High plasticity of multispecific DNA methyltransferases in the region carrying DNA target recognizing enzyme modules

Beta class amino methyltransferases from bacteria to humans: evolution and structural consequences

S-adenosyl-l-methionine dependent methyltransferases catalyze methyl transfers onto a wide variety of target molecules, including DNA and RNA. We discuss a family of methyltransferases, those that act on the amino groups of adenine or cytosine in DNA, have conserved motifs in a particular order in their amino acid sequence, and are referred to as class beta MTases. Members of this class include M.EcoGII and M.EcoP15I from Escherichia coli, Caulobacter crescentus cell cycle-regulated DNA methyltransferase (CcrM), the MTA1-MTA9 complex from the ciliate Oxytricha, and the mammalian MettL3-MettL14 complex. These methyltransferases all generate N6-methyladenine in DNA, with some members having activity on single-stranded DNA as well as RNA. The beta class of methyltransferases has a unique multimeric feature, forming either homo- or hetero-dimers, allowing the enzyme to use division of labor between two subunits in terms of substrate recognition and methylation. We suggest that M.EcoGII may represent an ancestral form of these enzymes, as its activity is independent of the nucleic acid type (RNA or DNA), its strandedness (single or double), and its sequence (aside from the target adenine).

Functional studies of the small subunit of EcoHK31I DNA methyltransferase

EcoHK31I DNA methyltransferase recognizes the sequence 5'-YGGCCR-3' and adds a methyl group to the fifth position of the internal cytosine to protect the DNA from cleavage by its cognate endonuclease. M.EcoHK31I is composed of polypeptides alpha and beta. Polypeptide beta only contains the conserved IX motif of the C5-MTase family, and provides a unique example to show that this motif alone may be dislocated to another polypeptide. By electromobility shift assay, we found that the alpha/beta complex recognizes specific oligonucleotide substrates. Polypeptide alpha formed aggregates with DNA, while polypeptide beta alone did not bind DNA. Therefore, polypeptide beta assists in the proper binding of polypeptide alpha to DNA substrate. The complex of polypeptide alpha and a polypeptide beta variant with an N-terminal deletion of 41 amino acids showed a 16-fold reduction in methylation activity. Further deletion resulted in an inactive methyltransferase. The dissociation equilibrium constant (Kd) of the alpha/beta complex was 56.4 nM, while the Kd value for the alpha/deltaN46-polypeptide beta complex was increased approximately 95-fold, caused by a drastic decrease in dissociate rate constant (kd) and an increase in the association rate constant (ka). This indicates that the N-terminal region of polypeptide beta takes part in subunit interaction, while the C-terminal region is involved in DNA binding.

DNA adenine methyltransferase influences the virulence of Aeromonas hydrophila

Among the various virulence factors produced by Aeromonas hydrophila, a type II secretion system (T2SS)-secreted cytotoxic enterotoxin (Act) and the T3SS are crucial in the pathogenesis of Aeromonas-associated infections. Our laboratory molecularly characterized both Act and the T3SS from a diarrheal isolate, SSU of A. hydrophila, and defined the role of some regulatory genes in modulating the biological effects of Act. In this study, we cloned, sequenced, and expressed the DNA adenine methyltransferase gene of A. hydrophila SSU (dam(AhSSU)) in a T7 promoter-based vector system using Escherichia coli ER2566 as a host strain, which could alter the virulence potential of A. hydrophila. Recombinant Dam, designated as M.AhySSUDam, was produced as a histidine-tagged fusion protein and purified from an E. coli cell lysate using nickel affinity chromatography. The purified Dam had methyltransferase activity, based on its ability to transfer a methyl group from S-adenosyl-l-methionine to N(6)-methyladenine-free lambda DNA and to protect methylated lambda DNA from digestion with DpnII but not against the DpnI restriction enzyme. The dam gene was essential for the viability of the bacterium, and overproduction of Dam in A. hydrophila SSU, using an arabinose-inducible, P(BAD) promoter-based system, reduced the virulence of this pathogen. Specifically, overproduction of M.AhySSUDam decreased the motility of the bacterium by 58%. Likewise, the T3SS-associated cytotoxicity, as measured by the release of lactate dehydrogenase enzyme in murine macrophages infected with the Dam-overproducing strain, was diminished by 55% compared to that of a control A. hydrophila SSU strain harboring the pBAD vector alone. On the contrary, cytotoxic and hemolytic activities associated with Act as well as the protease activity in the culture supernatant of a Dam-overproducing strain were increased by 10-, 3-, and 2.4-fold, respectively, compared to those of the control A. hydrophila SSU strain. The Dam-overproducing strain was not lethal to mice (100% survival) when given by the intraperitoneal route at a dose twice that of the 50% lethal dose, which within 2 to 3 days killed 100% of the animals inoculated with the A. hydrophila control strain. Taken together, our data indicated alteration of A. hydrophila virulence by overproduction of Dam.

Protein fragment complementation in M.HhaI DNA methyltransferase

The 5mC DNA methyltransferase M.HhaI can be split into two individually inactive N- and C-terminal fragments that together can form an active enzyme in vivo capable of efficiently methylating DNA. This active fragment pair was identified by creating libraries of M.HhaI gene fragment pairs and then selecting for the pairs that code for an active 5mC methyltransferase. The site of bisection for successful protein fragment complementation in M.HhaI was in the variable region near the target recognition domain between motif VIII and TRD. This same region is the location of bifurcation in the naturally split 5mC methyltransferase M.AquI, the location for circular permutation in M.BssHII, and the location for previously engineered split versions of M.BspRI.

DNA methyltransferases from Neisseria meningitidis and Neisseria gonorrhoeae FA1090 associated with mismatch nicking endonucleases

The genes encoding the DNA methyltransferases M.NmeDI and M.NmeAI from Neisseria meningitidis associated with the genes encoding putative Vsr endonucleases were overexpressed in Escherichia coli. The enzymes were purified to apparent homogeneity on Ni-NTA agarose columns, yielding proteins of 49+/-1 kDa and 39.6+/-1 kDa, respectively, under denaturing conditions. M.NmeDI recognizes the degenerate sequence 5'-RCCGGB-3'. It methylates the first 5' cytosine residue on both strands within the core sequence CCGG. The enzyme shows higher affinity with the hemimethylated degenerate sequence than with the unmethylated degenerate sequence. Comparison of the amino acid sequence of the target-recognizing domain of M.NmeDI with the closest neighbours recognizing the sequence 5'-RCCGGY-3' showed the presence of the homologous domain and an additional domain that may be responsible for recognizing the degenerate sequence. M.NmeAI recognizes the sequence 5'-CCGG-3' and methylates the second 5' cytosine residue on both DNA strands. In Neisseria gonorrhoeae strain FA1090 the homologues of these ORFs are truncated due to a variety of mutations.

The small subunit of M. AquI is responsible for sequence-specific DNA recognition and binding in the absence of the catalytic domain

AquI DNA methyltransferase (M. AquI) catalyzes the transfer of a methyl group from S-adenosyl-L-methionine to the C5 position of the outermost deoxycytidine base in the DNA sequence 5'-CCCGGG-3'. M. AquI is a heterodimer in which the polypeptide chain is separated at the junction between the two equivalent structural domains in the related enzyme M. HhaI. Recently, we reported the subcloning, overexpression, and purification of the subunits (alpha and beta) of M. AquI separately. Here we describe the DNA binding properties of M. AquI. The results presented here indicate that the beta subunit alone contains all of the information for sequence-specific DNA recognition and binding. The first step in the sequence-specific recognition of DNA by M. AquI involves the formation of binary complex with the target recognition domain in conjunction with conserved sequence motifs IX and X, found in all known C5 DNA methyltransferases, contained in the beta subunit. The alpha subunit enhances the binding of the beta subunit to DNA specifically and nonspecifically. It is likely that the addition of the alpha subunit to the beta subunit stabilizes the conformation of the beta subunit and thereby enhances its affinity for DNA indirectly. Addition of S-adenosyl-L-methionine and its analogues S-adenosyl-L-homocysteine and sinefungin enhances binding, but only in the presence of the alpha subunit. These compounds did not have any effect on DNA binding by the beta subunit alone. Using a 30-mer oligodeoxynucleotide substrate containing 5-fluorodeoxycytidine (5-FdC), it was found that the beta subunit alone did not form a covalent complex with its specific sequence in the absence or presence of S-adenosyl-L-methionine. However, the addition of the alpha subunit to the beta subunit led to the formation of a covalent complex with specific DNA sequence containing 5-FdC.

Plasticity in protein-peptide recognition: crystal structures of two different peptides bound to concanavalin A

The structures of concanavalin A (ConA) in complex with two carbohydrate-mimicking peptides, 10-mer (MYWYPYASGS) and 15-mer (RVWYPYGSYLTASGS) have been determined at 2.75 A resolution. In both crystal structures four independent peptide molecules bind to each of the crystallographically independent subunits of ConA tetramer. The peptides exhibit small but significant variability in conformations and interactions while binding to ConA. The crystal structure of another similar peptide, 12-mer (DVFYPYPYASGS), in complex with ConA has been determined (Jain, D., K. J. Kaur, B. Sundaravadivel, and D. M. Salunke. 2000. Structural and functional consequences of peptide-carbohydrate mimicry. J. Biol. Chem. 275:16098-16102). Comparison of the three complexes shows that the peptides bind to ConA at a common binding site, using different contacting residues and interactions depending on their sequence and the local environment at the binding site. The binding is also optimized by corresponding plasticity of the peptide binding site on ConA. The diversity in conformation and interactions observed here are in agreement with the structural leeway concerning plasticity of specific molecular recognition in biological processes. The adaptability of peptide-ConA interactions may also be correlated with the carbohydrate-mimicking property of these peptides.

Specificity of DNA binding and methylation by the M.FokI DNA methyltransferase

The M.FokI adenine-N(6) DNA methyltransferase recognizes the asymmetric DNA sequence GGATG/CATCC. It consists of two domains each containing all motifs characteristic for adenine-N(6) DNA methyltransferases. We have studied the specificity of DNA-methylation by both domains using 27 hemimethylated oligonucleotide substrates containing recognition sites which differ in one or two base pairs from GGATG or CATCC. The N-terminal domain of M.FokI interacts very specifically with GGATG-sequences, because only one of the altered sites is modified. In contrast, the C-terminal domain shows lower specificity. It prefers CATCC-sequences but only two of the 12 star sites (i.e. sites that differ in 1 bp from the recognition site) are not accepted and some star sites are modified with rates reduced only 2-3-fold. In addition, GGATGC- and CGATGC-sites are modified which differ at two positions from CATCC. DNA binding experiments show that the N-terminal domain preferentially binds to hemimethylated GGATG/C(m)ATCC sequences whereas the C-terminal domain binds to DNA with higher affinity but without specificity. Protein-protein interaction assays show that both domains of M.FokI are in contact with each other. However, several DNA-binding experiments demonstrate that DNA-binding of both domains is mutually exclusive in full-length M.FokI and both domains do not functionally influence each other. The implications of these results on the molecular evolution of type IIS restriction/modification systems are discussed.

Sequence comparison of the EcoHK31I and EaeI restriction-modification systems suggests an intergenic transfer of genetic material

The genes coding for the EcoHK31I and EaeI restriction-modification (R-M) systems from Escherichia coli strain HK31 and Enterobacter aerogenes, respectively, have been cloned and sequenced. Both ENases recognize and cleave Y/GGCCR leaving 4 nucleotide 5'-protruding ends, while the MTases modify the internal cytosine. The systems were isolated on a 2.3kb AseI fragment for EcoHK31I, and a 4.6 kb HindIII fragment for EaeI. The R and M genes of both systems converge and overlap by 14 nucleotides. Previously, we found that M.EcoHK31I consisted of two subunits, (alpha and beta), with the beta subunit being translated from an alternative open reading frame within the gene encoding the alpha subunit. Sequence comparison between the EcoHK31I and EaeI systems reveals striking similarity. The eaeIM gene also encodes alpha and beta polypeptides of 309 and 176 amino acids which share 96% and 97% identity, respectively, with those of ecoHK31IM. ecoHK31IR and eaeIR encode proteins of 318 and 315 aa, respectively, which share 92% identity but are otherwise unique in the GenBank database. The EaeI and the EcoHK31I R-M systems were found to be flanked by genes coding for integrases. It is possible that these integrases have facilitated the transfer of this system among different bacterial species.

Cloning and characterization of the gene encoding PspPI methyltransferase from the Antarctic psychrotroph Psychrobacter sp. strain TA137. Predicted interactions with DNA and organization of the variable region

The gene (pspPIM) encoding the PspPI DNA methyltransferase (MTase) associated with the PspPI restriction-modification (R-M) system (5'-GGNCC-3') of Psychrobacter species TA137 has been cloned and expressed in E. coli, and its nucleotide (nt) sequence has been determined. The coding region was 1248 nt in length and capable of specifying a 46826-Da protein of 416 amino acids (aa). The predicted sequence of the MTase protein displays ten sequence motifs characteristic of all prokaryotic m5C-MTases and shows the highest similarity to other MTases that methylate the GGNCC sequence, namely M . Eco47II and M . Sau96I. All three MTases methylate the internal cytosine within their recognition sequence. Sequence similarities between M . PspPI and its isospecific M . Eco47II and M . Sau96I as well as with four other m5C-MTases that methylate the related GGWCC sequence, namely M . SinI, M . HgiCII, M . HgiBI, M . HgiEI have been also found within the variable region of these proteins. On the basis of structural information from M . HhaI and M . HaeIII, several M . PspPI residues that are expected to interact with DNA can be predicted. Furthermore, an organization of the variable region of m5C-MTases into two segments exhibiting a pattern of conserved residues and a considerable degree of structural homologies is described.

Identification of a subdomain within DNA-(cytosine-C5)-methyltransferases responsible for the recognition of the 5' part of their DNA target

In previous work on DNA-(cytosine-C5)-methyltransferases (C5-MTases), domains had been identified which are responsible for the sequence specificity of the different enzymes (target-recognizing domains, TRDs). Here we have analyzed the DNA methylation patterns of two C5-MTases containing reciprocal chimeric TRDs, consisting of the N- and C-terminal parts derived from two different parental TRDs specifying the recognition of 5'-CC(A/T)GG-3' and 5'-GCNGC-3'. Sequences recognized by these engineered MTases were non-symmetrical and degenerate, but contained at their 5' part a consensus sequence which was very similar to the 5' part of the target recognized by the parental TRD which contributed the N-terminal moiety of the chimeric TRD. The results are discussed in connection with the present understanding of the mechanism of DNA target recognition by C5-MTases. They demonstrate the possibility of designing C5-MTases with novel DNA methylation specificities.

Exact size and organization of DNA target-recognizing domains of multispecific DNA-(cytosine-C5)-methyltransferases

A large portion of the sequences of type II DNA-(cytosine-C5)-methyltransferases (C5-MTases) represent highly conserved blocks of amino acids. General steps in the methylation reaction performed by C5-MTases have been found to be mediated by some of these domains. C5-MTases carry, in addition at the same relative location, a region variable in size and amino acid composition, part of which is associated with the capacity of each C5-MTase to recognize its characteristic target. Individual target-recognizing domains (TRDs) for the targets CCGG (M), CC(A/T)GG (E), GGCC (H), GCNGC (F) and G(G/A/T)GC(C/A/T)C (B) could be identified in the C-terminal part of the variable region of multispecific C5-MTases. With experiments reported here, we have established the organization of the variable regions of the multispecific MTases M.SPRI, M.phi3TI, M.H2I and M.rho 11SI at the resolution of individual amino acids. These regions comprise 204, 175, 268 and 268 amino acids, respectively. All variable regions are bipartite. They contain at their N-terminal side a very similar sequence of 71 amino acids. The integrity of this sequence must be assured to provide enzyme activity. Bracketed by 6-10 'linker' amino acids, they have, depending on the enzyme studied, towards their C-terminal end ensembles of individual TRDs of 38 (M), 39 (E), 40 (H), 44 (F) and 54 (B) amino acids. TRDs of different enzymes with equal specificity have the same size. TRDs do not overlap but are either separated by linker amino acids or abut each other.

Altered sequence recognition specificity of a C5-DNA methyltransferase carrying a chimeric 'target recognizing domain'

A MTase with a chimeric TRD with the N-terminal half derived from a TRD recognizing GCNGC, the C-terminal half from one with CCWGG recognition, was constructed. Its target specificity is reported.

CpNpG methylation in mammalian cells

In vertebrate DNA, 3% to 5% of cytosine residues are present as 5-methylcytosine, and it is generally accepted that essentially all of this methylation occurs at cytosines which are contained in the symmetrical dinucleotide CpG. In this report we demonstrate, using bisulphite genomic sequencing, that the methylation machinery of mammalian cells is capable of both maintenance and de novo methylation at CpNpG sites. The existence of inherited CpNpG methylation in mammalian cells has important implications in gene regulation and in the aetiology of disease.

DNA modification by methyltransferases

Enzymatic methylation of DNA plays important roles in both prokaryotes and eukaryotes. Structural study of the HhaI DNA methyltransferase has provided considerable insight into the chemistry of C5-cytosine methylation. The DNA-protein complex reveals a substrate cytosine flipped out of the double helix during the reaction, and a novel two-loop DNA-binding motif used for both sequence recognition and flipping the base. Structural comparison of HhaI C5-cytosine methyltransferase, TaqI N6-adenine methyltransferase, and catechol O-methyltransferase reveals a common catalytic domain structure, which might be universal among S-adenosyl-L-methionine (SAM)-dependent methyltransferases.

Identification of the XorII methyltransferase gene and a vsr homolog from Xanthomonas oryzae pv. oryzae

The gene encoding the XorII methyltransferase (M.XorII) was cloned from Xanthomonas oryzae pv. oryzae and characterized in Escherichia coli. The M.XorII activity was localized to a 3.1 kb BamHI-BstXI fragment, which contained two open reading frames (ORFs) of 1272 nucleotides (424 amino acids) and 408 nucleotides (136 amino acids). Ten polypeptide domains conserved in other M5 cytosine methyltransferases (MTases) were identified in the deduced amino acid sequence of the 1272 ORF. E. coli Mrr+ strains were transformed poorly by plasmids containing the XorII MTase gene, indicating the presence of at least one MCG in the recognition sequence for M.XorII (CGATCG). The 408 nucleotide ORF was 36% identical at the amino acid level to sequences of the E. coli dem-vsr gene, which is required for very short patch repair. X. oryzae pv. oryzae genomic DNA that is resistant to digestion by PvuI and XorII hybridizes with a 7.0 kb fragment containing the XorII MTase gene and vsr homolog, whereas DNA from strains that lack M.XorII activity do not hybridize with the fragment.

The NlaIV restriction and modification genes of Neisseria lactamica are flanked by leucine biosynthesis genes

The genes encoding the Neisseria lactamica restriction endonuclease IV (R.NlaIV) and its cognate DNA methyltransferase (M.NlaIV), both of which recognize the sequence GGNNCC, have been cloned in Escherichia coli and overexpressed using the T7 polymerase/promoter system. Analysis of a sequenced 3.58 kb fragment established the gene order, leuD-M.NlaIV-R.NlaIV-leuB. The predicted primary sequence of M.NlaIV (423 amino acids) shows the highest degree of identity to a pair of cytosine-specific methyltransferases, M.BanI (44.9%) and M.HgiCI (44.3%), which recognize the sequence GGYRCC (Y, pyrimidines; R, purines). In contrast, the R.NlaIV protein sequence (243 amino acids) is unique in the existing data-base, a situation that holds for most endonucleases. Flanking the NlaIV modification and restriction genes are homologues of the leuD and leuB genes of enteric bacteria, which code for enzymes in the leucine biosynthesis pathway. This gene context implies a possible new mode of gene regulation for the RM.NlaIV system, which would involve a mechanism similar to the recently discovered leucine/Lrp regulon in E. coli.