive to digestion with AfeI (Fig. 3B). In addition, duplex 13 was used, which contains two AluI sites – one overlapping with the HP0050 MTase site (GAAG) and other 15 bp away from it. When duplex 13 was methylated by HP0050 MTase and then digested with AluI, two fragments were obtained. It was observed that, upon methylation, the AluI site overlapping with the HP0050 MTase cognate sequence became resistant to AluI digestion. However, three fragments of same size were obtained when unmethylated duplex 13 was digested with AluI (Fig. S3B). To eliminate the possibil- ity that AluI is blocked by modification immediately outside its recognition site, duplex 18 was used. Duplex 18 has a GAAG site with an overlapping AluI site and in which the first A was methylated (GmAAG). Duplex 18 was completely digested with AluI, suggesting that the modification immediately outside the recognition site of AluI has no effect on its activity (Fig. S3C). 1600 1200 800 400 0 0 5 10 15 Oligonucleotide (µ M ) Methyl groups transfered (mol·min –1 ) GAGG GAAG GGAG GmAAG GAmAG GTGG GAGA GmAmAG Fig. 2. Specificity of HP0050 MTase. (A) Methylation activity of HP0050 MTase as a function of increasing concentrations of differ- ent 26-mer duplex DNA species. Methylation assays were carried out in reactions containing 2.0 l M [ 3 H]AdoMet and increasing con- centrations of 26-mer duplex DNA (2.5–15 l M), with one GAGG site or with a modified GAGG site, in standard reaction buffer at 37 °C. HP0050 MTase (100 n M) was added to start the reaction. (d, GAGG; , GAAG; , GGAG; ., GmAAG; r, GAmAG; s, GTGG; h, GAGA; D, GmAmAG). mA, methyl adenine. Table 3. Kinetic parameters for HP0050 N 6 adenine methyltransferase. K m (M) k cat (s )1 ) k cat ⁄ K m (M )1 Æs )1 ) pUC19 1.9 ± 0.3 · 10 )8 0.5 ± 0.05 · 10 )2 2.6 · 10 5 Duplex 1 (GAGG) 5.2 ± 1.0 · 10 )6 2.4 ± 0.05 · 10 )2 4.6 · 10 3 Duplex 2 (GGAG) 17.0 ± 2.0 · 10 )6 0.8 ± 0.03 · 10 )2 0.47 · 10 3 Duplex 3 (GAAG) 13.0 ± 3.0 · 10 )6 1.1 ± 0.02 · 10 )2 0.84 · 10 3 Duplex 6 (GAmAG) 27.0 ± 1.0 · 10 )6 0.3 ± 0.03 · 10 )2 0.11 · 10 3 Duplex 7 (GmAAG) 29.0 ± 2.0 · 10 )6 0.2 ± 0.02 · 10 )2 0.07 · 10 3 N 6 adenine methyltransferase from H. pylori 26695 R. Kumar et al. 1670 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS To further confirm the methylation of adjacent ade- nines in GAAG, we used duplex 17, which contains a FokI site (GGATG) and GAAG, which is 7 bp away from the FokI site. Duplex 17 was used as a substrate and the methylation reaction was carried out in the presence of [ 3 H] AdoMet. The methylated duplex was purified and then digested by FokI, which resulted in two fragments, each containing one adenine from GAAG. These fragments were separated by electro- phoresis on a 20% polyacrylamide gel and then checked for the incorporation of radiolabel. It was found that both fragments were labelled, confirming that HP0050 MTase indeed methylates both the ade- nines in 5¢-GAAG -3¢ (Fig. 4). It is worth mentioning here that the Type IIS MnlI R-M system, comprising N 6 adenine and C 5 cytosine MTase and a restriction endonuclease, recognizes the nonpalindromic nucleo- tide sequence 5¢-CCTC(N)7 ⁄ 6-3¢. While the C 5 MTase modifies the first cytosine base within the 5¢-CCTC-3¢ sequence, the N 6 adenine MTase methylates the bottom strand of the MnlI target, resulting in 5¢-G mAGG. Interestingly, these two MTases share the greatest degree of similarity with HP0050 MTase and HP0051 MTase from H. pylori 26695 [20]. In the case of the FokI MTase, two domains are responsible for methylating two adenine residues – one in the upper strand and one in the lower strand [24].Yet another variation is seen in the case of MmeI, where it has been reported that MmeI modifies the adenine in the top strand of the recognition sequence 5¢-TCC RAC-3¢ and uses modification only on one of the two DNA strands for host protection [25]. Interestingly, M.Alw261, M.Eco31l and M.Esp3l methylate both strands of their recognition sites, yielding C 5 methyl AluITaqI Unmethylated duplex 10 Methylated duplex 10 M * 29 mer 12 mer/13 mer AluI AluI+TaqI TaqI UD AluI AluI+TaqI TaqI UD Unmethylated duplex 12 Methylated duplex 12 M 50 bp 30 bp 12 bp AfeI A feI UD A feI UD A B Fig. 3. Comparison of restriction digestion patterns of methylated and unmethylated duplex DNA. (A) Restriction digestion of 29-mer duplex 10. M, molecular mass marker, AluI and TaqI denote digestion of duplex 10 with these respective enzymes. Schematic representation of the 29-mer duplex 10 is shown with HP0050 MTase and overlapping AluI and TaqI sites. (B) Restriction digestion of the 30-mer duplex 12. AfeI denotes digestion of duplex 12. Schematic representation of the 30-mer duplex 12 is shown with HP0050 MTase and an overlapping AfeI site. The underlined region of the oligonucleotide represents the HP0050 MTase recognition sequence. UD, undigested duplex. * Corresponds to a 50-bp band in the marker. R. Kumar et al. N 6 adenine methyltransferase from H. pylori 26695 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS 1671 cytosine and N 6 methyl adenine on opposite strands [26]. To the best of our knowledge, M.CviPII is the only other DNA MTase that modifies adjacent resi- dues in the cognate sequence. In addition to modifyng the first cytosine in CCD (D = A, G or T) sequences, M.CviPII also modifies both the cytosines in CCAA and CCCG sites [27]. Processivity of DNA methylation HP0050 MTase methylates adjacent adenines in GAAG. The methylation can take place either in a sin- gle binding event or in two separate binding events. To address this, 100 nm HP0050 MTase was pre-incu- bated with 5 lm AdoMet for 10 min at 37 °C to pro- mote the formation of the protein–AdoMet complex. This complex was then made catalytically competent by adding 2 lm duplex 15 and incubated for an addi- tional 5 min on ice to allow the formation of a ternary complex. Following the second incubation, the reac- tion mixture was split in two and 40 lm duplex 3 was added in one set as a trap and the other set was allowed to proceed without the DNA trap. Both the reaction mixes were incubated at 25 °C, and reaction aliquots withdrawn at 2 min intervals were checked for methylation. The reaction mixes were incubated at 25 °C in order to decrease the turnover rate so that the first turnover could be monitored. If HP0050 MTase methylates in a nonprocessive manner, it would dissociate from the substrate mole- cule after each round of methylation and would re-associate in the next round of catalysis. If, however, the MTase works in a processive manner, then it would dissociate from the substrate molecule after methylating both the adenines in GAAG (duplex 15). A biotin– avidin microplate assay was used to separate biotiny- lated substrate from nonbiotinylated duplex DNA and to monitor the methylation of biotinylated substrate. The addition of a molar excess of duplex 3 to duplex 15 at different time-points of the modification reaction of the GAAG substrate resulted in a decrease in the rate of methylation of duplex 15 (Fig. 5A). This result clearly suggests that HP0050 methylates adjacent adenines in a nonprocessive manner. To determine if HP0050 MTase methylates the duplex with two recognition sites in a nonprocessive manner, we used duplex 16 containing two GAGG sites (duplex 14 with a 5¢ biotin tag) as a substrate and duplex 14 as competitor DNA. The biotin–avidin mi- croplate assay was used to separate biotinylated sub- strate from nonbiotinylated duplex DNA and to monitor the methylation of biotinylated substrate. It was observed that, in the presence of a 20-fold excess of nonbiotinylated duplex DNA, the extent of the methylation reaction did not increase, but in the absence of nonbiotinylated competitor, methylation was observed (Fig. 5B). This suggests a distributive mechanism of methylation. In this assay, EcoDam was used as a positive control for the processive mecha- nism of methylation (data not shown). Yet another approach was used to show the proces- sivity of HP0050. A 294 bp dsDNA containing a GAAG site with overlapping AluI and TaqI sites (simi- lar to duplex 10) was used for the methylation assay. The master mix (400 lL) containing 1 lm HP0050 MTase was incubated with 5 lm [ 3 H] AdoMet and Fragment 1 Fragment 2 12 Duplex 17 FokI 29 bp 1 19 10 2200 ± 150 Fragment c.p.m. Duplex 17 Fragment 1 (19 bp) 950 ± 100 Fragment 2 (10 bp) 850 ± 100 20% PAGE Fig. 4. Analysis of the methylation pattern of HP0050 MTase. Duplex 17 was methylat- ed by HP0050 MTase using [ 3 H]AdoMet. After cleavage with FokI, the DNA was electrophoresed through a 20% polyacryl- amide gel and specific restriction fragments (Fragments 1 and 2) were isolated. The labelled methyl group contents of the fragments are shown in counts per minute (c.p.m.). N 6 adenine methyltransferase from H. pylori 26695 R. Kumar et al. 1672 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS 3 lm of 294 bp dsDNA at 25 °C (to decrease the turn- over). Two aliquots (of 25 lL each) were withdrawn at 3-min intervals up to 15 min and the reactions were stopped by snap-freezing in liquid nitrogen. One ali- quot from each time-point was analyzed for protection from AluI digestion and the other for protection from TaqI digestion (Fig. 6). If HP0050 methylates adjacent adenines in a processive manner there should not be any difference in the resistance to AluI digestion and to TaqI digestion. However, if HP0050 methylates adjacent adenines in a nonprocessive manner then the substrate DNA should show early resistance to TaqI digestion compared with AluI digestion. It is evident from Fig. 6 that after 3 min the sub- strate starts showing resistance to TaqI digestion but shows AluI resistance only at the 6-min time-point. These results suggest that HP0050 methylates adjacent residues in a nonprocessive manner. In general, DNA MTases accompanied with a restriction enzyme, such as M.EcoRI exhibit a nonprocessive mechanism of action, whereas solitary MTases, such as T4 Dam and EcoDam, methylate DNA in a processive manner [3]. Purification and characterization of AdoMet-binding motif (F195S) and catalytic motif (Y32L) HP0050 mutant proteins All N 6 adenine MTases have conserved characteristic motifs such as the AdoMet-binding motif (FXGXG) and the catalytic motif (DPPY). Several research groups have performed mutational studies on amino acids in these motifs, which, in turn, have revealed the significance of these motifs in catalysis [1,3]. For instance, Pues et al. [28] have shown in the case of M.TaqI that replacement of Y108 with alanine or glycine resulted in mutant MTases with reduced enzy- matic activities, which highlights the importance of tyrosine in the methylation activity. It was shown that the replacement of F39 with alanine in the AdoMet- binding motif of M.EcoRV abrogated AdoMet binding [29]. Site-directed mutagenesis was performed to replace F195 and Y32 of HP0050 MTase by serine and lysine, respectively. Both the AdoMet-binding motif (F195S) and the catalytic motif (Y32L) HP0050 mutant proteins were purified to near homogeneity and analyzed on an SDS-polyacrylamide gel for altera- tions in their electrophoretic mobilities. Both mutant proteins fractionated like the wild-type HP0050 protein and no apparent changes were detected. To determine the size and subunit structure of the HP0050 mutant proteins in solution, gel-filtration chromatography was performed and it was found that the mutant proteins eluted as monomers with a molecular mass of 28 kDa (data not shown). Analysis of the wild-type, F195S and Y32L mutants did not reveal significant differ- ences in the CD spectra (data not shown), indicating that the amino acid exchanges did not affect the over- all structure of the mutant proteins. HP0050+AdoMet HP0050-AdoMet Binary complex at 37°C +Duplex 15/16 HP0050-AdoMet-DNA Ternary complex at 4°C + competitor – competitor 0 0 2468 Time (min) Without competitor Without competitor Competitor added at 0 min Competitor added at 0 min Competitor added at 4 min Competitor added at 4 min Competitor added at 8 min Competitor added at 8 min 10 12 14 16 02468 Time (min) 10 12 14 16 100 Methyl groups transfered (fmol) Methyl groups transfered (fmol) 200 300 400 600 400 200 0 AB Fig. 5. Nonprocessive methylation catalyzed by HP0050 MTase. HP0050 MTase (100 nM) was incubated with 5 lM [ 3 H]AdoMet at 37 °C for 10 min to facilitate formation of the HP0050–AdoMet binary complex and then the 2 l M duplex 15 or duplex 16 was added. The mixture was incubated on ice for 5 min to allow the formation of a ternary complex. Then, the mixture was divided into two sets and 40 l M duplex 3 or duplex 14 was added to one set at different time-points ( , 0 min; , 4 min; and ., 8 min) and the other set was allowed to proceed without a DNA trap (d), as described in the Experimental procedures. The reaction was monitored at 2-min time intervals by processing 25 lL of the reaction mixture in duplicate. (A) Duplex 15; (B) duplex 16. R. Kumar et al. N 6 adenine methyltransferase from H. pylori 26695 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS 1673 The methylation activity of both the mutant proteins was analysed as a function of increasing enzyme concentration. It was found that both the mutant proteins were catalytically inactive compared with wild-type HP0050 MTase (Fig. 7A). The loss of activity could be a result of the inability of these mutant proteins to bind to one or both substrates. To investigate the AdoMet binding of the F195S mutant, fluorescence emission spectra and fluorescence intensi- ties were measured in the presence of different concen- trations of AdoMet. The F195S mutant protein showed significantly less quenching in the presence of AdoMet (up to 80 lm) compared with wild-type HP0050 MTase. The K a value for AdoMet was calcu- lated (using a modified Stern-Volmer plot) as 7 lm for the wild-type protein and (using a Stern–Volmer plot) as 64 lm for the F195S mutant (Fig. S4), which is nine times higher than that obtained for the wild-type MTase. This result showed that the F195S mutant was not able to bind to the AdoMet as effectively as the wild-type protein, therefore resulting in the loss of activity. When the Y32L mutant protein was analysed for its AdoMet-binding property, it was found to binds to AdoMet as efficiently as wild-type HP0050 MTase (Fig. S4) but was catalytically inactive (Fig. 7A). DNA distortion induced by wild-type HP0050 MTase upon binding to 2-aminopurine-containing duplexes Most DNA MTases flip the target base within the cog- nate sequence [30]. The fluorescence of 2-aminopurine (2AP) is often used as a signal for base flipping because it shows enhanced fluorescence when its envi- ronment is perturbed. However, it is now well estab- lished that the enhancement of 2AP fluorescence is a more general measure of DNA distortion [31]. To study the change in DNA conformation in the enzyme–DNA complex, we used the 2AP fluorescence- based assay. Irradiation of oligonucleotide (upper strand, duplex 9) containing 2AP at a target base instead of at an adenine base, at 320 nm produced a strong fluorescence emission spectrum with a k max at 375 nm (Fig. 7B). Annealing of this oligonucleotide with the complementary strand resulted in a decrease of approximately threefold in fluorescence intensity at 375 nm. When HP0050 MTase (100 nm) was incubated with 200 nm double-stranded 2AP DNA (duplex 9, Table 2), a fivefold increase in 2AP fluorescence was observed. The increased fluorescence observed upon enzyme binding was more substantial than the fluores- cence of the single-stranded 2AP oligonucleotide. This suggests that the increased fluorescence was not just caused by an enzyme-induced local unwinding of the helix resulting in a region of single-stranded DNA sur- rounding the 2AP, but possibly a result of DNA dis- tortion caused by binding of the protein. The addition of 1 lm sinefungin (an AdoMet analog) resulted in further enhancement of fluorescence. Interestingly, the addition of sinefungin shifted the fluorescence emission spectrum 10 nm towards a longer wavelength. This could be because of a change in the environment of the adenine base upon the addition of sinefungin. By contrast, the Y32L mutant of the HP0050 MTase failed to show any increase in fluorescence, suggesting that, unlike the wild-type MTase, the mutant protein was not able to interact with DNA, and this could be the reason for being catalytically inactive. When the F195S mutant was incubated with double-stranded 2AP DNA (duplex 9, Table 2), an increase in 2AP fluorescence was observed, but the addition of 1 lm sinefungin did not lead to further enhancement of fluorescence. This is in agreement with the observation TaqI AluI TaqI 294 bp 150/146 bp 1.6% Agarose gel AluI 0 0 5 10 15 20 10 20 30 40 50 Taql Alul Time (min) DNA protected from cleavage (%) Fig. 6. Nonprocessive methylation of adjacent adenines in 5¢- GAAG-3¢ by HP0050 MTase. HP0050 MTase (1 l M) was incubated with 5 l M [ 3 H]AdoMet and 3 lM of 294-bp dsDNA at 25°. Two aliquots (each 25 lL) were withdrawn at 3-min intervals up to 15 min and the reactions were stopped by snap-freezing in liquid nitrogen. One of these aliquots was analyzed for protection from digestion with AluI( ) and the other was analyzed for protection from digestion with TaqI(d). N 6 adenine methyltransferase from H. pylori 26695 R. Kumar et al. 1674 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS that the F195S protein does not bind AdoMet. Using the 2AP fluorescence-based assay, enhancements in flu- orescence upon enzyme binding to canonical sequences have been reported with other MTases, such as EcoDam [32] and T4Dam [33]. Distribution of hp0050 in clinical H. pylori isolates The strains of H. pylori (26695, J99 and HPAG1) for which the genome sequence is available were isolated from patients with superficial gastritis, duodenal ulcer and chronic atrophic gastritis, respectively. In the pres- ent study a number of clinical isolates of H. pylori were screened for the presence of hp0050. hp0050 was found to be present in 97.14% of strains obtained from patients [n = 73 (Kolkata strains)] compared with 90.63% of strains from healthy volunteers [n = 32 (Santhali strains)] (data not shown). Primers 3 and 4 (Table 1) were used to amplify hp0050 homo- logs. The functionality of HP0050 MTase in the strains was checked by digestion with MnlI. If a strain has a functional MTase then the genomic DNA will be resis- tant to digestion with MnlI. It was found that all strains which were positive for the presence of hp0050 by PCR were resistant to digestion with MnlI (data not shown). Kolkata strains are H.pylori isolates from patients suffering from ulcer, gastritis or cancer, whereas Santhali strains are isolates from healthy vol- unteers [34]. The hp0050 gene from two clinical isolates (strain PG227 isolated from a patient suffering from duodenal ulcers and strain 128 isolated from a patient with antral gastritis) was cloned into the BamHI and XhoI sites of pET28a, overexpressed and the proteins purified as mentioned in the Experimental procedures. Both were found to be as active as wild-type HP0050 MTase (from H. pylori 26695), and, in the presence of 1 lm sinefungin, which is a competitive inhibitor of all MTases, methylation activity was inhibited by 70%, similarly to the wild-type MTase (data not shown). The hp0050 gene from H. pylori strains PG227 and 128 was sequenced and found to be 89% similar to its homolog from strain 26695 (Fig. 8A). Interestingly, when HP0050 MTase homologs were checked for their specificity, it was found that HP0050 MTase from strains PG227 and 128 methylate GAGG but do not methylate GAAG or GGAG (Fig. 8B). A dot-blot assay was performed to further confirm this observation using duplexes 1, 2 and 3 (Table 2) (Fig. 8C–D). These observations suggest that because of mutations, hp0050 from strain 26695 has evolved relaxed specificity. HP0050 MTase from strain 26695 is able to methylate GAAG and GGAG, whereas its homologs from strains PG227 and 128 lack this speci- ficity as they methylate only GAGG. Because HP0050 is an orphan MTase and lacks a cognate restriction enzyme, it can afford to undergo mutations that result in changed specificity. Isolation and characterization of the Dhp0050 derivative of H. pylori Transcriptional regulation by methylation patterns has been described for a number of prokaryotes, where promoter methylation alters the interaction of regula- Methyl groups transferred (femtomol·min –1 ) Duplex 9 + wild type HP0050 MTase + Sf a Enzyme (n M) Wild type a b Duplex 9 + F195S c Duplex 9 + F195S + Sf d F195S Y32L Duplex 9 + wild type HP0050 MTase d Duplex 9 + Y32L e ss 2AP DNA f g Duplex 9 h Wild type HP0050 + Sf B A 800 Wild type Y32L F195S 550 440 Relative intensity 330 220 110 0 350 400 450 500 Wavelength (nm) 600 400 200 0 020406080 Fig. 7. Characterization of HP0050 MTase Y32L and F195S mutants. (A) Initial velocity versus enzyme concentration. Increasing concentra- tions of wild-type or mutant HP0050 MTase (10–80 n M) were incubated with 80 nM pUC19 and 2.0 lM AdoMet in the presence of 10 mM Tris ⁄ HCl, pH 8.0, containing 5 mM b-mercaptoethanol, at 37 °C for 15 min. The reactions were stopped and analyzed as described in the Experimental procedures. ( ) wild type, (•) Y32L, ( ) F195S. (B) Steady-state fluorescence emission spectra of 2AP-substituted DNA with HP0050 MTase. Spectra were recorded after incubating 100 n M enzyme and 200 nM duplex 9 for 15 min on ice in 10 mM Tris ⁄ HCl, pH 8.0, containing 5 m M b-mercaptoethanol. The total volume of the reaction mixture was 400 lL. Curve a, HP0050 MTase with duplex 9 in the presence of 1 l M sinefungin; curve b, F195S mutant with duplex 9; curve c, F195S mutant with duplex 9 in the presence of 1 lM sinefungin; curve d, HP0050 MTase with duplex 9; curve e, Y32L mutant with duplex 9; curve f, 2AP ssDNA; curve g, duplex 9; curve h, HP0050 MTase with sinefungin. R. Kumar et al. N 6 adenine methyltransferase from H. pylori 26695 FEBS Journal 277 (2010) 1666–1683 ª 2010 The Authors Journal compilation ª 2010 FEBS 1675 [...]... III and IV; lane 2, amplification of the hp0050 locus from a chloramphenicol-resistant colony using primers III and IV; lane 3, 1-kb DNA ladder (C) Lane 1, amplification of the hp0050 locus from a chloramphenicol-resistant colony using primers I and II; lane 2, amplification of the hp0050 locus from a colony of the wild-type strain 227 using primers I and II; lane 3, 1-kb DNA ladder H pylori strain 227 and... compilation ª 2010 FEBS N6 adenine methyltransferase from H pylori 26695 R Kumar et al England Biolabs, USA We thank Dr Anand Swaroop for the methylated oligonucleotides All members of the DNR laboratory are acknowledged for critical reading of the manuscript and useful discussions The work was aided by a grant from the Department of Biotechnology, Government of India, to DNR and AKM We thank the DBT... check the specificity of HP0050 from different strains of H pylori (C) Dot-blot assay of HP0050 from strain 26695 versus HP0050 from strain PG227 (D) Dot-blot assay of HP0050 from strain 26695 versus HP0050 from strain 128 tory proteins with their target DNA [4,5] To investigate the role of the HP0050 MTase in gene regulation, a knockout of hp0050 was constructed in H pylori strain 227, as described in the. .. E H pylori strain 227 H pylori strain 227Δhp0050 Fig 9 Construction and characterization of Helicobacter pylori strain 227Dhp0050 (A) Approximate positions of PCR primers flanking hp0050 used for screening the hp0050 knockout are indicated by arrows Screening the hp0050 knockout by PCR is shown in panels B and C (B) Lane 1, amplification of the hp0050 locus from a colony of the wild-type strain 227 using... the ratio of Vmax ⁄ [E] Similarly, initial velocity experiments were carried out by varying the concentration of [3H]AdoMet in the range of 0.3–2.4 lm while keeping the DNA concentration fixed at 50 nm and keeping other reaction conditions identical The nonlinear regression analysis of initial velocity versus AdoMet concentration allowed the determination of Km (AdoMet) Data were plotted by nonlinear... of wild-type and mutant HP0050 N6 adenine MTases E coli BL21 (DE3) pLysS cells harboring pET28a–hp0050 constructs were grown in 600 mL of LB broth, containing 50 lgÆmL)1 of kanamycin, to an D260 of 0.6 and the expression of HP0050 protein was induced by the addition of IPTG, to a final concentration of 0.5 mm, at 30 °C After 4 h of induction at 30 °C, the culture was cooled on ice and the cells were... volume of v = 0.73 cm3Æg)1 M is the molecular mass of the protein, and the Rtheoh value can be estimated using Eqn (2) where N is the Avogadro constant Equation (2) is based on the assumption of a spherical shape for the investigated molecule The diameter of a water molecule is approximately 0.3 nm By adding this value to the calculated Rtheoh of the protein, hydration can be taken into account The frictional... and H pylori strain 227Dhp0050 growing on brain heart infusion agar (BHIA) plates are shown in panels D and E (D) H pylori strain 227 (E) H pylori strain 227Dhp0050 (dcm)(lon) (DE3) pLysS (camR) cells were used to express wild-type and mutant HP0050 proteins PCR amplification and cloning of the hp0050 gene of H pylori 26695 and other clinical isolates The 699 bp hp0050 gene was amplified from the genomic... regulation of pathogenesis thus makes N6 adenine MTase an interesting target for drug design [38] The lack of adenine methylation in higher eukaryotes has sparked interest in targeting adenine MTases for the development of new antibiotics A detailed understanding of both the structure and mechanism of N6 adenine MTases thus becomes important Experimental procedures Strains and plasmids H pylori 26695. .. DNA of H pylori 26695 by PCR with Pfu polymerase using primers 1 and 2 (Table 1) Primers 3 and 4 were used to amplify the hp0050 homolog from strains PG227 and 128 The primers were designed with the help of the annotated complete genome sequence of H pylori 26695, considering the putative gene sequence of hp0050, obtained from TIGR The amplified PCR fragment was cloned into the bacterial expression vector . Characterization of an N 6 adenine methyltransferase from Helicobacter pylori strain 26695 which methylates adjacent adenines on the same strand Ritesh. modifies the adenine in the top strand of the recognition sequence 5¢-TCC RAC-3¢ and uses modification only on one of the two DNA strands for host protection