EcoA and EcoE: alternatives to the EcoK family of type I restriction and modification systems of Escherichia coli

Characterization of a restriction modification system from the commensal Escherichia coli strain A0 34/86 (O83:K24:H31)

Background

Type I restriction-modification (R-M) systems are the most complex restriction enzymes discovered to date. Recent years have witnessed a renaissance of interest in R-M enzymes Type I. The massive ongoing sequencing programmes leading to discovery of, so far, more than 1 000 putative enzymes in a broad range of microorganisms including pathogenic bacteria, revealed that these enzymes are widely represented in nature. The aim of this study was characterisation of a putative R-M system EcoA0ORF42P identified in the commensal Escherichia coli A0 34/86 (O83: K24: H31) strain, which is efficiently used at Czech paediatric clinics for prophylaxis and treatment of nosocomial infections and diarrhoea of preterm and newborn infants.

Results

We have characterised a restriction-modification system EcoA0ORF42P of the commensal Escherichia coli strain A0 34/86 (O83: K24: H31). This system, designated as EcoAO83I, is a new functional member of the Type IB family, whose specificity differs from those of known Type IB enzymes, as was demonstrated by an immunological cross-reactivity and a complementation assay. Using the plasmid transformation method and the RM search computer program, we identified the DNA recognition sequence of the EcoAO83I as GGA(8N)ATGC. In consistence with the amino acids alignment data, the 3' TRD component of the recognition sequence is identical to the sequence recognized by the EcoEI enzyme. The A-T (modified adenine) distance is identical to that in the EcoAI and EcoEI recognition sites, which also indicates that this system is a Type IB member. Interestingly, the recognition sequence we determined here is identical to the previously reported prototype sequence for Eco377I and its isoschizomers.

Conclusion

Putative restriction-modification system EcoA0ORF42P in the commensal Escherichia coli strain A0 34/86 (O83: K24: H31) was found to be a member of the Type IB family and was designated as EcoAO83I. Combination of the classical biochemical and bacterial genetics approaches with comparative genomics might contribute effectively to further classification of many other putative Type-I enzymes, especially in clinical samples.

Phosphorylation of Type IA restriction-modification complex enzyme EcoKI on the HsdR subunit

Phosphorylation of Type I restriction-modification (R-M) enzymes EcoKI, EcoAI, and EcoR124I - representatives of IA, IB, and IC families, respectively - was analysed in vivo by immunoblotting of endogenous phosphoproteins isolated from Escherichia coli strains harbouring the corresponding hsd genes, and in vitro by a phosphorylation assay using protein kinase present in subcellular fractions of E. coli. From all three R-M enzymes, the HsdR subunit of EcoKI system was the only subunit that was phosphorylated. Further, evidence is presented that HsdR is phosphorylated in vivo only when coproduced with HsdM and HsdS subunits - as part of assembled EcoKI restriction endonuclease, while the individually produced HsdR subunit is not phosphorylated. In vitro phosphorylation of the HsdR subunit of purified EcoKI endonuclease occurs on Thr, and is strictly dependent on the addition of a catalytic amount of cytoplasmic fraction isolated from E. coli. So far this is the first case of phosphorylation of a Type I R-M enzyme reported.

Diversity within the Campylobacter jejuni type I restriction-modification loci

The type I restriction-modification (hsd) systems of 73 Campylobacter jejuni strains were characterized according to their DNA and amino acid sequences, and/or gene organization. A number of new genes were identified which are not present in the sequenced strain NCTC 11168. The closely related organism Helicobacter pylori has three type I systems; however, no evidence was found that C. jejuni strains contain multiple type I systems, although hsd loci are present in at least two different chromosomal locations. Also, unlike H. pylori, intervening ORFs are present, in some strains, between hsdR and hsdS and between hsdS and hsdM. No definitive function can be ascribed to these ORFs, designated here as rloA-H (R-linked ORF) and mloA-B (M-linked ORF). Based on parsimony analysis of amino acid sequences to assess character relatedness, the C. jejuni type I R-M systems are assigned to one of three families: 'IAB', 'IC' or 'IF'. This study confirms that HsdM proteins within a family are highly conserved but share little homology with HsdM proteins from other families. The 'IC' hsd loci are >99 % identical at the nucleotide level, as are the 'IF' hsd loci. Additionally, whereas the nucleotide sequences of the 'IAB' hsdR and hsdM genes show a high degree of similarity, the nucleotide sequences of the 'IAB' hsdS and rlo genes vary considerably. This diversity suggests that recombination between 'IAB' hsd loci would lead not only to new hsdS alleles but also to the exchange of rlo genes; five C. jejuni hsd loci are presumably the result of such recombination. The importance of these findings with regard to the evolution of C. jejuni type I R-M systems is discussed.

KpnBI is the prototype of a new family (IE) of bacterial type I restriction-modification system

KpnBI is a restriction-modification (R-M) system recognized in the GM236 strain of Klebsiella pneumoniae. Here, the KpnBI modification genes were cloned into a plasmid using a modification expression screening method. The modification genes that consist of both hsdM (2631 bp) and hsdS (1344 bp) genes were identified on an 8.2 kb EcoRI chromosomal fragment. These two genes overlap by one base and share the same promoter located upstream of the hsdM gene. Using recently developed plasmid R-M tests and a computer program RM Search, the DNA recognition sequence for the KpnBI enzymes was identified as a new 8 nt sequence containing one degenerate base with a 6 nt spacer, CAAANNNNNNRTCA. From Dam methylation and HindIII sensitivity tests, the methylation loci were predicted to be the italicized third adenine in the 5' specific region and the adenine opposite the italicized thymine in the 3' specific region. Combined with previous sequence data for hsdR, we concluded that the KpnBI system is a typical type I R-M system. The deduced amino acid sequences of the three subunits of the KpnBI system show only limited homologies (25 to 33% identity) at best, to the four previously categorized type I families (IA, IB, IC, and ID). Furthermore, their identity scores to other uncharacterized putative genome type I sequences were 53% at maximum. Therefore, we propose that KpnBI is the prototype of a new 'type IE' family.

Cellular localization of Type I restriction-modification enzymes is family dependent

Cellular localization of Type I restriction-modification enzymes EcoKI, EcoAI, and EcoR124I-the most frequently studied representatives of IA, IB, and IC families-was analyzed by immunoblotting of subcellular fractions isolated from Escherichia coli strains harboring the corresponding hsd genes. EcoR124I shows characteristics similar to those of EcoKI. The complex enzymes are associated with the cytoplasmic membrane via DNA interaction as documented by the release of the Hsd subunits from the membrane into the soluble fraction following benzonase treatment. HsdR subunits of the membrane-bound enzymes EcoKI and EcoR124I are accessible, though to a different extent, at the external surface of cytoplasmic membrane as shown by trypsinization of intact spheroplasts. EcoAI strongly differs from EcoKI and EcoR124I, since neither benzonase nor trypsin affects its association with the cytoplasmic membrane. Possible reasons for such a different organization are discussed in relation of the control of the restriction-modification activities in vivo.

Tetra-amino-acid tandem repeats are involved in HsdS complementation in type IC restriction-modification systems

All known type I restriction and modification (R-M) systems of Escherichia coli and Salmonella enterica belong to one of four discrete families: type IA, IB, IC or ID. The classification of type I systems from a wide range of other genera is mainly based on complementation and molecular evidence derived from the comparison of the amino acid similarity of the corresponding subunits. This affiliation was seldom based on the strictest requirement for membership of a family, which depends on relatedness as demonstrated by complementation tests. This paper presents data indicating that the type I NgoAV R-M system from Neisseria gonorrhoeae, despite the very high identity of HsdM and HsdR subunits with members of the type IC family, does not show complementation with E. coli type IC R-M systems. Sequence analysis of the HsdS subunit of several different potential type IC R-M systems shows that the presence of different tetra-amino-acid sequence repeats, e.g. TAEL, LEAT, SEAL, TSEL, is characteristic for type IC R-M systems encoded by distantly related bacteria. The other regions of the HsdS subunits potentially responsible for subunit interaction are also different between a group of distantly related bacteria, but show high similarity within these bacteria. Complementation between the NgoAV R-M system and members of the EcoR124 R-M family can be restored by changing the tetra-amino-acid repeat within the HsdS subunit. The authors propose that the type IC family of R-M systems could consist of several complementation subgroups whose specificity would depend on differences in the conserved regions of the HsdS polypeptide.

Structure of Ocr from bacteriophage T7, a protein that mimics B-form DNA

We have solved, by X-ray crystallography to a resolution of 1.8 A, the structure of a protein capable of mimicking approximately 20 base pairs of B-form DNA. This ocr protein, encoded by gene 0.3 of bacteriophage T7, mimics the size and shape of a bent DNA molecule and the arrangement of negative charges along the phosphate backbone of B-form DNA. We also demonstrate that ocr is an efficient inhibitor in vivo of all known families of the complex type I DNA restriction enzymes. Using atomic force microscopy, we have also observed that type I enzymes induce a bend in DNA of similar magnitude to the bend in the ocr molecule. This first structure of an antirestriction protein demonstrates the construction of structural mimetics of long segments of B-form DNA.

Domain structure and subunit interactions in the type I DNA methyltransferase M.EcoR124I

The type IC DNA methyltransferase M.EcoR124I is a trimeric enzyme of 162 kDa consisting of two modification subunits, HsdM, and a single specificity subunit, HsdS. Studies have been largely restricted to the HsdM subunit or to the intact methyltransferase since the HsdS subunit is insoluble when over-expressed independently of HsdM. Two soluble fragments of the HsdS subunit have been cloned, expressed and purified; a 25 kDa N-terminal fragment (S3) comprising the N-terminal target recognition domain together with the central conserved domain, and a 8.6 kDa fragment (S11) comprising the central conserved domain alone. Analytical ultracentrifugation shows that the S3 subunit exists principally as a dimer of 50 kDa. Gel retardation and competition assays show that both S3 and S11 are able to bind to HsdM, each with a subunit stoichiometry of 1:1. The tetrameric complex (S3/HsdM)(2) is required for effective DNA binding. Cooperative binding is observed and at low enzyme concentration, the multisubunit complex dissociates, leading to a loss of DNA binding activity. The (S3/HsdM)(2) complex is able to bind to both the EcoR124I DNA recognition sequence GAAN(6)RTCG and a symmetrical DNA sequence GAAN(7)TTC, but has a 30-fold higher affinity binding for the latter DNA sequence. Exonuclease III footprinting of the (S3/HsdM)(2) -DNA complex indicates that 29 nucleotides are protected on each strand, corresponding to a region 8 bp on both the 3' and 5' sides of the recognition sequence bound by the (S3/HsdM)(2) complex.

Families of restriction enzymes: an analysis prompted by molecular and genetic data for type ID restriction and modification systems

Current genetic and molecular evidence places all the known type I restriction and modification systems of Escherichia coli and Salmonella enterica into one of four discrete families: type IA, IB, IC or ID. StySBLI is the founder member of the ID family. Similarities of coding sequences have identified restriction systems in E.coli and Klebsiella pneumoniae as probable members of the type ID family. We present complementation tests that confirm the allocation of EcoR9I and KpnAI to the ID family. An alignment of the amino acid sequences of the HsdS subunits of StySBLI and EcoR9I identify two variable regions, each predicted to be a target recognition domain (TRD). Consistent with two TRDs, StySBLI was shown to recognise a bipartite target sequence, but one in which the adenine residues that are the substrates for methylation are separated by only 6 bp. Implications of family relationships are discussed and evidence is presented that extends the family affiliations identified in enteric bacteria to a wide range of other genera.

Analysis of type I restriction modification systems in the Neisseriaceae: genetic organization and properties of the gene products

The hsd locus (host specificity of DNA) was identified in the Neisseria gonorrhoeae genome. The DNA fragment encoding this locus produced an active restriction and modification (R/M) system when cloned into Escherichia coli. This R/M system was designated NgoAV. The cloned genomic fragment (7800 bp) has the potential to encode seven open reading frames (ORFs). Several of these ORFs had significant homology with other proteins found in the databases: ORF1, the hsdM, a methylase subunit (HsdM); ORF2, a homologue of dinD; ORF3, a homologue of hsdS; ORF4, a homologue of hsdS; and ORF5, an endonuclease subunit hsdR. The endonuclease and methylase subunits possessed strongest protein sequence homology to the EcoR124II R/M system, indicating that NgoAV belongs to the type IC R/M family. Deletion analysis showed that only ORF3 imparted the sequence specificity of the RM.NgoAV system, which recognizes an interrupted palindrome sequence (GCAN(8-)TGC). The genetic structure of ORF3 (208 amino acids) is almost identical to the structure of the 5' truncated hsdS genes of EcoDXXI or EcoR124II R/M systems obtained by in vitro manipulation. Genomic sequence analysis allowed us to identify hsd loci with a very high homology to RM.NgoAV in two strains of Neisseria meningitidis. However, significant differences in the organization and structure of the hsdS genes in both these systems suggests that, if functional, they would possess recognition sites that differ from the gonococcus and from themselves.

Target recognition by EcoKI: the recognition domain is robust and restriction-deficiency commonly results from the proteolytic control of enzyme activity

We report a genetic and biochemical analysis of a target recognition domain (TRD) of EcoKI, a type I restriction and modification enzyme. The TRDs of type I R-M systems are within the specificity subunit (HsdS) and HsdS confers sequence specificity to a complex endowed with both restriction and modification activities. Random mutagenesis has revealed that most substitutions within the amino TRD of EcoKI, a region comprising 157 amino acid residues, have no detectable effect on the phenotype of the bacterium, even when the substitutions are non- conservative. The structure of the TRD appears to be robust. All but one of the six substitutions that confer a restriction-deficient, modification-deficient (r(-)m(-)) phenotype were found to be in the interval between residues 80 and 110, a region predicted by sequence comparisons to form part of the protein-DNA interface. Additional site-directed mutations affecting this interval commonly impair both restriction and modification. However, we show that an r(-) phenotype cannot be taken as evidence that the EcoKI complex lacks endonuclease activity; in response to even a slightly impaired modification efficiency, the endonuclease activity of EcoKI is destroyed by a process dependent upon the ClpXP protease. Enzymes from mutants with an r(-)m(-) phenotype commonly retain some sequence-specific activity; methylase activity can be detected on hemimethylated DNA substrates and residual endonuclease activity is implied whenever the viability of the r(-)m(-) bacterium is dependent on ClpXP. Conversely, the viability of ClpX(-) r(-)m(-) bacteria can be used as evidence for little, or no, endonuclease activity. Of 14 mutants with an r(-)m(-) phenotype, only six are viable in the absence of ClpXP. The significance of four of the six residues (G91, G105, F107 and G141) is enhanced by the finding that even conservative substitutions for these residues impair modification, thereby conferring an r(-)m(-) phenotype.

The proteolytic control of restriction activity in Escherichia coli K-12

The endonuclease activity of EcoKI is regulated by the ClpXP-dependent degradation of the subunit that is essential for restriction, but not modification. We monitored proteolysis in mutants blocked at different steps in the restriction pathway. Mutations that prevent DNA translocation render EcoKI refractory to proteolysis, whereas those that permit DNA translocation, but block endonuclease activity, do not. Although proteolysis alleviates restriction in a mutant that lacks modification activity, some restriction activity remains; our evidence indicates residual EcoKI associated with the membrane fraction. ClpXP protects the bacterial chromosome, but little effect was detected on unmodified foreign DNA within the cytoplasm of a restriction-proficient cell. The molecular basis for the distinction between unmodified resident and foreign DNA remains to be determined.

ATP-dependent restriction enzymes

The phenomenon of restriction and modification (R-M) was first observed in the course of studies on bacteriophages in the early 1950s. It was only in the 1960s that work of Arber and colleagues provided a molecular explanation for the host specificity. DNA restriction and modification enzymes are responsible for the host-specific barriers to interstrain and interspecies transfer of genetic information that have been observed in a variety of bacterial cell types. R-M systems comprise an endonuclease and a methyltransferase activity. They serve to protect bacterial cells against bacteriophage infection, because incoming foreign DNA is specifically cleaved by the restriction enzyme if it contains the recognition sequence of the endonuclease. The DNA is protected from cleavage by a specific methylation within the recognition sequence, which is introduced by the methyltransferase. Classic R-M systems are now divided into three types on the basis of enzyme complexity, cofactor requirements, and position of DNA cleavage, although new systems are being discovered that do not fit readily into this classification. This review concentrates on multisubunit, multifunctional ATP-dependent restriction enzymes. A growing number of these enzymes are being subjected to biochemical and genetic studies that, when combined with ongoing structural analyses, promise to provide detailed models for mechanisms of DNA recognition and catalysis. It is now clear that DNA cleavage by these enzymes involves highly unusual modes of interaction between the enzymes and their substrates. These unique features of mechanism pose exciting questions and in addition have led to the suggestion that these enzymes may have biological functions beyond that of restriction and modification. The purpose of this review is to describe the exciting developments in our understanding of how the ATP-dependent restriction enzymes recognize specific DNA sequences and cleave or modify DNA.

Regulation of endonuclease activity by proteolysis prevents breakage of unmodified bacterial chromosomes by type I restriction enzymes

ClpXP-dependent proteolysis has been implicated in the delayed detection of restriction activity after the acquisition of the genes (hsdR, hsdM, and hsdS) that specify EcoKI and EcoAI, representatives of two families of type I restriction and modification (R-M) systems. Modification, once established, has been assumed to provide adequate protection against a resident restriction system. However, unmodified targets may be generated in the DNA of an hsd(+) bacterium as the result of replication errors or recombination-dependent repair. We show that ClpXP-dependent regulation of the endonuclease activity enables bacteria that acquire unmodified chromosomal target sequences to survive. In such bacteria, HsdR, the polypeptide of the R-M complex essential for restriction but not modification, is degraded in the presence of ClpXP. A mutation that blocks only the modification activity of EcoKI, leaving the cell with approximately 600 unmodified targets, is not lethal provided that ClpXP is present. Our data support a model in which the HsdR component of a type I restriction endonuclease becomes a substrate for proteolysis after the endonuclease has bound to unmodified target sequences, but before completion of the pathway that would result in DNA breakage.

Single amino acid substitutions in the HsdR subunit of the type IB restriction enzyme EcoAI uncouple the DNA translocation and DNA cleavage activities of the enzyme

Type I restriction enzymes bind to specific DNA sequences but subsequently translocate non-specific DNA past the complex in a reaction coupled to ATP hydrolysis and cleave DNA at any barrier that can halt the translocation process. The restriction subunit of these enzymes, HsdR, contains a cluster of seven amino acid sequence motifs typical of helicase superfamily II, that are believed to be relevant to the ATP-dependent DNA translocation. Alignment of all available HsdR sequences reveals an additional conserved region at the protein N-terminus with a consensus sequence reminiscent of the P-D.(D/E)-X-K catalytic motif of many type II restriction enzymes. To investigate the role of these conserved residues, we have produced mutants of the type IB restriction enzyme Eco AI. We have found that single alanine substitutions at Asp-61, Glu-76 and Lys-78 residues of the HsdR subunit abolished the enzyme's restriction activity but had no effect on its ATPase and DNA translocation activities, suggesting that these residues are part of the active site for DNA cleavage.

The DNA recognition subunit of the type IB restriction-modification enzyme EcoAI tolerates circular permutions of its polypeptide chain

The DNA specificity subunit (HsdS) of type I restriction-modification enzymes is composed of two independent target recognition domains and several regions whose amino acid sequence is conserved within an enzyme family. The conserved regions participate in intersubunit interactions with two modification subunits (HsdM) and two restriction subunits (HsdR) to form the complete endonuclease. It has been proposed that the domains of the HsdS subunit have a circular organisation providing the required symmetry for their interaction with the other subunits and with the bipartite DNA target. To test this model, we circularly permuted the HsdS subunit of the type IB R-M enzyme EcoAI at the DNA level by direct linkage of codons for original termini and introduction of new termini elsewhere along the N-terminal and central conserved regions. By analysing the activity of mutant enzymes, two circularly permuted variants of HsdS that had termini located at equivalent positions in the N-terminal and central repeats, respectively, were found to fold into a functional DNA recognition subunit with wild-type specificity, suggesting a close proximity of the N and C termini in the native protein. The wild-type HsdS subunit was purified to homogeneity and shown to form a stable trimeric complex with HsdM, M2S1, which was fully active as a DNA methyltransferase. Gel electrophoretic mobility shift assays revealed that the HsdS protein alone was not able to form a specific complex with a 30-mer oligoduplex containing a single EcoAI recognition site. However, addition of stoichiometric amounts of HsdM to HsdS led to efficient specific DNA binding. Our data provide evidence for the circular organisation of domains of the HsdS subunit. In addition, they suggest a possible role of HsdM subunits in the formation of this structure.

Localization of a protein-DNA interface by random mutagenesis

The type I restriction and modification enzymes do not possess obvious DNA-binding motifs within their target recognition domains (TRDs) of 150-180 amino acids. To identify residues involved in DNA recognition, changes were made in the amino-TRD of EcoKI by random mutagenesis. Most of the 101 substitutions affecting 79 residues had no effect on the phenotype. Changes at only seven positions caused the loss of restriction and modification activities. The seven residues identified by mutation are not randomly distributed throughout the primary sequence of the TRD: five are within the interval between residues 80 and 110. Sequence analyses have led to the suggestion that the TRDs of type I restriction enzymes include a tertiary structure similar to the TRD of the HhaI methyltransferase, and to a model for a DNA-protein interface in EcoKI. In this model, the residues within the interval identified by the five mutations are close to the protein-DNA interface. Three additional residues close to the DNA in the model were changed; each substitution impaired both activities. Proteins from twelve mutants were purified: six from mutants with partial or wild-type activity and six from mutants lacking activity. There is a strong correlation between phenotype and DNA-binding affinity, as determined by fluorescence anisotropy.

ClpX and ClpP are essential for the efficient acquisition of genes specifying type IA and IB restriction systems

Efficient acquisition of genes that encode a restriction and modification (R-M) system with specificities different from any already present in the recipient bacterium requires the sequential production of the new modification enzyme followed by the restriction activity in order that the chromosome of the recipient bacterium is protected against attack by the restriction endonuclease. We show that ClpX and ClpP, the components of ClpXP protease, are necessary for the efficient transmission of the genes encoding EcoKI and EcoAI, representatives of two families of type I R-M systems, thus implicating ClpXP in the modulation of restriction activity. Loss of ClpX imposed a bigger barrier than loss of ClpP, consistent with a dual role for ClpX, possibly as a chaperone and as a component of the ClpXP protease. Transmission of genes specifying EcoKI was more dependent on ClpX and ClpP than transmission of the genes for EcoAI. Sensitivity to absence of the protease was also influenced by the mode of gene transfer; conjugative transfer and transformation were more dependent on ClpXP than transduction. In the absence of either ClpX or ClpP transfer of the EcoKI genes by P1-mediated transduction was impaired, transfer of the EcoAI genes was not.

Protein-protein and protein-DNA interactions in the type I restriction endonuclease R.EcoR124I

The type I restriction-modification system EcoR124I recognizes and binds to the split DNA recognition sequence 5'-GAAN(6)RTCG-3'. The methyltransferase, consisting of HsdM and HsdS subunits with the composition M2S, can interact with one or more subunits of the HsdR subunit to form the endonuclease. The interaction of the methyltransferase with HsdR has been investigated by surface plasmon resonance, showing that there are two non-equivalent binding sites for HsdR which differ in binding affinity by at least two orders of magnitude. DNA footprinting experiments using Exonuclease III suggest that the addition of HsdR to the methyltransferase (at a stoichiometry of either 1:1 or 2:1) increases the stability of the resulting DNA-protein complex but does not increase the size of the footprint. More extensive in situ footprinting experiments using copper-phenanthroline on the DNA-protein complexes formed by M2S, R1M2S and R2M2S also show no difference in the detailed cleavage pattern, with approximately 18 nucleotides protected on both strands in each complex. Thus the HsdR subunit(s) of the endonuclease stabilise the interaction of the M2S complex with DNA, but do not directly contribute to DNA binding. In addition, the thymidine nucleotide in the tetranucleotide recognition sequence GTCG is hyper-reactive to cleavage in each case, suggesting that the DNA structure in this region is altered in these complexes.

Expression and characterisation of the N-terminal fragment of the HsdS subunit of M.EcoR124I

The type IC modification methyltransferase M.EcoR124I is a trimeric enzyme of 162 kDa consisting of two copies of the modification subunit, HsdM, and a single DNA specificity subunit, HsdS. Studies to date have been largely restricted to the HsdM subunit or the intact methyltransferase, since the HsdS subunit is insoluble when expressed independently of HsdM. Using PCR, we have cloned and expressed 13 fragments of the gene for the HsdS subunit, including the sequences encoding each of the variable and conserved domains and various combinations of these. Only two of these fragments were found to be soluble, a 8.6 kDa fragment (S11) comprising the central conserved domain and a 25 kDa N-terminal fragment (S3) containing the N-terminal variable domain and the central conserved domain. Analysis of the larger of these fragments by gel retardation shows that the protein binds DNA in the presence of HsdM at a subunit stoichiometry of 1:1. Gel filtration and CD spectroscopy indicate that the protein is monomeric and predominantly alpha-helical.

The restriction-modification genes of Escherichia coli K-12 may not be selfish: they do not resist loss and are readily replaced by alleles conferring different specificities

Type II restriction and modification (R-M) genes have been described as selfish because they have been shown to impose selection for the maintenance of the plasmid that encodes them. In our experiments, the type I R-M system EcoKI does not behave in the same way. The genes specifying EcoKI are, however, normally residents of the chromosome and therefore our analyses were extended to monitor the deletion of chromosomal genes rather than loss of plasmid vector. If EcoKI were to behave in the same way as the plasmid-encoded type II R-M systems, the loss of the relevant chromosomal genes by mutation or recombination should lead to cell death because the cell would become deficient in modification enzyme and the bacterial chromosome would be vulnerable to the restriction endonuclease. Our data contradict this prediction; they reveal that functional type I R-M genes in the chromosome are readily replaced by mutant alleles and by alleles encoding a type I R-M system of different specificity. The acquisition of allelic genes conferring a new sequence specificity, but not the loss of the resident genes, is dependent on the product of an unlinked gene, one predicted [Prakash-Cheng, A., Chung, S. S. & Ryu, J. (1993) Mol. Gen. Genet. 241, 491-496] to be relevant to control of expression of the genes that encode EcoKI. Our evidence suggests that not all R-M systems are evolving as "selfish" units; rather, the diversity and distribution of the family of type I enzymes we have investigated require an alternative selective pressure.

The specificity of sty SKI, a type I restriction enzyme, implies a structure with rotational symmetry

The type I restriction and modification (R-M) enzyme from Salmonella enterica serovar kaduna ( Sty SKI) recognises the DNA sequence 5'-CGAT(N)7GTTA, an unusual target for a type I R-M system in that it comprises two tetranucleotide components. The amino target recognition domain (TRD) of Sty SKI recognises 5'-CGAT and shows 36% amino acid identity with the carboxy TRD of Eco R124I which recognises the complementary, but degenerate, sequence 5'-RTCG. Current models predict that the amino and carboxy TRDs of the specificity subunit are in inverted orientations within a structure with 2-fold rotational symmetry. The complementary target sequences recognised by the amino TRD of Sty SKI and the carboxy TRD of Eco R124I are consistent with the predicted inverted positions of the TRDs. Amino TRDs of similar amino acid sequence have been shown to recognise the same nucleotide sequence. The similarity reported here, the first example of one between amino and carboxy TRDs, while consistent with a conserved mechanism of target recognition, offers additional flexibility in the evolution of sequence specificity by increasing the potential diversity of DNA targets for a given number of TRDs. Sty SKI identifies the first member of the IB family in Salmonella species.

Regulation of the activity of the type IC EcoR124I restriction enzyme

Restriction-modification (R-M) systems must regulate the expression of their genes so that the chromosomal genome is modified at all times by the methyltransferase to protect the host cell from the potential lethal action of the cognate restriction endonuclease. Since type I R-M systems can be transferred to non-modified Escherichia coli cells by conjugation or transformation without killing the recipient, they must have some means to regulate their restriction activity upon entering a new host cell to avoid restriction of unprotected host DNA and cell death. This is especially true for EcoR124I, a type IC family member, which is coded for by a conjugative plasmid. Control of EcoR124I restriction activity is most likely at the post-translational level as the transfer of the EcoR124I system into a recipient cell that already expressed the HsdR subunit of this system was not a lethal event. Additionally, the kinetics of restriction activity upon transfer of the genes coding for the EcoR124I RM system to a recipient cell are the same, irrespective of the modification state of the recipient cell or the presence or absence of the EcoR124I HsdR subunit in the new host cells. The mechanism controlling the restriction activity of a type IC R-M system upon transfer to a new host cell is different from that controlling the chromosomally coded type IA and IB R-M systems. The previously discovered hsdC mutant, which affects the establishment of the type IA system EcoKI, was shown to affect the establishment of the type IB system EcoAI, but to have no influence on EcoR124I.

High resolution footprinting of a type I methyltransferase reveals a large structural distortion within the DNA recognition site

The type I DNA methyltransferase M.EcoR124I is a multi-subunit enzyme that binds to the sequence GAAN6RTCG, transferring a methyl group from S-adenosyl methionine to a specific adenine on each DNA strand. We have investigated the protein-DNA interactions in the complex by DNase I and hydroxyl radical footprinting. The DNase I footprint is unusually large: the protein protects the DNA on both strands for at least two complete turns of the helix, indicating that the enzyme completely encloses the DNA in the complex. The higher resolution hydroxyl radical probe shows a smaller, but still extensive, 18 bp footprint encompassing the recognition site. Within this region, however, there is a remarkably hyper-reactive site on each strand. The two sites of enhanced cleavage are co-incident with the two adenines that are the target bases for methylation, showing that the DNA is both accessible and highly distorted at these sites. The hydroxyl radical footprint is unaffected by the presence of the cofactor S-adenosyl methionine, showing that the distorted DNA structure induced by M.EcoR124I is formed during the initial DNA binding reaction and not as a transient intermediate in the reaction pathway.

The restriction-modification system of Pasteurella haemolytica is a member of a new family of type I enzymes

Genes encoding the type I restriction-modification (R-M) system of the bovine pathogen, Pasteurella haemolytica, have been identified immediately downstream of a locus that encodes a transcriptional activator of P. haemolytica leukotoxin expression. Type I enzymes are encoded by three genes called hsdM, hsdS and hsdR, and have fallen into three groups, called Ia, Ib and Ic. HsdS provides a sequence recognition function which in concert with HsdM forms an active methyltransferase (MTase). Inclusion of the HsdR subunit in the complex creates an active restriction endonuclease (ENase) capable of cleaving unmethylated target DNA. The P. haemolytica hsdMSR genes were mapped using transposon Tn10d-Cam insertions, and bacteriophage restriction and modification assays in Escherichia coli. We determined the nucleotide sequences of hsdM, hsdS and hsdR, and observed that the deduced amino acid (aa) sequences were very similar to predicted R-M subunits in the respiratory pathogen, Haemophilus influenzae. Phylogenetic comparisons of all known Hsd aa sequences placed the P. haemolytica and H. influenzae proteins into a new group which we labeled the Type Id R-M family. Expression of the P. haemolytica R-M genes in E. coli was inefficient and is likely to be a consequence of the unusual codon usage in P. haemolytica genes.

Tyrosine 27 of the specificity polypeptide of EcoKI can be UV crosslinked to a bromodeoxyuridine-substituted DNA target sequence

The specificity (S) subunit of the restriction enzyme EcoKI imparts specificity for the sequence AAC(N6)GTGC. Substitution of thymine with bromodeoxyuridine in a 25 bp DNA duplex containing this sequence stimulated UV light-induced covalent crosslinking to the S subunit. Crosslinking occurred only at the residue complementary to the first adenine in the AAC sequence, demonstrating a close contact between the major groove at this sequence and the S subunit. Peptide sequencing of a proteolytically-digested, crosslinked complex identified tyrosine 27 in the S subunit as the site of crosslinking. This is consistent with the role of the N-terminal domain of the S subunit in recognizing the AAC sequence. Tyrosine 27 is conserved in the S subunits of the three type I enzymes that share the sequence AA in the trinucleotide component of their target sequence. This suggests that tyrosine 27 may make a similar DNA contact in these other enzymes.

Overproduction of the Hsd subunits leads to the loss of temperature-sensitive restriction and modification phenotype

The genes hsdM and hsdS for M. EcoKI modification methyltransferase and the complete set of hsdR, hsdM and hsdS genes coding for R. EcoKI restriction endonuclease, both with and without a temperature-sensitive (ts) mutation in hsdS gene, were cloned in pBR322 plasmid and introduced into E. coli C (a strain without a natural restriction-modification (R-M) system). The strains producing only the methyltransferase, or together with the endonuclease, were thus obtained. The hsdSts-1 mutation, mapped previously in the distal variable region of the hsdS gene with C1 245-T transition has no effect on the R-M phenotype expressed from cloned genes in bacteria grown at 42 degrees C. In clones transformed with the whole hsd region an alleviation of R-M functions was observed immediately after the transformation, but after subculture the transformants expressed the wild-type R-M phenotype irrespective of whether the wild-type or the mutant hsdS allele was present in the hybrid plasmid. Simultaneous overproduction of HsdS and HsdM subunits impairs the ts effect of the hsdSts-1 mutation on restriction and modification.

Effect of site-specific modification on restriction endonucleases and DNA modification methyltransferases

Restriction endonucleases have site-specific interactions with DNA that can often be inhibited by site-specific DNA methylation and other site-specific DNA modifications. However, such inhibition cannot generally be predicted. The empirically acquired data on these effects are tabulated for over 320 restriction endonucleases. In addition, a table of known site-specific DNA modification methyltransferases and their specificities is presented along with EMBL database accession numbers for cloned genes.

The expression and regulation of hsdK genes after conjugative transfer

The type I restriction and modification genes of Escherichia coli can be transferred to other non-modified strains by conjugation without killing the recipient, implying that the restriction function must be regulated. In this study, two isogenic F' plasmids (r+K and r-K) served as donors in quantitative conjugation experiments with various restriction-deficient strains of E. coli and Salmonella typhimurium. Conjugation studies with hsd::lacZ operon fusions in F' plasmids indicate that both of the hsdK promoters, p(res) and pmod, express simultaneously following conjugative transfer. Thus these genes do not appear to be regulated at the transcriptional level. A spontaneous mutant of E. coli C was discovered that is presumably killed upon conjugative transfer of the hsdK genes (defined as a Crc- phenotype). The gene that is defective in the mutant was tentatively designated hsdC (control). Hfr gene replacement studies led to the localization of the putative hsdC gene between 6 and 16 min on the E. coli genetic map.

Substrate recognition and selectivity in the type IC DNA modification methylase M.EcoR124I

The type I DNA modification methylase M.EcoR124I binds sequence specifically to DNA and protects a 25bp fragment containing its cognate recognition sequence from digestion by exonuclease III. Using modified synthetic oligonucleotide duplexes we have investigated the catalytic properties of the methylase, and have established that a specific adenine on each strand of DNA is the site of methylation. We show that the rate of methylation of each adenine is increased at least 100 fold by prior methylation at the other site. However, this is accompanied by a significant decrease in the affinity of the methylase for these substrates according to competitive gel retardation assays. In contrast, methylation of an adenine in the recognition site which is not a target for the enzyme results in only a small decrease in both DNA binding affinity and rate of methylation by the enzyme.

A deletion mutant of the type IC restriction endonuclease EcoR1241 expressing a novel DNA specificity

We have developed a complementation assay which allows us to distinguish between mutations affecting subunit assembly and mutations affecting DNA binding in the DNA recognition subunit (HsdS) of the multimeric restriction endonuclease EcoR1241. A number of random point mutations were constructed to test the validity of this assay. Two of the mutants produced were found to be truncated polypeptides that were still capable of complementation with the EcoR1241 Hsd subunits to give an active restriction enzyme of novel DNA specificity. The N-terminal variable domain (responsible for recognition of GAA from the EcoR1241 recognition sequence GAAnnnnnnRTCG) and the spacer region (central conserved region) is intact in both of these mutants. One of these mutant genes (hsdS(delta 50) has been cloned as an active Mtase. Purification of the Mtase proved to be difficult because the complex is weak. However, Mtase activity was obtained from a soluble cell extract, and this allowed us to determine the DNA recognition sequence of the Mtase to be GAAnnnnnnnTTC. This recognition sequence is an inverted repeat of 5'-end of the EcoR1241 recognition sequence. This suggests that the mutant Mtase is assembled from two inverted HsdS(D50) subunits, possibly held together by the HsdM subunits.

Delayed expression of in vivo restriction activity following conjugal transfer of Escherichia coli hsdK (restriction-modification) genes

Following conjugal transfer of the hsdK genes (hsdRK, hsdMK, and hsdSK) of Escherichia coli K-12, restriction activity was first detected only after approximately 15 generations, whereas modification activity was observed immediately. This sequential expression explains the establishment of hsdK genes in a nonmodified host and suggests regulation of restriction activity after conjugal transfer.

Conservation of motifs within the unusually variable polypeptide sequences of type I restriction and modification enzymes

Type I restriction enzymes comprise three subunits encoded by genes designated hsdR, hsdM, and hsdS; S confers sequence specificity. Three families of enzymes are known and within families, but not between, hsdM and hsdR are conserved. Consequently, interfamily comparisons of M and R sequences focus on regions of putative functional significance, while both inter- and intrafamily comparisons address the origin, nature and role of diversity of type I restriction systems. We have determined the sequence of the hsdR gene for EcoA, thus making available sequences of all three hsd genes of one representative from each family. The predicted R polypeptide sequences share conserved regions with one superfamily of putative helicases, so-called 'DEAD box' proteins; these conserved sequences may be associated with the ATP-dependent translocation of DNA that precedes restriction. We also present hsdM and hsdR sequences for EcoE, a member of the same family as EcoA. The sequences of the M and R genes of EcoA and EcoE are at least as divergent as typical genes from Escherichia coli and Salmonella, perhaps as the result of selection favouring diversity of restriction specificities combined with lateral transfer among different species.

Biology of DNA restriction

Our understanding of the evolution of DNA restriction and modification systems, the control of the expression of the structural genes for the enzymes, and the importance of DNA restriction in the cellular economy has advanced by leaps and bounds in recent years. This review documents these advances for the three major classes of classical restriction and modification systems, describes the discovery of a new class of restriction systems that specifically cut DNA carrying the modification signature of foreign cells, and deals with the mechanisms developed by phages to avoid the restriction systems of their hosts.

Cloning, production and characterisation of wild type and mutant forms of the R.EcoK endonucleases

The hsdR, hsdM and hsdS genes coding for R.EcoK restriction endonuclease, both with and without a temperature sensitive mutation (ts-1) in the hsdS gene, were cloned in pBR322 plasmid and introduced into E.coli C3-6. The presence of the hsdSts-1 mutation has no effect on the R-M phenotype of this construct in bacteria grown at 42 degrees C. However, DNA sequencing indicates that the mutation is still present on the pBR322-hsdts-1 operon. The putative temperature-sensitive endonuclease was purified from bacteria carrying this plasmid and the ability to cleave and methylate plasmid DNA was investigated. The mutant endonuclease was found to show temperature-sensitivity for restriction. Modification was dramatically reduced at both the permissive and non-permissive temperatures. The wild type enzyme was found to cleave circular DNA in a manner which strongly suggests that only one endonuclease molecule is required per cleavage event. Circular and linear DNA appear to be cleaved using different mechanisms, and cleavage of linear DNA may require a second endonuclease molecule. The subunit composition of the purified endonucleases was investigated and compared to the level of subunit production in minicells. There is no evidence that HsdR is prevented from assembling with HsdM and HsdSts-1 to produce the mutant endonuclease. The data also suggests that the level of HsdR subunit may be limiting within the cell. We suggest that an excess of HsdM and HsdS may produce the methylase in vivo and that assembly of the endonuclease may be dependent upon the prior production of this methylase.

Roles of selection and recombination in the evolution of type I restriction-modification systems in enterobacteria

Restriction-modification systems can protect bacteria against viral infection. Sequences of the hsdM gene, encoding one of the three subunits of type I restriction-modification systems, have been determined for four strains of enterobacteria. Comparison with the known sequences of EcoK and EcoR124 indicates that all are homologous, though they fall into three families (exemplified by EcoK, EcoA, and EcoR124), the first two of which are apparently allelic. The extent of amino acid sequence identity between EcoK and EcoA is so low that the genes encoding them might be better termed pseudoalleles; this almost certainly reflects genetic exchange among highly divergent species. Within the EcoK family the ratio of intra- to interspecific divergence is very high. The extent of divergence between the genes from Escherichia coli K-12 and Salmonella typhimurium LT2 is similar to that for other genes with the same level of codon usage bias. In contrast, intraspecific divergence (between E. coli strains B and K-12) is extremely high and may reflect the action of frequency-dependent selection mediated by bacteriophages. There is also evidence of lateral transfer of a short sequence between E. coli and S. typhimurium.

Mutation in the specificity polypeptide of the type I restriction endonuclease R.EcoK that affects subunit assembly

We describe the isolation and characterization of a temperature-sensitive mutation within the hsdS gene of the type I restriction and modification system EcoK. This mutation appears to affect the ability of the HsdR subunit to interact with the HsdS subunit when forming an active endonuclease. We discuss the possibility that this mutant, together with another mutation described previously, may define a discontinuous domain, involved in protein-protein interactions, within the HsdS polypeptide.

Mutations that confer de novo activity upon a maintenance methyltransferase

DNA methyltransferases are not only sequence specific in their action, but they also differentiate between the alternative methylation states of a target site. Some methyltransferases are equally active on either unmethylated or hemimethylated DNA and consequently function as de novo methyltransferases. Others are specific for hemimethylated target sequences, consistent with the postulated role of a maintenance methyltransferase in perpetuating a pattern of DNA modification. The molecular basis for the difference between de novo and maintenance methyltransferase activity is unknown, yet fundamental to cellular activities that are affected by different methylation states of the genome. The methyltransferase activity of the type I restriction and modification system, EcoK, is the only known prokaryotic methyltransferase shown to be specific for hemimethylated target sequences. We have isolated mutants of Escherichia coli K-12 which are able to modify unmethylated target sequences efficiently in a manner indicative of de novo methyltransferase activity. Consistent with this change in specificity, some mutations shift the balance between DNA restriction and modification as if both activities now compete at unmethylated targets. Two genes encode the methyltransferase and all the mutations are loosely clustered within one of them.

Organization of restriction-modification systems

The genes for over 100 restriction-modification systems have now been cloned, and approximately one-half have been sequenced. Despite their similar function, they are exceedingly heterogeneous. The heterogeneity is evident at three levels: in the gene arrangements; in the enzyme compositions; and in the protein sequences. This paper summarizes the main features of the R-M systems that have been cloned.

Conservation of organization in the specificity polypeptides of two families of type I restriction enzymes

We have identified the recognition sequence for the Citrobacter freundii restriction endonuclease CfrA, a member of the A-family of type I R-M enzymes. This bipartite target sequence differs in both its components from those of other type I enzymes. We determined the nucleotide sequence of its specificity gene (hsdS) and a comparison of this with its relative EcoA identifies two extensive variable regions, an organization analogous to that found in the K-family of type I R-M enzymes. The specificity polypeptides of the A-family, unlike those of K, have an N-terminal conserved region, and this includes a sequence repeated within the central conserved region. A second repeat sequence, identified at the amino acid level, coincides with the only sequence similarity common to all type I S polypeptides. Sequences immediately downstream from the hsdS genes of EcoA, CfrA, EcoK, B and D are almost identical, consistent with an allelic chromosomal location.

Conservation of complex DNA recognition domains between families of restriction enzymes

One polypeptide, designated S, confers sequence-specificity to the multisubunit type I restriction enzymes. Two families of such enzymes, K and A, include members that recognize diverse, bipartite, target sequences. The S polypeptides of the K family, while having areas of near identity, also contain two extensive regions of variable sequence. We now show that one of these, comprising the N-terminal 150 amino acids, specifies recognition of one component of the bipartite target sequence. We have determined the sequence recognized by EcoE, a member of the A family. This sequence, 5'GAG(N7)ATGC, has the trinucleotide GAG in common with EcoA and with StySB of the K family. We determined the nucleotide sequences of the S genes of EcoA and EcoE, and compared their predicted amino acid sequences with each other and with those of the five members of the K family. There is no general sequence similarity between families, but the domain of the S polypeptide of StySB, which specifies GAG, shows nearly 50 per cent identity with the amino variable region of the S polypeptides of EcoA and EcoE. A complex domain that recognizes and directs methylation of GAG is therefore common to enzymes of generally dissimilar amino acid sequence.

Basis for changes in DNA recognition by the EcoR124 and EcoR124/3 type I DNA restriction and modification enzymes

EcoR124 and EcoR124/3 are type I DNA restriction and modification systems. The EcoR124/3 system arose from the EcoR124 system some 15 years ago and at the electron microscopic DNA heteroduplex level the genes for both systems are still apparently identical. We have shown that the DNA sequences recognized by the two systems are GAA(N6)RTCG for EcoR124 and GAA(N7)RTCG for EcoR124/3. The sequences thus differ only in the length of the non-specific spacer. This difference nevertheless places the two specific domains of the EcoR124/3 recognition sequence 0.34 nm further apart and rotates them 36 degrees with respect to those of EcoR124, which implies major structural differences in the proteins recognizing these sequences. We have now determined the nucleotide sequences of the hsdS and hsdM genes of both systems and of the hsdR gene of EcoR124/3. The hsdS gene products provide DNA sequence specificity in both restriction and modification, the hsdM gene products are necessary for modification and all three hsd gene products are required for restriction. The only difference that we have detected between the two systems is that a 12 base-pair sequence towards the middle of the hsdS gene is repeated twice in the EcoR124 gene and three times in the EcoR124/3 gene. We have deleted one of the repeats in the EcoR124/3 gene and shown that this changes the specificity to that of EcoR124. Thus, the extra four amino acids in the middle of the EcoR124/3 hsdS gene product, which in an alpha-helical configuration would extend 0.6 nm, are sufficient to explain the differences in sequence recognition. We suggest that the EcoR124/3 system was generated by an unequal crossing over and argue that this kind of specificity change should not be rare in Nature.

Cloned restriction-modification systems--a review

The genes for numerous restriction endonucleases and modification methylases have been cloned into Escherichia coli. A summary is given for the clones isolated so far (115 entries) and of the procedures used to obtain them.

Complementation and hybridization evidence for additional families of type I DNA restriction and modification genes in Salmonella serotypes

Of eight Salmonella, serB-linked hsd genes for the restriction and modification of DNA transferred to Escherichia coli/Salmonella hybrids, only two--those with SM and ST (S. muenchen and S. thompson, respectively) specificities--may have weakly complemented rSB- and none complemented rK-. An A-specific DNA probe failed to hybridize to HindIII-restricted fragments of each of the hybrids, but an SB (S. typhimurium)-specific probe hybridized to DNA from the hybrid with ST specificity. These results indicate that additional families of the type I hsd genes may exist.

Distribution and diversity of hsd genes in Escherichia coli and other enteric bacteria

We screened Salmonella typhimurium, Citrobacter freundii, Klebsiella pneumoniae, Shigella boydii, and many isolates of Escherichia coli for DNA sequences homologous to those encoding each of two unrelated type I restriction and modification systems (EcoK and EcoA). Both K- and A-related hsd genes were identified, but never both in the same strain. S. typhimurium encodes three restriction and modification systems, but its DNA hybridized only to the K-specific probe which we know to identify the StySB system. No homology to either probe was detected in the majority of E. coli strains, but in C. freundii, we identified homology to the A-specific probe. We cloned this region of the C. freundii genome and showed that it encoded a functional, A-related restriction system whose specificity differs from those of known type I enzymes. Sequences immediately flanking the hsd K genes of E. coli K-12 and the hsd A genes of E. coli 15T- were shown to be homologous, indicating similar or even identical positions in their respective chromosomes. E. coli C has no known restriction system, and the organization of its chromosome is consistent with deletion of the three hsd genes and their neighbor, mcrB.