Functional cooperation between exonucleases and endonucleases--basis for the evolution of restriction enzymes

Many types of restriction enzymes cleave DNA away from their recognition site. Using the type III restriction enzyme, EcoP15I, which cleaves DNA 25–27 bp away from its recognition site, we provide evidence to show that an intact recognition site on the cleaved DNA sequesters the restriction enzyme and decreases the effective concentration of the enzyme. EcoP15I restriction enzyme is shown here to perform only a single round of DNA cleavage. Significantly, we show that an exonuclease activity is essential for EcoP15I restriction enzyme to perform multiple rounds of DNA cleavage. This observation may hold true for all restriction enzymes cleaving DNA sufficiently far away from their recognition site. Our results highlight the importance of functional cooperation in the modulation of enzyme activity. Based on results presented here and other data on well-characterised restriction enzymes, a functional evolutionary hierarchy of restriction enzymes is discussed.     

Real-time observation of DNA looping dynamics of Type IIE restriction enzymes NaeI and NarI

Many restriction enzymes require binding of two copies of a recognition sequence for DNA cleavage, thereby introducing a loop in the DNA. We investigated looping dynamics of Type IIE restriction enzymes NaeI and NarI by tracking the Brownian motion of single tethered DNA molecules. DNA containing two endonuclease recognition sites spaced a few 100 bp apart connect small polystyrene beads to a glass surface. The position of a bead is tracked through video microscopy. Protein-mediated looping and unlooping is then observed as a sudden specific change in Brownian motion of the bead. With this method we are able to directly follow DNA looping kinetics of single protein–DNA complexes to obtain loop stability and loop formation times. We show that, in the absence of divalent cations, NaeI induces DNA loops of specific size. In contrast, under these conditions NarI mainly creates non-specific loops, resulting in effective DNA compaction for higher enzyme concentrations. Addition of Ca²⁺ increases the NaeI-DNA loop lifetime by two orders of magnitude and stimulates specific binding by NarI. Finally, for both enzymes we observe exponentially distributed loop formation times, indicating that looping is dominated by (re)binding the second recognition site.     

Type III restriction endonucleases translocate DNA in a reaction driven by recognition site-specific ATP hydrolysis.

Type III restriction/modification systems recognize short non-palindromic sequences, only one strand of which can be methylated. Replication of type III-modified DNA produces completely unmethylated recognition sites which, according to classical mechanisms of restriction, should be signals for restriction. We have shown previously that suicidal restriction by the type III enzyme EcoP15I is prevented if all the unmodified sites are in the same orientation: restriction by EcoP15I requires a pair of unmethylated, inversely oriented recognition sites. We have now addressed the molecular mechanism of site orientation-specific DNA restriction. EcoP15I is demonstrated to possess an intrinsic ATPase activity, the potential driving force of DNA translocation. The ATPase activity is uniquely recognition site-specific, but EcoP15I-modified sites also support the reaction. EcoP15I DNA restriction patterns are shown to be predetermined by the enzyme-to-site ratio, in that site-saturating enzyme levels elicit cleavage exclusively between the closest pair of head-to-head oriented sites. DNA restriction is blocked by Lac repressor bound in the intervening sequence between the two EcoP15I sites. These results rule out DNA looping and strongly suggest that cleavage is triggered by the close proximity of two convergently tracking EcoP15I-DNA complexes.     

Type III restriction-modification enzymes: a historical perspective

Restriction endonucleases interact with DNA at specific sites leading to cleavage of DNA. Bacterial DNA is protected from restriction endonuclease cleavage by modifying the DNA using a DNA methyltransferase. Based on their molecular structure, sequence recognition, cleavage position and cofactor requirements, restriction–modification (R–M) systems are classified into four groups. Type III R–M enzymes need to interact with two separate unmethylated DNA sequences in inversely repeated head-to-head orientations for efficient cleavage to occur at a defined location (25–27 bp downstream of one of the recognition sites). Like the Type I R–M enzymes, Type III R–M enzymes possess a sequence-specific ATPase activity for DNA cleavage. ATP hydrolysis is required for the long-distance communication between the sites before cleavage. Different models, based on 1D diffusion and/or 3D-DNA looping, exist to explain how the long-distance interaction between the two recognition sites takes place. Type III R–M systems are found in most sequenced bacteria. Genome sequencing of many pathogenic bacteria also shows the presence of a number of phase-variable Type III R–M systems, which play a role in virulence. A growing number of these enzymes are being subjected to biochemical and genetic studies, which, when combined with ongoing structural analyses, promise to provide details for mechanisms of DNA recognition and catalysis.     

Translocation-independent dimerization of the EcoKI endonuclease visualized by atomic force microscopy

Bacterial type I restriction/modification systems are capable of performing multiple actions in response to the methylation pattern on their DNA recognition sequences. The enzymes making up these systems serve to protect the bacterial cells against viral infection by binding to their recognition sequences on the invading DNA and degrading it after extensive ATP-driven translocation. DNA cleavage has been thought to occur as the result of a collision between two translocating enzyme complexes. Using atomic force microscopy (AFM), we show here that EcoKI dimerizes rapidly when bound to a plasmid containing two recognition sites for the enzyme. Dimerization proceeds in the absence of ATP and is also seen with an EcoKI mutant (K477R) that is unable to translocate DNA. Only monomers are seen when the enzyme complex binds to a plasmid containing a single recognition site. Based on our results, we propose that the binding of EcoKI to specific DNA target sequences is accompanied by a conformational change that leads rapidly to dimerization. This event is followed by ATP-dependent translocation and cleavage of the DNA.     

Unidirectional translocation from recognition site and a necessary interaction with DNA end for cleavage by Type III restriction enzyme

Type III restriction enzymes have been demonstrated to require two unmethylated asymmetric recognition sites oriented head-to-head to elicit double-strand break 25–27 bp downstream of one of the two sites. The proposed DNA cleavage mechanism involves ATP-dependent DNA translocation. The sequence context of the recognition site was suggested to influence the site of DNA cleavage by the enzyme. In this investigation, we demonstrate that the cleavage site of the R.EcoP15I restriction enzyme does not depend on the sequence context of the recognition site. Strikingly, this study demonstrates that the enzyme can cleave linear DNA having either recognition sites in the same orientation or a single recognition site. Cleavage occurs predominantly at a site proximal to the DNA end in the case of multiple site substrates. Such cleavage can be abolished by the binding of Lac repressor downstream (3′ side) but not upstream (5′ side) of the recognition site. Binding of HU protein has also been observed to interfere with R.EcoP15I cleavage activity. In accordance with a mechanism requiring two enzyme molecules cooperating to elicit double-strand break on DNA, our results convincingly demonstrate that the enzyme translocates on DNA in a 5′ to 3′ direction from its recognition site and indicate a switch in the direction of enzyme motion at the DNA ends. This study demonstrates a new facet in the mode of action of these restriction enzymes.     

DNA looping and translocation provide an optimal cleavage mechanism for the type III restriction enzymes

EcoP15I is a type III restriction enzyme that requires two recognition sites in a defined orientation separated by up to 3.5 kbp to efficiently cleave DNA. The mechanism through which site-bound EcoP15I enzymes communicate between the two sites is unclear. Here, we use atomic force microscopy to study EcoP15I-DNA pre-cleavage complexes. From the number and size distribution of loops formed, we conclude that the loops observed do not result from translocation, but are instead formed by a contact between site-bound EcoP15I and a nonspecific region of DNA. This conclusion is confirmed by a theoretical polymer model. It is further shown that translocation must play some role, because when translocation is blocked by a Lac repressor protein, DNA cleavage is similarly blocked. On the basis of these results, we present a model for restriction by type III restriction enzymes and highlight the similarities between this and other classes of restriction enzymes.     

DNA as a possible target for antitumor ruthenium(III) complexes

The interaction of two experimental ruthenium(III)-containing antitumor complexes-Na[trans-RuCl(4)(DMSO)(Im)] (NAMI) and dichloro(1,2-propylendiaminetetraacetate)ruthenium(III) (RAP)-with DNA was investigated through a number of spectroscopic and molecular biology techniques, including spectrophotometry, circular dichroism, gel shift analysis, and restriction enzyme inhibition. It was found that both complexes slightly alter DNA conformation, modify its electrophoretic mobility, and inhibit DNA recognition and cleavage by some restriction enzymes, though they were less effective than cisplatin in producing such effects. Notably, the effects produced by NAMI on DNA were much larger than those induced by RAP. Implications of these results for the mechanism of action of ruthenium(III) antitumor complexes are discussed.     

S-adenosyl-L-methionine is required for DNA cleavage by type III restriction enzymes.

The requirement of S-adenosyl-L-methionine (AdoMet) in the cleavage reaction carried out by type III restriction-modification enzymes has been investigated. We show that DNA restriction by EcoPI restriction enzyme does not take place in the absence of exogenously added AdoMet. Interestingly, the closely related EcoP15I enzyme has endogenously bound AdoMet and therefore does not require the addition of the cofactor for DNA cleavage. By employing a variety of AdoMet analogs, which differ structurally from AdoMet, this study demonstrates that the carboxyl group and any substitution at the epsilon carbon of methionine is absolutely essential for DNA cleavage. Such analogs could bring about the necessary conformational change(s) in the enzyme, which make the enzyme proficient in DNA cleavage. Our studies, which include native polyacrylamide gel electrophoresis, molecular size exclusion chromatography, UV, fluorescence and circular dichroism spectroscopy, clearly demonstrate that the holoenzyme and apoenzyme forms of EcoP15I restriction enzyme have different conformations. Furthermore, the Res and Mod subunits of the EcoP15I restriction enzyme can be separated by gel filtration chromatography in the presence of 2 M NaCl. Reconstitution experiments, which involve mixing of the isolated subunits, result in an apoenzyme form, which is restriction proficient in the presence of AdoMet. However, mixing the Res subunit with Mod subunit deficient in AdoMet binding does not result in a functional restriction enzyme. These observations are consistent with the fact that AdoMet is required for DNA cleavage. In vivo complementation of the defective mod allele with a wild-type mod allele showed that an active restriction enzyme could be formed. Furthermore, we show that while the purified c2-134 mutant restriction enzyme is unable to cleave DNA, the c2-440 mutant enzyme is able to cleave DNA albeit poorly. Taken together, these results suggest that AdoMet binding causes conformational changes in the restriction enzyme and is necessary to bring about DNA cleavage.     

Simultaneous binding of three recognition sites is necessary for a concerted plasmid DNA cleavage by EcoRII restriction endonuclease

According to the current paradigm type IIE restriction endonucleases are homodimeric proteins that simultaneously bind to two recognition sites but cleave DNA at only one site per turnover: the other site acts as an allosteric locus, activating the enzyme to cleave DNA at the first. Structural and biochemical analysis of the archetypal type IIE restriction enzyme EcoRII suggests that it has three possible DNA binding interfaces enabling simultaneous binding of three recognition sites. To test if putative synapsis of three binding sites has any functional significance, we have studied EcoRII cleavage of plasmids containing a single, two and three recognition sites under both single turnover and steady state conditions. EcoRII displays distinct reaction patterns on different substrates: (i) it shows virtually no activity on a single site plasmid; (ii) it yields open-circular DNA form nicked at one strand as an obligatory intermediate acting on a two-site plasmid; (iii) it cleaves concertedly both DNA strands at a single site during a single turnover on a three site plasmid to yield linear DNA. Cognate oligonucleotide added in trans increases the reaction velocity and changes the reaction pattern for the EcoRII cleavage of one and two-site plasmids but has little effect on the three-site plasmid. Taken together the data indicate that EcoRII requires simultaneous binding of three rather than two recognition sites in cis to achieve concerted DNA cleavage at a single site. We show that the orthodox type IIP enzyme PspGI which is an isoschisomer of EcoRII, cleaves different plasmid substrates with equal rates. Data provided here indicate that type IIE restriction enzymes EcoRII and NaeI follow different mechanisms. We propose that other type IIE restriction enzymes may employ the mechanism suggested here for EcoRII.     

DNA translocation blockage, a general mechanism of cleavage site selection by type I restriction enzymes

Type I restriction enzymes bind to a specific DNA sequence and subsequently translocate DNA past the complex to reach a non-specific cleavage site. We have examined several potential blocks to DNA translocation, such as positive supercoiling or a Holliday junction, for their ability to trigger DNA cleavage by type I restriction enzymes. Introduction of positive supercoiling into plasmid DNA did not have a significant effect on the rate of DNA cleavage by EcoAI endonuclease nor on the enzyme's ability to select cleavage sites randomly throughout the DNA molecule. Thus, positive supercoiling does not prevent DNA translocation. EcoR124II endonuclease cleaved DNA at Holliday junctions present on both linear and negatively supercoiled substrates. The latter substrate was cleaved by a single enzyme molecule at two sites, one on either side of the junction, consistent with a bi-directional translocation model. Linear DNA molecules with two recognition sites for endonucleases from different type I families were cut between the sites when both enzymes were added simultaneously but not when a single enzyme was added. We propose that type I restriction enzymes can track along a DNA substrate irrespective of its topology and cleave DNA at any barrier that is able to halt the translocation process.     

S-Adenosyl-L-methionine-dependent restriction enzymes

Restriction-modification (R-M) enzymes are classified into type I, II, III, and IV, based on their recognition sequence, subunit composition, cleavage position, and cofactor requirements. While the role of S-Adenosyl-L-methionine (AdoMet) as the methyl group donor in the methylation reaction is undisputed, its requirement in DNA cleavage reaction has been subject to intense study. AdoMet is a prerequisite for the DNA cleavage by most type I enzymes known so far, with the exception of R.EcoR124I. A number of new type II restriction enzymes belonging to the type IIB and IIG family were found to show AdoMet dependence for their cleavage reaction. The type III enzymes have been found to require AdoMet for their restriction function. AdoMet functions as an allosteric effector of the DNA cleavage reaction and has been shown to bring about conformational changes in the protein upon binding.     

Scanning force microscopy of DNA translocation by the Type III restriction enzyme EcoP15I

Type III restriction enzymes are multifunctional heterooligomeric enzymes that cleave DNA at a fixed position downstream of a non-symmetric recognition site. For effective DNA cleavage these restriction enzymes need the presence of two unmethylated, inversely oriented recognition sites in the DNA molecule. DNA cleavage was proposed to result from ATP-dependent DNA translocation, which is expected to induce DNA loop formation, and collision of two enzyme-DNA complexes. We used scanning force microscopy to visualise the protein interaction with linear DNA molecules containing two EcoP15I recognition sites in inverse orientation. In the presence of the cofactors ATP and Mg(2+), EcoP15I molecules were shown to bind specifically to the recognition sites and to form DNA loop structures. One of the origins of the protein-clipped DNA loops was shown to be located at an EcoP15I recognition site, the other origin had an unspecific position in between the two EcoP15I recognition sites. The data demonstrate for the first time DNA translocation by the Type III restriction enzyme EcoP15I using scanning force microscopy. Moreover, our study revealed differences in the DNA-translocation processes mediated by Type I and Type III restriction enzymes.     

Inhibition of restriction endonuclease Nci I cleavage by phosphorothioate groups and its application to oligonucleotide-directed mutagenesis.

M13 RF IV DNA where phosphorothioate groups are incorporated at restriction endonuclease Nci I recognition sites in the (-)strand is efficiently nicked by the action of this enzyme. Incubation of such nicked DNA with exonuclease III produces gapped DNA. The gap can be filled by reaction with deoxynucleoside triphosphates and DNA polymerase I. When this sequence of reactions is performed with DNA containing a mismatch oligonucleotide primer in the (-)-strand mutational frequencies of 70-90% can be obtained upon transformation. The general nature of this methodology has been further shown to be applicable to other restriction enzymes such as Hind II, Pst I and Fsp I. The mutational frequency obtained using these enzymes is between 40-80% mainly because of less efficient nicking and gapping. Studies on inhibition of Nci I cleavage show that in addition to a phosphorothioate group at the position of cleavage an additional group in the 5'-neighbouring position is necessary for complete inhibition.     

DNA cleavage by type III restriction-modification enzyme EcoP15I is independent of spacer distance between two head to head oriented recognition sites

The type III restriction-modification enzyme EcoP15I requires the interaction of two unmethylated, inversely oriented recognition sites 5'-CAGCAG in head to head configuration to allow an efficient DNA cleavage. It has been hypothesized that two convergent DNA-translocating enzyme-substrate complexes interact to form the active cleavage complex and that translocation is driven by ATP hydrolysis. Using a half-automated, fluorescence-based detection method, we investigated how the distance between two inversely oriented recognition sites affects DNA cleavage efficiency. We determined that EcoP15I cleaves DNA efficiently even for two adjacent head to head or tail to tail oriented target sites. Hence, DNA translocation appears not to be required for initiating DNA cleavage in these cases. Furthermore, we report here that EcoP15I is able to cleave single-site substrates. When we analyzed the interaction of EcoP15I with DNA substrates containing adjacent target sites in the presence of non-hydrolyzable ATP analogues, we found that cleavage depended on the hydrolysis of ATP. Moreover, we show that cleavage occurs at only one of the two possible cleavage positions of an interacting pair of target sequences. When EcoP15I bound to a DNA substrate containing one recognition site in the absence of ATP, we observed a 36 nucleotide DNaseI-footprint that is asymmetric on both strands. All of our footprinting experiments showed that the enzyme did not cover the region around the cleavage site. Analyzing a DNA fragment with two head to head oriented recognition sites, EcoP15I protected 27-33 nucleotides around the recognition sequence, including an additional region of 26 bp between both cleavage sites. For all DNA substrates examined, the presence of ATP caused altered footprinting patterns. We assume that the altered patterns are most likely due to a conformational change of the enzyme. Overall, our data further refine the tracking-collision model for type III restriction enzymes.     

HineI is an isoschizomer of HinfIII restriction endonuclease

HineI is a restriction enzyme isolated from Haemophilus influenzae strain Re. Like other type III restriction endonucleases it requires ATP for cleavage and S-adenosyl-methionine for methylation of DNA. This enzyme recognises the same sequence as HinfIII (Piekarowicz et al., 1981) and cleaves and methylates DNA in a manner similar to all type III restriction enzymes.     

Tailoring the activity of restriction endonuclease PleI by PNA-induced DNA looping

DNA looping is one of the key factors allowing proteins bound to different DNA sites to signal one another via direct contacts. We demonstrate that DNA looping can be generated in an arbitrary chosen site by sequence-directed targeting of double-stranded DNA with pseudocomplementary peptide-nucleic acids (pcPNAs). We designed pcPNAs to mask the DNA from cleavage by type IIs restriction enzyme PleI while not preventing the enzyme from binding to its primary DNA recognition site. Direct interaction between two protein molecules (one bound to the original recognition site and the other to a sequence-degenerated site) results in a totally new activity of PleI: it produces a nick near the degenerate site. The PNA-induced nicking efficiency varies with the distance between the two protein-binding sites in a phase with the DNA helical periodicity. Our findings imply a general approach for the fine-tuning of proteins bound to DNA sites well separated along the DNA chain.     

Evolutionary relationship between different subgroups of restriction endonucleases

The type II restriction endonuclease SsoII shows sequence similarity with 10 other restriction endonucleases, among them the type IIE restriction endonuclease EcoRII, which requires binding to an effector site for efficient DNA cleavage, and the type IIF restriction endonuclease NgoMIV, which is active as a homotetramer and cleaves DNA with two recognition sites in a concerted reaction. We show here that SsoII is an orthodox type II enzyme, which is active as a homodimer and does not require activation by binding to an effector site. Nevertheless, it shares with EcoRII and NgoMIV a very similar DNA-binding site and catalytic center as shown here by a mutational analysis, indicative of an evolutionary relationship between these three enzymes. We suggest that a similar relationship exists between other orthodox type II, type IIE, and type IIF restriction endonucleases. This may explain why similarities may be more pronounced between members of different subtypes of restriction enzymes than among the members of a given subtype.     

S-adenosyl methionine prevents promiscuous DNA cleavage by the EcoP1I type III restriction enzyme

DNA cleavage by the type III restriction endonuclease EcoP1I was analysed on circular and catenane DNA in a variety of buffers with different salts. In the presence of the cofactor S-adenosyl methionine (AdoMet), and irrespective of buffer, only substrates with two EcoP1I sites in inverted repeat were susceptible to cleavage. Maximal activity was achieved at a Res2Mod2 to site ratio of approximately 1:1 yet resulted in cleavage at only one of the two sites. In contrast, the outcome of reactions in the absence of AdoMet was dependent upon the identity of the monovalent buffer components, in particular the identity of the cation. With Na+, cleavage was observed only on substrates with two sites in inverted repeat at elevated enzyme to site ratios (>15:1). However, with K+ every substrate tested was susceptible to cleavage above an enzyme to site ratio of approximately 3:1, including a DNA molecule with two directly repeated sites and even a DNA molecule with a single site. Above an enzyme to site ratio of 2:1, substrates with two sites in inverted repeat were cleaved at both cognate sites. The rates of cleavage suggested two separate events: a fast primary reaction for the first cleavage of a pair of inverted sites; and an order-of-magnitude slower secondary reaction for the second cleavage of the pair or for the first cleavage of all other site combinations. EcoP1I enzymes mutated in either the ATPase or nuclease motifs did not produce the secondary cleavage reactions. Thus, AdoMet appears to play a dual role in type III endonuclease reactions: Firstly, as an allosteric activator, promoting DNA association; and secondly, as a "specificity factor", ensuring that cleavage occurs only when two endonucleases bind two recognition sites in a designated orientation. However, given the right conditions, AdoMet is not strictly required for DNA cleavage by a type III enzyme.     

Nucleoside triphosphate-dependent restriction enzymes

The known nucleoside triphosphate-dependent restriction enzymes are hetero-oligomeric proteins that behave as molecular machines in response to their target sequences. They translocate DNA in a process dependent on the hydrolysis of a nucleoside triphosphate. For the ATP-dependent type I and type III restriction and modification systems, the collision of translocating complexes triggers hydrolysis of phosphodiester bonds in unmodified DNA to generate double-strand breaks. Type I endonucleases break the DNA at unspecified sequences remote from the target sequence, type III endonucleases at a fixed position close to the target sequence. Type I and type III restriction and modification (R-M) systems are notable for effective post-translational control of their endonuclease activity. For some type I enzymes, this control is mediated by proteolytic degradation of that subunit of the complex which is essential for DNA translocation and breakage. This control, lacking in the well-studied type II R-M systems, provides extraordinarily effective protection of resident DNA should it acquire unmodified target sequences. The only well-documented GTP-dependent restriction enzyme, McrBC, requires methylated target sequences for the initiation of phosphodiester bond cleavage.