Restriction Enzymes: A History

By Wil A.M. Loenen, Leiden University Medical Center
April 2019 · 346 pages, illustrated (38 color and 26 B&W)
ISBN 978-1-621821-05-2

<<  Chapter 5   —   Chapter 7  >>

Chapter 6

Chapter doi:10.1101/restrictionenzymes_6

Variety in Mechanisms and Structures of Restriction Enzymes: ∼1982–1993


During the 1970s and 1980s Type II REases became the workhorses of genetic engineers, making the identification of novel enzymes important. By 1993, 188 Type II specificities had been identified, among a total of nearly 2400 enzymes, but still only a small number of Type I and III REases (Roberts and Macelis 1993a,b) rising to approximately 3700 REases by 2004 (Pingoud et al. 2005). It was the largest of any group of nucleases known at the time. It was estimated that ∼25% of bacteria from all genera would carry one or more Type II systems (Roberts and Halford 1993; Roberts and Macelis 1993a,b). The presence of Type I or Type III systems in bacterial strains remained difficult to establish. One of the most important new Type II enzymes identified was EcoRV (Bougueleret et al. 1984, 1985), which would be exhaustively studied over the next three decades (reviewed in Pingoud et al. 2014). Were REase–DNA interactions truly different from those mediated by known DNA recognition modules such as zinc fingers or helix-turn-helix found in transcription factors, repressors, etc.? Was the view, held at the time, that DNA recognition involved only a few such mechanisms mistaken and naive? Also, how did REases find their specific sites among so many kilobases of DNA: Did they use one or more of the various mechanisms proposed by von Hippel and Berg (Berg and von Hippel 1985; von Hippel and Berg 1989)? And what did the long-awaited crystal structures look like?

Despite difficulties encountered with the discovery and analysis of REases, all three types of enzymes had enthusiastic followers who were fascinated with the genetics, molecular biology, and biochemistry of these highly specific enzymes (Bennett and Halford 1989; Wilson 1991; Wilson and Murray 1991; Anderson 1993; Heitman 1993). During the 1980s and early 1990s, the groups in Bristol, Pittsburgh, Edinburgh, and Basel (Chapter 5) and Alfred Pingoud and colleagues in Hannover made important progress toward the understanding of the action of EcoRI and EcoRV and the ATP-dependent Type I and III enzymes. At the same time, other REases became subject to intense study (e.g., BamHI and EcoR124). FokI was different and appeared to contain separate domains for DNA sequence recognition and cleavage (Sugisaki and Kanazawa 1981; Sugisaki et al. 1989), whereas NaeI required interaction with two recognition sequences, such as EcoRII (Van Cott and Wilson 1988; Topal et al. 1991). Like some other “oddball systems, such as BcgI and Eco57I” (Roberts and Halford 1993) these enzymes would become the prototypes of new subclasses in 2003 (Roberts et al. 2003).

Another interesting enzyme was MmeI. Isolated by Imperial Chemistry Industries as AS1 (Aggravated Sludge 1), this bacterial strain was highly resistant to changes in temperature and pressure during growth in bulk. During the oil crisis in the early 1970s, the idea was developed to grow AS1 (which proved to be Methylophilus methylotrophus) in batches of 500,000 liters (!) and use them dried as protein-rich feed for cows. Chris Boyd in Bill Brammar's laboratory identified the R-M system in this strain, MmeI, which he lovingly called “My Mimi,” and worked out the recognition sequence using his own computer program (Boyd et al. 1986). This was still not very common at the time and sometimes Bill would ask in (mock) despair: “Yes, Chris, could you perhaps also do an experiment at the bench to back up your computer data?” The oil crisis ended, and MmeI was shelved until two decades later Richard (Rick) Morgan discovered that it was a rather interesting enzyme (Morgan et al. 2008, 2009; Callahan et al. 2011; Morgan 2013; Chapter 8).

One goal of many workers at the time would prove elusive—that of the generation of new specificities of Type II REases, particularly longer recognition sequences, by genetic manipulation of existing systems. Perhaps enzymes such as EcoRI were simply not “malleable” to reengineering (Roberts and Halford 1993) in this way.

With respect to the Type I systems, the role of these enzymes was under strong debate: Why spend so much metabolic energy on these enzymes, as the vast majority of individual bacterial cells were unlikely to experience phage infection or conjugation outside the laboratory (Bickle 1993)? Did these enzymes play a role in the cellular economy (e.g., in genetic recombination [Price and Bickle 1986]), as well as offer protection against potential phage infections in nature?


Characterization of New Specificities of Type II Enzymes

Initially, the Type II systems identified were R-M systems like EcoRI, a homodimeric REase, which required Mg2+ as cofactor and cleaved at a specific recognition sequence 4–8 bp in length, and a monomeric MTase, which recognized and methylated the same (“cognate”) 4–8-bp site and by doing so protected the site against the REase. This view changed over the years, as the superfamily grew—for example, in some cases two MTases were associated with the REase (e.g., DpnI and DpnII [de la Campa et al. 1987], Esp3I [Bitinaite et al. 1992], and MboI [Ueno et al. 1993]).

By 1993 the rate of discovery of new specificities had dropped to about eight per year (http://library.cshl.edu/Meetings/restriction-enzymes/v-Roberts.php). More iso- and neoschizomer pairs like MboI/Sau3AI and HpaII/MspI were identified that were differentially sensitive to methylation, either within or outside their recognition site (Nelson et al. 1993; Roberts and Halford 1993). This led to progress in cloning in E. coli (Chapter 5) and investigations in the methylation state of the CG dinucleotide in and around genes in higher organisms, a tool toward understanding the organization and transcription of eukaryotic genes (Doerfler 1983).

The Type IIS Enzymes

In addition to the Type II REases that cut at a recognition site with dyad symmetry, a substantial number of the new enzymes cleaved at a short distance away from an asymmetric sequence. These were termed Type IIS (for shifted) (Szybalski et al. 1991) to distinguish them from the EcoRI-like enzymes (Type IIP for palindrome [Roberts et al. 2003]). By 1991, only a small percentage of the more than 1000 Type II REases belonged to the Type IIS class (35 specificities and 80 isoschizomers) (Szybalski et al. 1991).

The Type IIS REases had separate recognition and cleavage domains, somewhat like the Type III enzymes. However, the Type IIS REases appeared to be monomeric proteins, whereas Type III REases were heterodimeric with Mod and Res subunits. In support of this separation of recognition and cleavage modules, deletion of the carboxyl terminus of FokI prevented cleavage but not DNA binding (Li et al. 1993). The distance between the recognition sequence and the cleavage site varied from enzyme to enzyme: FokI cut GGATG (9/13)—that is, it cleaved 9 nt beyond 5′-GGATG on the same strand and 13 nt beyond this sequence on the complementary strand (Sugisaki and Kanazawa 1981). Isoschizomer StsI cleaved 1 nt further along: GGATG (10/14). Also, some enzymes showed variable cleavage distance, either as wild-type or mutant enzyme (Li and Chandrasegaran 1993; Roberts and Halford 1993; see below).

In the Type IIS systems, cognate methylation was due to two activities—one for methylation of each strand of the recognition sequence. In the case of M·FokI the two MTases were fused into a single protein (Looney et al. 1989; Sugisaki et al. 1989), whereas the first Type IIS REase to be discovered, HgaI (found in 1974, but not recognized as such), had two separate cognate MTases (Barsomian et al. 1990). Would the REases have two cutting domains, one for each strand?

The separation between recognition and cleavage domains made Type IIS useful for various manipulations (Szybalski et al. 1991) and would lead to engineering of FokI to obtain hybrid REases (Kim et al. 1998; Sanders et al. 2009; Guo et al. 2010; Halford et al. 2011; Li et al. 2011; Chapter 8). By 1991, it was possible to generate mutants by cutting selected DNAs with FokI, filling in or removing the single-stranded (ss) ends, then cutting again, and again, as one wished, since the recognition site remained intact. One could mix and match sequences—for example, reassemble HgaI-derived fragments of phage f1 replicative form (RF) DNA (Moses and Horiuchi 1979)—and various other applications such as precise excision and amplification (see Szybalski et al. 1991 for details).

Other Type II Enzymes

Among the Type II REases, some enzymes had unusual features. Examples were BcgI, which required SAM as cofactor and cleaved symmetrically on both sides of its asymmetric recognition sequence (GCA[N6]TCG [Kong et al. 1993]), and Sgr20I (Orekhov et al. 1982). The latter recognized the same sequence as EcoRII (CC(A/T)GG, written as CCWGG) but appeared to cleave on both sides of this sequence, as it produced bands on gels that were slightly smaller than those generated with EcoRII (Roberts and Halford 1993). These would be the first of the Type IIC and Type IIH (BcgI), Type IIF (Sgr20I), and Type IIE (EcoRII) REase subgroups that would lead to the new classification in 2003 in 11 subclasses (Roberts et al. 2003).

Determination of Cleavage Sites of Type II Enzymes

New computer programs combined with the availability of sequenced DNA from plasmid, phage, and other sources came to the aid of scientists to characterize the recognition and cleavage sites of new REases. Analysis of the size of bands produced by the REases on sequenced DNA helped to identify potential restriction sites (Gingeras et al. 1978; Tolstoshev and Blakesley 1982; Boyd et al. 1986). Experimental proof was usually obtained using primed synthesis reactions on a couple of sites (Brown and Smith 1980): A primer close to the site was extended with DNA polymerase on a given template. The product was cleaved with the REase, and polymerase and dNTPs were added to one-half of the sample. If the REase produced blunt ends, no dNTPs would be incorporated; if the enzyme produced sticky ends, then a 5′ extension would be repaired, and a 3′ extension trimmed back by the polymerase. In the second and third case the fragment would become longer or shorter than in the first case, which would not change in length. The precise sites of cleavage on both strands could then be determined by running the treated and untreated samples in parallel on a sequencing gel.

Genes and Organization of Type II Enzymes

A comprehensive survey of the cloned R-M systems was compiled by Geoff Wilson at New England Biolabs (NEB) in 1991 (Wilson 1991), showing different organizations, but always tight linkage between the REase and MTase genes. For example, the EcoRII genes were convergently transcribed from separate promoters on opposite DNA strands, in contrast to those encoding EcoRI (Kosykh et al. 1980, 1989; Som et al. 1987; Bhagwat et al. 1990; Reuter et al. 1999), which were in-line. By 1993, more than 60 REase and 100 MTase genes had been sequenced. The m5C-MTases clearly shared a common architecture (Lauster et al. 1989; Pósfai et al. 1989). The m4C- and m6A-MTases also showed similarities, although they were less pronounced (Klimasauskas et al. 1989; Lauster et al. 1989). The molecular weight (MW) of the REases, already known to vary considerably, proved to vary even more substantially: PvuII was only 18 kDa (Athanasiadis et al. 1990; Tao and Blumenthal 1992) and BsuRI was 66 kDa (Kiss et al. 1985). All this reinforced the idea that most enzymes evolved independently and used a variety of mechanisms to recognize DNA. Only in a few pairs was significant homology found (e.g., RsrI and EcoRI, which recognize the same sequence) (Aiken 1986; Stephenson et al. 1989).

The evidence that, like EcoRI, many Type II REases acted as dimers and the MTases as monomers was consolidated and supported by the first structures of Type II enzymes, those of EcoRI and EcoRV.

DNA Binding by Type II Enzymes

Virtually all Type II REases required Mg2+ for cleavage. However, many REases could form stable DNA–protein complexes in the absence of Mg2+—for example, BamHI, EcoRI, EcoRII, EcoRV, RsrI, and TaqI (Terry et al. 1987; Aiken et al. 1991a; Taylor et al. 1991; Xu and Schildkraut 1991; Gabbara and Bhagwat 1992; Zebala et al. 1992a). Fortuitously, both EcoRI and EcoRV first bound DNA before binding Mg2+ (Halford 1983; Taylor and Halford 1989). Hence, it was relevant to the reaction pathway to study the complexes formed in the absence of Mg2+. Gel-shift assays superseded the filter-binding method used for EcoRI (Halford and Johnson 1980; Jack et al. 1982; Chapter 5). Such assays revealed that binding of EcoRI to DNA with two recognition sites resulted in two complexes, with either one or two enzyme molecules bound to the DNA (Terry et al. 1985). Footprinting and a preferential cleavage assay were also used to monitor binding of REases to their substrates (Jack et al. 1982; Lu et al. 1983; Terry et al. 1983; Becker et al. 1988; Lesser et al. 1990; Roberts and Halford 1993).

EcoRI could bind very tightly to its recognition site, even on DNA molecules as long as 40 kb (Halford and Johnson 1980; Terry et al. 1983). Sets of synthetic oligonucleotide duplexes were used to study the effect of single-base changes in the recognition site (Lesser et al. 1990; Thielking et al. 1990). The equilibrium binding constants varied with each substitution, but these were much lower than at the recognition site—typically ∼5000-fold! BamHI and RsrI gave similar results (Aiken et al. 1991a; Xu and Schildkraut 1991), but a surprise came in the shape of EcoRV. In contrast to EcoRI, in the absence of Mg2+, EcoRV bound all DNA sequences with equal affinity (Fig. 1) irrespective of whether an EcoRV site was present or not. Gel shifts revealed a series of complexes due to the binding of 1, 2, 3, …, n molecules of protein per molecule of DNA (Taylor et al. 1991; Roberts and Halford 1993).


FIGURE 1. EcoRV binds all DNA sequences with equal affinity in the absence of Mg2+. Gel shifts with increasing concentrations of EcoRV added to 0.1 nM 32P-labeled DNA in EDTA buffer (no Mg2+). DNA (381 bp) with one EcoRV site. The enzyme concentrations (nM) are shown above each lane. The arrow marks the mobility of the free DNA. The same DNA without an EcoRV site gave the same results (not shown). The band above the arrow is the only band seen with specific DNA when Ca2+ was added (Vipond and Halford 1995; Halford 2013). With 50-, 100-, and 200-bp DNA, an increasing number of retarded bands appeared: 3, 6, and 12, respectively (Taylor et al. 1991; Halford 2013). (Reprinted, with permission, from Taylor et al. 1991, © American Chemical Society.)


This suggested that, in the absence of Mg2+, EcoRV bound the DNA indiscriminately. Also, the equilibrium constants all had the same value, even if the DNA contained an EcoRV site. (Note that although surprising, this is not relevant in vivo, as cells have ample Mg2+.) Like EcoRV, TaqI and Cfr9I (and also other DNA-binding proteins) could bind DNA without preference for their recognition sequences in the absence of Mg2+(Zebala et al. 1992a; Roberts and Halford 1993).

EcoRI appeared to bind nonspecifically to the DNA and then transfer to its specificity site. In general, such transfers could occur by sliding, hopping, or jumping (Fig. 2): Sliding is linear diffusion from nonspecific to specific sites; hopping refers to tiny dissociations/reassociations within the same DNA molecule; and jumping indicates total release of the DNA (von Hippel and Berg 1989). EcoRI probably used sliding, as this is fast: The association rate was very fast, and “too fast” for 3D “intersegmental transfer” (or “hopping”) (Jack et al. 1982; Terry et al. 1985; Halford 2013). Was hopping unlikely for REases with only one DNA-binding site?


FIGURE 2. Scheme for target site location (Fig. 1 in Halford and Marko 2004, as proposed by von Hippel and Berg in 1989 [von Hippel and Berg 1989; Halford and Marko 2004], legend adapted). Three routes for transfer of a protein from one site to another along a long DNA molecule: “sliding,” “hopping,” and “intersegmental transfer” (also called “jumping”). (Top) A protein might “slide” along the double helix from base pair to base pair without dissociation from the DNA. Repeated sliding results in 1D diffusion of the protein along the DNA. (Middle) If dissociation occurs, the protein might “hop” onto the same DNA a little further along. (Bottom) Beyond 150 bp (50 nm), the DNA can bend back on itself, and the enzyme could bind transiently to both DNAs, and then move to the other DNA, a process called “intersegmental transfer” or “jumping.” (Reprinted from Halford and Marko 2004.)


DNA Cleavage by Type II Enzymes

During this decade, increasing support was obtained for variation in restriction efficiencies at different sites and the role of flanking sequences (Ehbrecht et al. 1985; Terry et al. 1985; Nardone et al. 1986). In a few cases the effects of these flanking sequences were examined systematically (Taylor and Halford 1992; Yang and Topal 1992). For kinetic studies, it was important to use DNA with single sites, and the length of DNA in contact with the protein had to be longer than the recognition sequence, so as to include at least some flanking DNA (Lu et al. 1983; Becker et al. 1988; Rosenberg 1991). EcoRII and NaeI became the prototypes of REases that had to interact simultaneously with two or more copies of the recognition sequence before cleaving DNA (Krüger et al. 1988; Conrad and Topal 1989; Oller et al. 1991; Roberts et al. 2003). These enzymes contained two distinct binding sites for their recognition site: One proved to be an allosteric effector that activated the other for DNA cleavage. This explained why some EcoRII sites were refractory to restriction (Krüger et al. 1988). This activation was crucial for cleavage, as without this the EcoRII–DNA complex was stable even in the presence of Mg2+ (Gabbara and Bhagwat 1992). The two sequences could be supplied in cis or in trans (Conrad and Topal 1989; Pein et al. 1991; Topal et al. 1991; Yang and Topal 1992).

Plasmid DNA was conclusive to the determination of whether the enzyme cleaved one strand first or both strands at the same time. If the enzyme nicked a supercoiled DNA molecule, an open circle appeared on an agarose gel with a different mobility to that of a linear molecule (produced by a double-strand cut). In this way the reaction could be followed in time, providing an answer to the question of whether the enzyme first cut one strand and then the next or both strands at the same time. In the latter case no open circles would be produced. Gel-shift assays could also be used to study single turnover reactions (Halford 1983; Halford and Johnson 1983; Terry et al. 1987; Bennett and Halford 1989; Zebala et al. 1992a).

Studies on the kinetics of DNA cleavage by EcoRI yielded conflicting results (Roberts and Halford 1993), but results with EcoRV were clear: no open circles, hence double-strand cuts (Halford and Goodall 1988). However, changing normal assay conditions (low pH or low MgCl2) did produce open circles. The explanation was that in these cases Mg2+ bound only one of the two EcoRV subunits of the homodimer. As SalI produced similar results (Maxwell and Halford 1982), these results were consistent with a general requirement for coupled reactions to have Mg2+ bound to both subunits and that failure to do so would alter the mode of DNA cleavage (Bennett and Halford 1989; Hensley et al. 1990; Zebala et al. 1992a). Under optimal conditions the steady state reactions produced kcat values and Km values that approached the theoretical limit for kcat/Km in enzyme-catalyzed reactions (Roberts and Halford 1993).

In addition to plasmid DNA, short synthetic DNA duplexes (8–20 nt long) were used for kinetic studies. Time courses and analysis of reaction products (by electrophoresis, chromatography, or UV spectrophotometry [Aiken and Gumport 1991; Waters and Connolly 1992]) indicated that such short substrates posed fundamentally different problems for the REases compared to those composed of longer DNA molecules (Roberts and Halford 1993).

Specificity of Type II Enzymes

During the 1980s, research continued into the remarkable ability of the REases to discriminate the recognition sequence from all DNA sites. In vivo, a double-strand break may kill the cell and must be avoided, as only cognate sites are protected by methylation by the cognate MTase (Heitman and Model 1990b; Taylor et al. 1990; Smith et al. 1992). However, in vitro, REases cleaved DNA both at their recognition sites and some other sites (usually with one different base), albeit at a low level at noncognate sites under standard reaction conditions (Taylor and Halford 1989; Lesser et al. 1990; Thielking et al. 1990). As mentioned in Chapter 5, different conditions led to “star” (*) activity of, for example, EcoRI (Polisky et al. 1975) and other enzymes (Bennett and Halford 1989). To analyze this further, the cognate site for EcoRV (GATATC) in pAT153 (a derivative of pBR322) was compared with the preferred noncognate site (GTTATG). The latter site was flanked by alternating purines and pyrimidines, which conferred flexibility to the DNA structure (Taylor and Halford 1992). Under standard reaction conditions, the activity of EcoRV at this GTTATG was a formidable 106 times lower than at the cognate site, but this ratio changed to 103 in the presence of 10% DMSO and to just 6 in the presence of Mn2+ (Taylor and Halford 1989; Vermote and Halford 1992). In the case of EcoRI, systematic analysis of all nine possible single base pair substitutions in the recognition sequence caused a 105- to 109-fold reduction in kcat/Km for DNA cleavage (Lesser et al. 1990; Thielking et al. 1990). The enzymes usually cleaved noncognate sites via two successive nicks, even when the enzyme produced double-strand breaks at the cognate site (Barany 1988; Taylor and Halford 1989; Thielking et al. 1990).

E. coli DNA ligase was known to rapidly seal nicks in duplex DNA but joined double-stranded (ds) breaks much more slowly (Lehman 1974). This suggested that in vivo this enzyme might repair damage by REase action at the noncognate site but not at the cognate site. Would addition of ligase to the reaction mixture prevent product formation at the noncognate site? The answer to that was yes (Taylor and Halford 1989; Roberts and Halford 1993)!

Crystallography of Type II Enzymes1

The structure of EcoRI was the first to be published in 1986 (McClarin et al. 1986) but carried a mistake in the chain tracing and was revised 4 years later (Kim et al. 1990). One structure of EcoRI was solved in complex with a 12-bp duplex containing the recognition site (in the absence of Mg2+, which prevents cleavage [Kim et al. 1990]). John Rosenberg's group was lucky. On soaking the crystals with either Mg2+ or Mn2+, the enzyme became active and cleaved the duplex; in this way the postreactive enzyme-product complex could be analyzed, because the crystals remained intact (Rosenberg 2013)!

By 1993, crystal structures for EcoRV and BamHI had also been solved at high resolution (Fig. 3; Kim et al. 1990; Strzelecka et al. 1990, 1994; Winkler et al. 1993), with five others in progress (Roberts and Halford 1993). The structure of PvuII was published by the group of John Anderson at Cold Spring Harbor Laboratory in 1994 (Cheng et al. 1994), shortly after the publication of the review in the Nucleases book (Roberts and Halford 1993). This led to the idea of two types of structures, one with EcoRI and BamHI as prototypes producing sticky ends and the other pair with EcoRV and PvuII as prototypes that generate blunt end fragments (Jack et al. 1991; Roberts and Halford 1993; Winkler et al. 1993; Newman et al. 1994; Strzelecka et al. 1994). In the case of EcoRV, three structures were solved; free protein, the protein bound to a 10-bp duplex with the EcoRV site, and the enzyme bound to noncognate DNA (Winkler et al. 1993). This provided highly valuable information, and revealed the secret to the specificity of the enzyme for cognate sites.


FIGURE 3. Crystal structures of (A) EcoRI, (B) EcoRV, and (C) BamHI (Kim et al. 1990; Strzelecka et al. 1990; Winkler et al. 1993; Strzelecka et al. 1994). (Courtesy of Aneel Aggarwal.)


Protein Structures of the Type II REases EcoRI and EcoRV with DNA

The crystal structures of EcoRI and EcoRV have been extensively reviewed over the years. In 1993, the first results were summarized by Rich Roberts and Steve Halford and a shorter version of this text follows below (Roberts and Halford 1993).

a. EcoRI. The EcoRI–DNA enzyme complex has a single dyad symmetry relating the two subunits and the two halves of the palindromic DNA (Kim et al. 1990), as expected (Kelly and Smith 1970). At the dyad axis, the major groove of the DNA faces the protein. In each subunit, two arms extend from the main body of the protein to wrap around the DNA, but these remain within the major groove. Critical contacts to the DNA are made with a bundle of four α-helices, two from each subunit, aligned almost perpendicular to the DNA with the amino terminus of each helix poking into the major groove (Kim et al. 1990).

b. EcoRV. In its complexes with either specific or nonspecific DNA, EcoRV consists of two L-shaped subunits that interact with each other over a small surface area, to create a U-shaped dimer with a deep cleft between the subunits (Winkler et al. 1993). The fold of the polypeptide appeared to be completely different from that of EcoRI. In both complexes the DNA is located in the cleft, with its minor groove facing the base of the cleft (i.e., the opposite way around from EcoRI [but, as Alfred Pingoud was to point out later, the enzymes share a common β-sheet edge on to the active site]). The principal contacts to the DNA are made by two peptide loops per subunit. One loop, the R (recognition)-loop, is located toward the top of the cleft above the DNA. In the complex with the cognate site, the R-loop is positioned deep within the major groove, but in the nonspecific complex the R-loop is more distant from the DNA (Fig. 4). The second loop at the base of the cleft contains several glutamines (hence called the Q-loop) (Winkler et al. 1993). This loop approaches the minor groove of the DNA and contacts primarily phosphates rather than bases. The available structure suggests that EcoRV must undergo a series of conformational changes (Winkler et al. 1993), a prediction that would prove to be correct.


FIGURE 4. Distortion of the 12-bp duplex DNA containing the EcoRI recognition site by EcoRI (Kim et al. 1990). (Left) The 12-bp duplex: B-DNA on its own. (Right) The 12-bp duplex + EcoRI: The enzyme kinks and distorts the DNA (Rosenberg 2013). (Courtesy of John Rosenberg.)


The First DNA Structures in Protein–DNA Complexes of Type II REases

In the case of the EcoRIDNA complex (with a 12-bp duplex 5′CGCGAATTCGCG), the DNA is distorted from the regular B-DNA structure (Fig. 4; Kim et al. 1990). The distortion is primarily an untwisting at the center of the sequence, with concomitant unstacking of the central 2 bp. This widens the major groove, thus improving access to the bases.

In the case of EcoRV, a 10-bp duplex (5′GGGATATCCC) is radically distorted from B-DNA (Winkler et al. 1993) The most marked feature of the distortion is a sharp bend, directed toward the protein, in the axis of the DNA helix at the center of the recognition site. Like EcoRI, the middle 2 bp in the recognition site are unstacked, but, in this case, the roll is in the opposite direction. The bound substrate for EcoRV has a deep and narrow major groove and a correspondingly shallow minor groove. The Winkler group would further refine the structure of EcoRV (see Winkler and Prota 2004 for review and Chapter 7).

DNA–Protein Interfaces of Type II REases EcoRI and EcoRV

Each base pair in the duplex DNA possesses, on its edge facing the major groove, a unique array of three hydrogen-bonding functions (Seeman et al. 1976). The 5-methyl group of thymidine can also be used to distinguish DNA sequences. EcoRI uses 16 (out of 18 possible) hydrogen-bonding functions in its 6-bp recognition sequence and makes van der Waals' contacts with all of the thymidine methyl groups (it was at that time unique to have so many contacts). In contrast, EcoRV only uses four amino acids for sequence-specific binding (see Roberts and Halford 1993 for further details).

DNA Cutting by Type II REases EcoRI and EcoRV

The overall structure of EcoRI bound to its recognition sequence was radically different from that of EcoRV. But one striking similarity was already noted, raising the possibility that these enzymes did use the same mechanism to hydrolyze the phosphodiester bond. In both enzymes the bond cleaved was surrounded by a proline, two acidic residues, and a lysine in the same relative positions (Selent et al. 1992; Winkler 1992). The sequence motif PD …(D/E)XK was noted in several other REases, but its significance still had to be established (Anderson 1993); by 2001, seven out of 12 REases had been shown to have the PD motif and nine out of 12 REases the (D/E)XK motif in their catalytic centers (Pingoud and Jeltsch 2001). The precise mechanism of the chemical reaction catalyzed by EcoRI, EcoRV, and almost all other REases remained to be determined (namely, deprotonation of a H2O molecule, in-line attack by the resulting OH leading to formation of an unstable pentacovalent phosphate intermediate, and dissolution of the 3′ bond by protonation is well established, but the details, and perhaps the order of these events, likely vary from enzyme to enzyme).

DNA Recognition Functions

To dissect recognition and catalysis by REases, one could mutate either the protein or the DNA. To alter the DNA, base analogs (Brennan et al. 1986), or replacement of phosphates with phosphorothioates (Connolly et al. 1984), were employed. In this way it was shown that EcoRI was less active toward GAAUTC than GAATTC, suggesting a role for the methyl group on the inner thymidine (Brennan et al. 1986). Replacing Gln-115 in the protein gave a mutant with the same (lower) activity toward both sequences (Jeltsch et al. 1993). This tied in with the crystal structure: The methylene side chain of Gln-115 interacted hydrophobically with the methyl group on the inner thymidine (Rosenberg 1991).

Altered Enzymes

Mutations in the R-loop of EcoRV all concurred with the crystal structure (Thielking et al. 1991; Vermote and Halford 1992; Halford et al. 1993). Initially, this was not the case for EcoRI, and this necessitated a revision of the original structure, as mentioned above (McClarin et al. 1986; Kim et al. 1990). However, these “wrong” mutants that retained specificity for the EcoRI site were useful, as they did affect nuclease activity (“secondary” functions; Rosenberg 1991; Heitman 1992).

In addition to specific mutagenesis, random mutagenesis followed by genetic selection yielded useful information about several REases (Yanofsky et al. 1987; King et al. 1989; Xu and Schildkraut 1991). Although many mutations affected the protein's affinity for DNA, for some the specificity remained unchanged. One interesting mutant of EcoRI had a Glu-111 → Gly-111 substitution, identifying this residue as a key amino acid in catalysis (King et al. 1989). A clever strategy to isolate mutants was devised by Joe Heitman and Peter Model (Heitman and Model 1990a). A strain carrying M·EcoRI and the lacZ gene under the control of an SOS-inducible promoter was transformed with mutagenized DNA. Induction of the SOS response indicated cutting at noncognate sites. This procedure yielded promiscuous mutants that were more active at noncognate sites than the wild-type enzyme (Heitman and Model 1990a). Moreover, their properties could be accounted for by reference to the crystal structure (Heitman 1992).

Altered Substrates

Base analogs alter the hydrogen-bonding interactions between bases and protein (Seeman et al. 1976). This allowed analysis of alterations in DNA cleavage by EcoRI, EcoRV, RsrI, and TaqI (Brennan et al. 1986; McLaughlin et al. 1987; Newman et al. 1990; Aiken et al. 1991b; Zebala et al. 1992b; Lesser et al. 1993; Waters and Connolly 1994). In the case of EcoRI and EcoRV, the loss of almost any one of the functional groups in the DNA (i.e., those that interacted in the crystal structure with the protein) reduced DNA cleavage rates relative to the cognate oligonucleotide. But other changes could also alter enzyme activity, indicating as yet unpredictable cooperativity between regions within the enzyme.

Incorporation of phosphorothioates in the DNA allowed the study of protein–DNA backbone interactions. Using three dNTPs and one dNTPαS (Potter and Eckstein 1984) on a ss template produced duplex DNA with the phosphorothioates in one strand only (the new strand). A phosphorothioate at the scissile bond reduced or abolished DNA cleavage, and thus these substrates amplified the difference between cleaving the first and second strands of the DNA (Potter and Eckstein 1984). A drawback was that phosphorothioates elsewhere in the recognition sequence or flanking DNA also reduced enzyme activity (Olsen et al. 1990; Lesser et al. 1992). Therefore, another method used chemical synthesis of oligonucleotides with the phosphorothioate placed at a specified site in the chain (Connolly et al. 1984). This method yielded valuable information about the stereochemistry of phosphodiester hydrolysis and interaction with phosphates in the distorted backbone (Connolly et al. 1984; Grasby and Connolly 1992; Lesser et al. 1992).

Coupling Recognition to Catalysis

Taking all results together, the question arose: Could one now account for the ability of REases to distinguish their recognition sites from all other DNA sequence?

a. EcoRI. In the case of EcoRI, several processes appeared to be involved in the recognition of the cognate site and the rejection of noncognate sites (Lesser et al. 1990; Heitman 1992). The combined data suggested that the conformation of the DNA (and/or protein) in the specific complex differed from that in the nonspecific complex (Lesser et al. 1990, 1993; Thielking et al. 1990; Heitman 1992), and that EcoRI thus generated its specificity by a subtle combination of both direct and indirect readouts (Roberts and Halford 1993).

b. EcoRV. In contrast, EcoRV showed no preference for binding to its recognition site (Taylor et al. 1991), although in this case also, distortion of the enzyme-bound DNA played a key role. The finding that EcoRV had a high affinity for Mg2+ when bound to the cognate, but not at nonspecific sites, was probably the main factor determining the different rates of DNA cleavage (Vipond and Halford 1993). This high affinity for Mg2+ at only cognate sites could be explained on the structures of the enzyme DNA complexes (Winkler et al. 1993; see Vermote and Halford 1992; Halford et al. 1993; Roberts and Halford 1993 for further discussion). The crucial role for Mg2+ at cognate sites was supported by the lack of discrimination by EcoRV in the presence of Mn2+ (Vermote and Halford 1992). Both cognate and noncognate complexes with EcoRV had high affinities for Mn2+ (with a ratio of DNA cleavage of 6, whereas it had been ∼106 with Mg2+ as cofactor).

The Structure of the PvuII REase with Cognate DNA

As mentioned above, the structure of PvuII was published shortly after the comprehensive review in the Nucleases book by Roberts and Halford (Roberts and Halford 1993). The enzyme binds as a dimer to the DNA (Cheng et al. 1994). The enzyme has three domains for dimerization, catalysis, and DNA recognition, respectively (Fig. 5). The catalytic domain resembles that of other REases and appears to share a conserved sequence with the active sites of EcoRI and EcoRV, whereas the direct contacts between the protein and the base pairs of the PvuII recognition site occur exclusively in the major groove via two antiparallel β-strands from the sequence recognition region of the protein. The catalytic regions of these REases appear to have been conserved in evolution (presumably reflecting common cleavage to yield 5′P and 3′OH groups); the subunit interface and DNA sequence recognition domains apparently are not conserved. The fact that EcoRI and BamHI produce four-base ssDNA overhangs on the major groove side of the DNA, whereas both EcoRV and PvuII generate blunt ends on the minor groove side of the DNA, may be the reason that with EcoRI and BamHI their DNA-binding cleft of the protein dimer faces the major groove, whereas in the case of EcoRV and PvuII the binding cleft faces the minor groove—thus, the difference in the directions in which EcoRI and EcoRV may be related to the positions of their scissile bonds (Anderson 1993). “The orientation of the catalytic region may need to be stabilized by the dimerization region for efficient cleavage, requiring that the DNA binding cleft face the side of the DNA the scissile bonds are on. This hypothesis is supported by the structure of PvuII-DNA, in which the enzyme binds to DNA from the minor groove side” (Cheng et al. 1994). The prediction would be that enzymes with 3′ extensions would approach DNA from the minor groove side as well (Cheng et al. 1994).

With a MW of only 18 kDa, PvuII was the shortest of these four structurally characterized REases, which could explain why the B-form DNA is not distorted after protein binding, whereas the same region in EcoRV is much bulkier. In the latter case, this might require a kink in the DNA to ensure similar distances between the catalytic residue(s) and the target, while at the same time avoiding steric collision between the protein and DNA (Cheng et al. 1994).


Structural Genes and Family Relationships of Type I Enzymes

During the 1980s, various Type I enzymes were cloned and sequenced, both from E. coli and Salmonella and from other species such as Citrobacter freundii (Bickle 1993). As the number of REases grew, Roman numerals were added to the enzymes (e.g., EcoK became EcoKI). In all cases, the results of genetic analysis corroborated the earlier findings: three hsd genes, tightly clustered, and transcribed from two promoters, one in front of hsdM (cotranscribing hsdS), and the other in front of hsdR (Sain and Murray 1980; Gough and Murray 1983; Suri and Bickle 1985; Loenen et al. 1987; Cowan et al. 1989; Kannan et al. 1989; Price et al. 1989). Although tight linkage was conserved in all cases, the relative order of the transcription units was not. Moreover, many of the new systems showed little or no homology with EcoKI or EcoBI, or with each other, whether by complementation analysis, DNA hybridization, or immunological cross-reactivity (Murray et al. 1982; Price et al. 1987a). This led to their classification into three families with 19 members in total by 1993 (Bickle 1993): Type IA (prototypes EcoKI and EcoBI), Type IB (prototype EcoAI), and Type IC (prototype R124I, variously called StyR124I and EcoR124I). At the time, all Type IA and Type IB genes were located on the bacterial chromosome near serB; the Type IC genes were plasmid-encoded. But all proteins were similar in structure and required ATP and SAM for activity. In later years, two more families were added (Type ID [prototype StyBLI and KpnAI (Titheradge et al. 2001; Murray 2002; Kasarjian et al. 2004)]) and Type IE (prototype KpnBI [Chin et al. 2004]), whereas the location on the chromosome or episome would prove not as strict as originally thought.

Sequence Homologies within and between Families of Type I Enzymes

Within a family, DNA sequences of the hsdM and hsdR genes are quite highly conserved (Murray et al. 1982; Daniel et al. 1988; Gubler et al. 1992; Sharp et al. 1992) with the strongest sequence identity ∼95% for hsdM (Sharp et al. 1992). In contrast, the hsdS genes contained two extensive regions of nonhomology, one at the 5′ end of the genes and the other toward the 3′ end, flanked by homologous regions (Gough and Murray 1983; Cowan et al. 1989; Gubler et al. 1992). The two nonhomologous “hypervariable” regions encoded protein domains that each recognized one-half of the recognition sequence. Those enzymes with the same 5′ moiety of their recognition sequence showed ∼50% identity in the amino-terminal hypervariable region (e.g., StySBI, EcoAI, and EcoEI all recognized 5′GAG) (Cowan et al. 1989). This was the first formal evidence for the current model of how Type I enzymes specifically recognize DNA. The conserved regions outside these nonhomologous regions were thought to provide protein–protein interactions with the HsdM and HsdR proteins.

Between families, the hsdM genes encoded ∼26%–33% identical amino acids, a degree of homology high enough to exclude an independent origin of the different Type I families. Some of these residues were found to be conserved in all m6A MTases (Lauster et al. 1987; Guschlbauer 1988; Narva et al. 1988; Smith et al. 1990).

Evolution of DNA Specificity of Type I Enzymes

Evolution of DNA Specificity by Homologous Recombination within the hsdS Gene of Type I Enzymes

In 1976, Len Bullas in the laboratory of Stuart Glover identified a new specificity, StySQ, after transduction of the hsd genes of S. potsdam (StySP) into S. typhimurium (StySB; a Roman I was later added to the names of all three enzymes) (Fig. 6A; Bullas et al. 1976). DNA heteroduplex analysis and DNA sequencing showed that indeed the StySQ system arose by recombination within the central conserved region of the parental hsdS genes (Fuller-Pace et al. 1984; Fuller-Pace and Murray 1986). The recognition sequences of the three proteins GAG(N6)RTAYG (StySBI), AAC(N6)GTRC (StySPI), and AAC(N6)RTAYG (StySQI) confirmed the hybrid nature of StySQI (Nagaraja et al. 1985a,b). These findings immediately led to the idea that hsdS genes encoded two DNA-binding domains: an amino-terminal domain that recognized the 5′ part of the recognition sequences and a carboxy-terminal domain that recognized the 3′ part. Thus, recombination in the central conserved fragment would allow domain swapping generating new sequence specificity. In line with this result, the reciprocal recombinant (StySJIb) between the StySBI and StySPI hsdS genes recognized GAG(N6)GTRC (Gann et al. 1987).


FIGURE 5. Structure of PvuII. Two subunits are shown in gray or in color, respectively, with a ball-and-stick model of the bound cognate DNA segment. Three regions are colored in red, green, and blue for dimer interaction, catalysis, and DNA recognition, respectively. (A) Front view of the protein–DNA complex. (B) Side view of the protein–DNA complex from an angle as indicated in A. (C) Same view as in B of dimer structure without DNA. The interaction between two H85 side chains closes off the DNA-binding cleft. (D) Locations of secondary structural elements in the amino acid sequence. (Reprinted from Cheng et al. 1994, with permission from EMBO.)


FIGURE 6. Evolution of Type I enzymes with new specificities. (A) Recombination between hsdS genes produces hybrid genes and chimeric S polypeptides. StySP1 and StyLTIII are naturally occurring Type I R-M systems. StySQ and StySJ have hybrid hsdS genes (Fuller-Pace et al. 1984; Gann et al. 1987). The regions originating from StySP1 are hatched and those originating from StyLTIII are stippled. Reassortment of the target recognition domains (TRDs) accordingly gave rise to recombinant recognition sequences (Nagaraja et al. 1985a; Gann et al. 1987). Site-directed mutagenesis of the central conserved region of the StySQ hsdS gene produced StySQ*, comprising only the amino-terminal variable region from StySP1 and the remainder from StyLTIII. The StySQ* target sequence confirms that the amino-terminal variable region is in fact a TRD responsible for recognition of the trinucleotide component of the sequence (Cowan et al. 1989). (B) Sequence specificity may also be altered by changing the length of the nonspecific spacer of the target sequence. The S polypeptides of EcoRI241 and EcoRI2411 differ only in the number of times a short amino acid motif (X = TAEL) is repeated within their central conserved regions (Price et al. 1989), resulting in extension of the spacer in the target sequence from 6 nt (N6) for EcoRI241 to N7 for EcoRI2411. The recognition sequence of EcoDXXI also contains a nonspecific spacer of 7 nt, corresponding to three TAEL repeats in its HsdS polypeptide (Gubler et al. 1992). Chimeric S polypeptides recognize the predicted target sequences (Gubler et al. 1992). (Reprinted from Murray 2002, with permission from Microbiology Society.)


Evolution of DNA Specificity by Unequal Crossing-Over within the hsdS Gene of Type I Enzymes

In addition to domain swaps via homologous recombination, another spontaneous change in specificity was found with plasmid EcoR124. Cells carrying this plasmid could express either EcoR124 (renamed EcoR124I) or EcoR124/3 (renamed EcoR124II), with recognition sequences GAA(N6)RTCG and GAA(N7)RTCG, respectively (Price et al. 1987b). This proved to be due to unequal crossing-over in the nonspecific spacer separating the two specific parts (Fig. 6B; Price et al. 1989). Surprisingly, this crossing-over occurred at a specific site in the conserved central region, where a 12-bp sequence (encoding four amino acids, TAEL) was repeated twice in EcoR124I and three times in EcoR124II. This increased the spacer from 6 to 7 bp, rotating the two domains by 36°, a far from trivial matter with respect to enzyme recognition. This effect of the increase in length of the conserved region on that of the spacer separating the two DNA recognizing domains led to the model in which the length, but not the exact amino acid sequence, was important for function. Mutations in the DNA encoding the TAEL repeats indeed did not affect activity or specificity, whereas altering the length of the repeated region did have drastic effects (Gubler and Bickle 1991).

Variants with either one or four copies of the repeat, for example, were virtually inactive in restriction (105–106 times less active than the wild type). However, they were still efficient MTases. The mutant with a single copy of the repeat methylated the EcoR124I recognition sequence, whereas a mutant with four repeats methylated both the EcoR124I and the EcoR124II sequences but would not methylate a putative recognition sequence with 8 bp in the nonspecific spacer (Gubler and Bickle 1991). It was speculated that the severe effects on restriction but not on modification might relate to the structure of the central conserved regions of the hsdS gene products. This region of the protein most likely had a dual function: a spacer between the DNA-binding domains of the protein, but it was also needed for interaction with the HsdR restriction subunit.

The idea that hsdS genes contained two DNA-binding domains separated by a spacer region whose length determined the number of base pairs separating the two components of the recognition sequences was tested using the Type IC enzyme, EcoDXXI (Piekarowicz et al. 1985; Skrzypek and Piekarowicz 1989). This enzyme recognized TCA(N7)RTTC and contained three copies of the 12-bp repeated sequence in the conserved region (Meister et al. 1993). Hybrids between the two halves of the hsdS genes of ecoDXXI and ecoR124 with either two or three copies of the 12-bp repeated sequence all were active in restriction and recognized DNA sequences consistent with this model.

Evolution of DNA Specificity by Transposition within the hsdS Gene of Type I Enzymes

A third mode of changing specificity of a Type I enzyme was also discovered in EcoDXXI. A Tn5 derivative in the ecoDXXI hsd region appeared to have an altered DNA sequence specificity. It turned out that Tn5 had inserted into the hsdS gene, just 3′ to the central conserved region. The hsdS gene product produced by the mutant was much shorter but retained the amino-terminal part of the protein. The sequence recognized was TCA(N8)TGA (Fig. 6B; Meister et al. 1993). The rotational symmetry of the site and the length of the truncated HsdS protein led to the inescapable conclusion that the enzyme apparently assembled two copies of the truncated HsdS protein (called EcoDXXsI) (Loenen 2003).

Enzyme Structures and Mechanisms of Type I Enzymes

As mentioned earlier, purification of the EcoKI and EcoBI enzymes resulted in various oligomeric protein complexes. New Type I enzymes of any of the three families behaved similarly (Suri et al. 1984a,b; Price et al. 1987a; Gubler and Bickle 1991; Taylor et al. 1992). Enzymes could be MTases with or without the REase, with stoichiometries reported for the HsdM and HsdS subunits as 2:1 or 1:1 for the methylase on its own, and most likely 2:2:1 for the pentameric REase, although some enzymes were unstable and aggregated or fell apart upon storage (Bickle 1993). The HsdS subunit of EcoKI could not be purified on its own, but that of EcoR124I was shown to be a sequence-specific DNA-binding protein without enzymatic activity (Kusiak et al. 1992).


The genes encoding the modification and restriction subunits of EcoKI were the first Type I HsdR and HsdM proteins to be sequenced and had sequence motifs typical of SAM- and ATP-binding proteins (Loenen et al. 1987). EcoKI had no affinity for DNA in the absence of cofactor SAM. After allosteric activation, the enzyme bound with high affinity to both modified and unmodified recognition sites (Bickle et al. 1978). In the absence of ATP, these complexes with DNA were relatively stable on both modified (t1/2 = 6 min), and unmodified sites (t1/2 = 22 min) (Yuan et al. 1975). The enzyme would modify the second strand of hemimethylated DNA with overall first-order reaction kinetics with rate constants of 3 × 10−3 sec−1 in the presence of ATP and 4 × 10−4 sec−1 in its absence (the enzyme could also modify unmethylated DNA, with a rate constant of 6 × 10−5 sec−1; Suri et al. 1984a; Bickle 1993). The addition of ATP to complexes with unmethylated DNA set the cleavage mode in action. After a massive conformational change, the enzyme remained bound to the recognition site, but cleaved randomly far from the recognition sites (400–7000 bp; Chapter 5). This was followed by massive ATP hydrolysis. This last aspect of the reaction mechanism was an utter mystery and generated considerable controversy (Bickle 1993).

Cleavage Models for Type I Enzymes

Different models were proposed for the mechanism whereby the enzyme cleaves DNA at loci distant from the recognition sequence. Based mainly on EM data on the EcoBI and EcoKI enzymes, the enzymes tracked along the DNA, forming ever-larger loops until the cleavage site was reached. Although EcoBI cleaved only to one side of the asymmetric recognition site (Rosamond et al. 1979), EcoKI translocated and cleaved the DNA in both directions (Yuan et al. 1980a). The “collision” model, which stated that restriction required at least two recognition sites in the DNA, was proposed in 1988 by William (Bill) Studier. An enzyme molecule would bind to each site, and the molecules would move along the DNA until they bumped into each other, at which point the DNA would be cut (Fig. 7; Studier and Bandyopadhyay 1988; Studier 2013).


FIGURE 7. The collision model for DNA breakage (Studier and Bandyopadhyay 1988). EcoKI bound to target sequences translocates DNA toward itself. Collision blocks translocation and stimulates the nicking of both DNA strands. REase activity may be stimulated when translocation is impeded by some other protein or structure (Janscak et al. 1999). (Reprinted from Murray 2002, with permission from Microbiology Society.)


How did Bill Studier arrive at this model? He studied phage T7 and tried to find out why this phage was resistant to cleavage by EcoKI (details about this and other restriction evasion strategies by plasmids and phages [extensively reviewed in Krüger and Bickle 1983; Bickle and Krüger 1993] will follow later in this book). Bill identified an early function (the product of gene 0.3, called Ocr [overcoming restriction]) that blocked EcoKI. Mutations in this gene led to restriction of T7 DNA by EcoKI. However, he found distinct restriction fragments on the gel, rather than a smear caused by random fragmentation. Did this mean that two enzymes would bind two recognition sites and translocate the DNA until they met in the middle? That cleavage occurred when they met, stalled, and cleaved (Studier 2013)? The model was very appealing, but if this collision model were true, how did one explain cleavage of DNA molecules with single recognition sites?

Whatever the exact mechanism, DNA cleavage was a two-step process: a nick in one strand, followed by a second cut, probably by another EcoKI molecule. In addition to the curious ATP hydrolysis, the DNA ends produced by Type I enzymes were a mystery too. They could not be labeled with polynucleotide kinase and had long 3′ protrusions (Eskin and Linn 1972; Murray et al. 1973; Endlich and Linn 1985).

Maintenance versus De Novo Methylation by Type I Enzymes

There were additional unexplained observations and differences between different Type I enzymes. Whereas EcoKI and EcoBI and other Type IA enzymes preferentially methylated hemimethylated substrates, the reaction with nonmethylated DNA was slow (see above), and the reaction with both substrates was inhibited by ATP (Suri and Bickle 1985). Type 1C enzymes showed the same substrate preferences as Type IA enzymes; however, for these enzymes, the reaction was stimulated by ATP (Price et al. 1987a). The Type IB enzyme EcoAI showed a completely different pattern. Hemimethylated and nonmethylated substrates were modified equally well, but the reaction was completely dependent on ATP (Suri and Bickle 1985).

An interesting finding was the small Ral (restriction alleviation) protein of lambda, which rescued superinfecting phages from restriction by EcoKI (Zabeau et al. 1980; Loenen and Murray 1986). (An analogous protein Lar is present on the cryptic lambdoid prophage, Rac [King and Murray 1995].) Ral appeared to switch EcoKI from a maintenance to a de novo MTase by enhancing methylation of unmodified sites (Loenen and Murray 1986). Perhaps even more interesting were the Ral-independent EcoKI m* mutants isolated in Noreen Murray's laboratory (Kelleher et al. 1991). They mapped to a few specific places in the HsdM subunit, in line with later evidence that assigned the discriminatory capacity with respect to the methylation state of the DNA to the M2S complex (see, e.g., Loenen 2003 for details).

Biochemical experiments were carried out with m* mutants (LL113Q, L134V [Winter 1997])—for example, the kinetic constants for L113Q versus wild type were determined on unmethylated DNA and on hemimethylated DNA with the methyl group on either the top or bottom strand (Tables 4.3 and 4.4 in Winter 1997).


Occurrence and Genetics of Type III Enzymes

By 1993, only four members of this family had been identified. In addition to EcoP1I, EcoP15I, and HinfIII (Chapter 5), the chromosomal StyLTI system was present in many Salmonella strains (Arber and Dussoix 1962; Arber and Wauters-Willems 1970; Colson et al. 1970; Piekarowicz and Kalinowska 1974; Bullas et al. 1980). The enzymes were encoded by the mod and res genes (Iida et al. 1983). mod encoded the MTase and recognized the DNA specificity site. The res gene product was essential for restriction in a complex with the MTase but lacked enzymatic activity on its own. The data on transcription were controversial. Were both genes transcribed from a single promoter located in front of the mod gene, as judged from transposon insertion experiments (Iida et al. 1983)? Or did other in vitro and in vivo studies suggest a more complex situation (Iida et al. 1983; Sharrocks and Hornby 1991)?

The sequences of the mod genes of EcoP1I, EcoP15I, and StyLTI, as well as the res genes of EcoP1I and StyLTI, were known (Humbelin et al. 1988; Dartois et al. 1993). DNA heteroduplex analysis indicated strong homology between the res genes of EcoP15I and EcoP1I (Iida et al. 1983). The mod genes were mosaics of conserved and nonconserved regions, a structure reminiscent of that of the Type I hsdS genes: a totally dissimilar central region and conserved 5′ and 3′ regions. It was thought that the conserved regions encoded protein domains that interacted with the res gene product and that the variable regions encoded sequence-specific DNA-binding domains. In line with this, mutations that led to loss of modification without affecting restriction (Rosner 1973) mapped in this region (Humbelin et al. 1988). These mutants were shown to have lost the ability to bind cofactor SAM (Rao et al. 1989a). The EcoP1I and StyLTI res sequences showed surprisingly little homology apart from a stretch of 50 amino acids toward the center, where the two proteins were virtually identical—a good candidate region for the interaction between the restriction and modification subunits (Dartois et al. 1993).

Enzyme Mechanisms of Type III Enzymes

All four REases recognized asymmetric DNA sequences and cut 25–26 bp downstream from the sequence: EcoP1I (AGACC), EcoP15I (CAGCAG), HinfIII (CGAAT), and StyLTI (CAGAG) (Bachi et al. 1979; Hadi et al. 1979; Piekarowicz et al. 1981; De Backer and Colson 1991). Surprisingly, these sequences had only adenine in the strand shown, and thus half of the sites following DNA replication completely lack modification. Yet, this was apparently not lethal to the cell. Experiments with phage T7 would solve that mystery in 1992, as detailed later.

The available data on the reaction mechanism of the Type III enzymes came mainly from studies on EcoP1I and EcoP15I. They were quite different from the Type I enzymes: They required ATP for restriction but did not hydrolyze as much ATP as the Type I enzymes. SAM was not essential, but stimulated cleavage, leading to competition between restriction and modification (Haberman 1974; Risser et al. 1974; Reiser and Yuan 1977; Kauc and Piekarowicz 1978). Both ATP and SAM were allosteric effectors for DNA cleavage, and nonhydrolyzable ATP analogs only weakly supported cleavage (Yuan and Reiser 1978; Yuan et al. 1980b).

Also in contrast to the Type I enzymes, the Type III MTase could bind either DNA or SAM first. An unusual feature of the kinetics of methylation was that the enzyme was inhibited by SAM concentrations of only slightly more than the Km values for SAM (Rao et al. 1989b). This suggested nonproductive binding of SAM to the methylated DNA–enzyme complex. A mutant, S240A, supported this idea, being more active than wild-type enzyme because it could no longer be inhibited by substrate SAM. This serine was important for activity, as a S240P mutation led to loss of SAM binding (Rao et al. 1989a).

DNA Cleavage by Type III Enzymes

As mentioned above, EcoP1I, EcoP15I, and StyLTI have a methylatable adenine in the top strand only (Hadi et al. 1979; Meisel et al. 1991). How did cells survive after DNA replication? The answer to this mystery came from the sequence of phage T7. This phage was not restricted by EcoP15I, although it contained 36 EcoP15I sites (Dunn and Studier 1983). Interestingly, these sites were all in the same orientation: the CAGCAG sequence in one DNA strand and its CTGCTG complement in the other strand (Schroeder et al. 1986). Did this make T7 DNA refractory to EcoP15I cleavage? If so, this would mean that EcoP15I restriction should require two recognition sites, and also that these two recognition sites should be in inverse orientation! M13 constructs with different numbers and orientations of EcoP15I sites were made and proved this idea to be correct. Single or multiple sites with any orientation could be methylated, but only unmodified sites in inverse orientation could be restricted (Meisel et al. 1992). Hence, the true recognition site for the EcoP15I REase complex consisted of a twofold rotationally symmetrical sequence interrupted by a nonspecific spacer of variable length. As all newly replicated sites would be in the same orientation, unmodified sites were not cleaved, but modified.

Although little evidence was available at the time, there was good reason to believe that this phenomenon would be a common characteristic of the Type III enzymes. This would be in line with some earlier observations with EcoP1I on phage lambda (Arber et al. 1963; Hattman et al. 1978; Bickle 1993).


Thomas (Tom) Bickle ends his 1993 review (Bickle 1993) with: “I believe that the prevalence of DNA restriction systems is a sign of genetic selection operating at the population level: A population whose individual members can prevent phage propagation (even if the infected individual is killed in the process) is fitter than one that cannot. Paradoxically, selection for function takes place in those cells in which the function is not used.”


Aiken CR, Gumport RI. 1991. Base analogs in study of restriction enzyme–DNA interactions. Methods Enzymol 208: 433–457. 10.1016/0076-6879(91)08023-B

Aiken C, Milarski-Brown K, Gumport RI. 1986. RsrI and EcoRI are two different restriction endonucleases that recognise the same DNA sequence. Fed Proc 45: 1914.

Aiken CR, Fisher EW, Gumport RI. 1991a. The specific binding, bending, and unwinding of DNA by RsrI endonuclease, an isoschizomer of EcoRI endonuclease. J Biol Chem 266: 19063–19069.

Aiken CR, McLaughlin LW, Gumport RI. 1991b. The highly homologous isoschizomers RsrI endonuclease and EcoRI endonuclease do not recognize their target sequence identically. J Biol Chem 266: 19070–19078.

Anderson JE. 1993. Restriction endonucleases and modification methylases. Curr Opin Struct Biol 3: 24–30. 10.1016/0959-440X(93)90197-S

Arber W, Dussoix D. 1962. Host specificity of DNA produced by Escherichia coli. I. Host controlled modification of bacteriophage λ. J Mol Biol 5: 18–36. 10.1016/S0022-2836(62)80058-8

Arber W, Wauters-Willems D. 1970. Host specificity of DNA produced by Escherichia coli. XII. The two restriction and modification systems of strain 15T. Mol Gen Genet 108: 203–217. 10.1007/BF00283350

Arber W, Hattman S, Dussoix D. 1963. On the host-controlled modification of bacteriophage λ. Virology 21: 30–35. 10.1016/0042-6822(63)90300-3

Athanasiadis A, Gregoriu M, Thanos D, Kokkinidis M, Papamatheakis J. 1990. Complete nucleotide sequence of the PvuII restriction enzyme gene from Proteus vulgaris. Nucleic Acids Res 18: 6434. 10.1093/nar/18.21.6434

Athanasiadis A, Vlassi M, Kotsifaki D, Tucker PA, Wilson KS, Kokkinidis M. 1994. Crystal structure of PvuII endonuclease reveals extensive structural homologies to EcoRV. Nat Struct Biol 1: 469–475. 10.1038/nsb0794-469

Bachi B, Reiser J, Pirrotta V. 1979. Methylation and cleavage sequences of the EcoP1 restriction-modification enzyme. J Mol Biol 128: 143–163. 10.1016/0022-2836(79)90123-2

Barany F. 1988. The TaqI ‘star’ reaction: strand preferences reveal hydrogen-bond donor and acceptor sites in canonical sequence recognition. Gene 65: 149–165. 10.1016/0378-1119(88)90452-0

Barsomian J, Landry D, Moran L, Slatko B, Feehery G, Jack W, Nwanko D, Wilson G. 1990. Organisation and evolution of Type IIS MTases. In Second NEB Workshop on Biological DNA Modification, Berlin, Vol. 11.

Becker MM, Lesser D, Kurpiewski M, Baranger A, Jen-Jacobson L. 1988. “Ultraviolet footprinting” accurately maps sequence-specific contacts and DNA kinking in the EcoRI endonuclease–DNA complex. Proc Natl Acad Sci 85: 6247–6251. 10.1073/pnas.85.17.6247

Bennett SP, Halford SE. 1989. Recognition of DNA by type II restriction enzymes. Curr Top Cell Regul 30: 57–104. 10.1016/B978-0-12-152830-0.50005-0

Berg OG, von Hippel PH. 1985. Diffusion-controlled macromolecular interactions. Annu Rev Biophys Biophys Chem 14: 131–160. 10.1146/annurev.bb.14.060185.001023

Bhagwat AS, Johnson B, Weule K, Roberts RJ. 1990. Primary sequence of the EcoRII endonuclease and properties of its fusions with β-galactosidase. J Biol Chem 265: 767–773.

Bickle TA. 1993. The ATP-dependent restriction enzymes. In Nucleases, 2nd ed. (ed. Linn SM, LLoyd SR, Roberts RJ), pp. 88–109. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Bickle TA, Krüger DH. 1993. Biology of DNA restriction. Microbiol Rev 57: 434–450.

Bickle TA, Brack C, Yuan R. 1978. ATP-induced conformational changes in the restriction endonuclease from Escherichia coli K-12. Proc Natl Acad Sci 75: 3099–3103. 10.1073/pnas.75.7.3099

Bitinaite J, Maneliene Z, Menkevicius S, Klimasauskas S, Butkus V, Janulaitis A. 1992. Alw26I, Eco31I and Esp3I–type IIs methyltransferases modifying cytosine and adenine in complementary strands of the target DNA. Nucleic Acids Res 20: 4981–4985. 10.1093/nar/20.19.4981

Bougueleret L, Schwarzstein M, Tsugita A, Zabeau M. 1984. Characterization of the genes coding for the Eco RV restriction and modification system of Escherichia coli. Nucleic Acids Res 12: 3659–3676. 10.1093/nar/12.8.3659

Bougueleret L, Tenchini ML, Botterman J, Zabeau M. 1985. Overproduction of the EcoR V endonuclease and methylase. Nucleic Acids Res 13: 3823–3839. 10.1093/nar/13.11.3823

Boyd AC, Charles IG, Keyte JW, Brammar WJ. 1986. Isolation and computer-aided characterization of MmeI, a type II restriction endonuclease from Methylophilus methylotrophus. Nucleic Acids Res 14: 5255–5274. 10.1093/nar/14.13.5255

Brennan CA, Van Cleve MD, Gumport RI. 1986. The effects of base analogue substitutions on the methylation by the EcoRI modification methylase of octadeoxyribonucleotides containing modified EcoRI recognition sequences. J Biol Chem 261: 7279–7286.

Brown NL, Smith M. 1980. A general method for defining restriction enzyme cleavage and recognition sites. Methods Enzymol 65: 391–404. 10.1016/S0076-6879(80)65050-2

Bullas LR, Colson C, Van Pel A. 1976. DNA restriction and modification systems in Salmonella. SQ, a new system derived by recombination between the SB system of Salmonella typhimurium and the SP system of Salmonella potsdam. J Gen Microbiol 95: 166–172. 10.1099/00221287-95-1-166

Bullas LR, Colson C, Neufeld B. 1980. Deoxyribonucleic acid restriction and modification systems in Salmonella: chromosomally located systems of different serotypes. J Bacteriol 141: 275–292.

Callahan SJ, Morgan RD, Jain R, Townson SA, Wilson GG, Roberts RJ, Aggarwal AK. 2011. Crystallization and preliminary crystallographic analysis of the type IIL restriction enzyme MmeI in complex with DNA. Acta Crystallogr, Sect F: Struct Biol Cryst Commun 67: 1262–1265. 10.1107/S1744309111028041

Cheng X, Balendiran K, Schildkraut I, Anderson JE. 1994. Structure of PvuII endonuclease with cognate DNA. EMBO J 13: 3927–3935. 10.1002/j.1460-2075.1994.tb06708.x

Chin V, Valinluck V, Magaki S, Ryu J. 2004. KpnBI is the prototype of a new family (IE) of bacterial type I restriction-modification system. Nucleic Acids Res 32: e138. 10.1093/nar/gnh134

Colson C, Colson AM, Van Pel A. 1970. Chromosomal location of host specificity in Salmonella typhimurium. J Gen Microbiol 60: 265–271. 10.1099/00221287-60-2-265

Connolly BA, Eckstein F, Pingoud A. 1984. The stereochemical course of the restriction endonuclease EcoRI-catalyzed reaction. J Biol Chem 259: 10760–10763.

Conrad M, Topal MD. 1989. DNA and spermidine provide a switch mechanism to regulate the activity of restriction enzyme Nae I. Proc Natl Acad Sci 86: 9707–9711. 10.1073/pnas.86.24.9707

Cowan GM, Gann AA, Murray NE. 1989. Conservation of complex DNA recognition domains between families of restriction enzymes. Cell 56: 103–109. 10.1016/0092-8674(89)90988-4

Daniel AS, Fuller-Pace FV, Legge DM, Murray NE. 1988. Distribution and diversity of hsd genes in Escherichia coli and other enteric bacteria. J Bacteriol 170: 1775–1782. 10.1128/jb.170.4.1775-1782.1988

Dartois V, De Backer O, Colson C. 1993. Sequence of the Salmonella typhimurium StyLT1 restriction-modification genes: homologies with EcoP1 and EcoP15 type-III R-M systems and presence of helicase domains. Gene 127: 105–110. 10.1016/0378-1119(93)90623-B

De Backer O, Colson C. 1991. Identification of the recognition sequence for the M.StyLTI methyltransferase of Salmonella typhimurium LT7: an asymmetric site typical of type-III enzymes. Gene 97: 103–107. 10.1016/0378-1119(91)90015-4

de la Campa AG, Kale P, Springhorn SS, Lacks SA. 1987. Proteins encoded by the DpnII restriction gene cassette. Two methylases and an endonuclease. J Mol Biol 196: 457–469. 10.1016/0022-2836(87)90024-6

Doerfler W. 1983. DNA methylation and gene activity. Ann Rev Biochem 52: 93–124. 10.1146/annurev.bi.52.070183.000521

Dunn JJ, Studier FW. 1983. Complete nucleotide sequence of bacteriophage T7 DNA and the locations of T7 genetic elements. J Mol Biol 166: 477–535. 10.1016/S0022-2836(83)80282-4

Ehbrecht HJ, Pingoud A, Urbanke C, Maass G, Gualerzi C. 1985. Linear diffusion of restriction endonucleases on DNA. J Biol Chem 260: 6160–6166.

Endlich B, Linn S. 1985. The DNA restriction endonuclease of Escherichia coli B. II. Further studies of the structure of DNA intermediates and products. J Biol Chem 260: 5729–5738.

Eskin B, Linn S. 1972. The deoxyribonucleic acid modification and restriction enzymes of Escherichia coli B. II. Purification, subunit structure, and catalytic properties of the restriction endonuclease. J Biol Chem 247: 6183–6191.

Fuller-Pace FV, Murray NE. 1986. Two DNA recognition domains of the specificity polypeptides of a family of type I restriction enzymes. Proc Natl Acad Sci 83: 9368–9372. 10.1073/pnas.83.24.9368

Fuller-Pace FV, Bullas LR, Delius H, Murray NE. 1984. Genetic recombination can generate altered restriction specificity. Proc Natl Acad Sci 81: 6095–6099. 10.1073/pnas.81.19.6095

Gabbara S, Bhagwat AS. 1992. Interaction of EcoRII endonuclease with DNA substrates containing single recognition sites. J Biol Chem 267: 18623–18630.

Gann AA, Campbell AJ, Collins JF, Coulson AF, Murray NE. 1987. Reassortment of DNA recognition domains and the evolution of new specificities. Mol Microbiol 1: 13–22. 10.1111/j.1365-2958.1987.tb00521.x

Gingeras TR, Milazzo JP, Roberts RJ. 1978. A computer assisted method for the determination of restriction enzyme recognition sites. Nucleic Acids Res 5: 4105–4127. 10.1093/nar/5.11.4105

Gough JA, Murray NE. 1983. Sequence diversity among related genes for recognition of specific targets in DNA molecules. J Mol Biol 166: 1–19. 10.1016/S0022-2836(83)80047-3

Grasby JA, Connolly BA. 1992. Stereochemical outcome of the hydrolysis reaction catalyzed by the EcoRV restriction endonuclease. Biochemistry 31: 7855–7861. 10.1021/bi00149a016

Gubler M, Bickle TA. 1991. Increased protein flexibility leads to promiscuous protein–DNA interactions in type IC restriction-modification systems. EMBO J 10: 951–957. 10.1002/j.1460-2075.1991.tb08029.x

Gubler M, Braguglia D, Meyer J, Piekarowicz A, Bickle TA. 1992. Recombination of constant and variable modules alters DNA sequence recognition by type IC restriction-modification enzymes. EMBO J 11: 233–240. 10.1002/j.1460-2075.1992.tb05046.x

Guo J, Gaj T, Barbas CF III, . 2010. Directed evolution of an enhanced and highly efficient FokI cleavage domain for zinc finger nucleases. J Mol Biol 400: 96–107. 10.1016/j.jmb.2010.04.060

Guschlbauer W. 1988. The DNA and S-adenosylmethionine-binding regions of EcoDam and related methyltransferases. Gene 74: 211–214. 10.1016/0378-1119(88)90289-2

Haberman A. 1974. The bacteriophage P1 restriction endonuclease. J Mol Biol 89: 545–563. 10.1016/0022-2836(74)90035-7

Hadi SM, Bachi B, Shepherd JC, Yuan R, Ineichen K, Bickle TA. 1979. DNA recognition and cleavage by the EcoP15 restriction endonuclease. J Mol Biol 134: 655–666. 10.1016/0022-2836(79)90372-3

Halford SE. 1983. How does EcoRI cleave its recognition site on DNA? Trends Biochem Sci 8: 455–460. 10.1016/0968-0004(83)90035-X

Halford SE, Goodall AJ. 1988. Modes of DNA cleavage by the EcoRV restriction endonuclease. Biochemistry 27: 1771–1777. 10.1021/bi00405a058

Halford SE, Johnson NP. 1980. The EcoRI restriction endonuclease with bacteriophage λ DNA. Equilibrium binding studies. Biochem J 191: 593–604. 10.1042/bj1910593

Halford SE, Johnson NP. 1983. Single turnovers of the EcoRI restriction endonuclease. Biochem J 211: 405–415. 10.1042/bj2110405

Halford SE, Marko JF. 2004. How do site-specific DNA-binding proteins find their targets? Nucleic Acids Res 32: 3040–3052. 10.1093/nar/gkh624

Halford SE, Taylor JD, Vermote CLM, Vipond IB. 1993. Mechanism of action of restriction endonuclease EcoRV. Nucleic Acids Mol Biol 7: 47–69. 10.1007/978-3-642-77950-3_4

Halford SE, Catto LE, Pernstich C, Rusling DA, Sanders KL. 2011. The reaction mechanism of FokI excludes the possibility of targeting zinc finger nucleases to unique DNA sites. Biochem Soc Trans 39: 584–588. 10.1042/BST0390584

Hattman S, Brooks JE, Masurekar M. 1978. Sequence specificity of the P1 modification methylase (M·Eco P1) and the DNA methylase (M·Eco dam) controlled by the Escherichia coli dam gene. J Mol Biol 126: 367–380. 10.1016/0022-2836(78)90046-3

Heitman J. 1992. How the EcoRI endonuclease recognizes and cleaves DNA. Bioessays 14: 445–454. 10.1002/bies.950140704

Heitman J. 1993. On the origins, structures and functions of restriction-modification enzymes. Genet Eng 15: 57–108. 10.1007/978-1-4899-1666-2_4

Heitman J, Model P. 1990a. Mutants of the EcoRI endonuclease with promiscuous substrate specificity implicate residues involved in substrate recognition. EMBO J 9: 3369–3378. 10.1002/j.1460-2075.1990.tb07538.x

Heitman J, Model P. 1990b. Substrate recognition by the EcoRI endonuclease. Proteins 7: 185–197. 10.1002/prot.340070207

Hensley P, Nardone G, Chirikjian JG, Wastney ME. 1990. The time-resolved kinetics of superhelical DNA cleavage by BamHI restriction endonuclease. J Biol Chem 265: 15300–15307.

Humbelin M, Suri B, Rao DN, Hornby DP, Eberle H, Pripfl T, Kenel S, Bickle TA. 1988. Type III DNA restriction and modification systems EcoP1 and EcoP15. Nucleotide sequence of the EcoP1 operon, the EcoP15 mod gene and some EcoP1 mod mutants. J Mol Biol 200: 23–29. 10.1016/0022-2836(88)90330-0

Iida S, Meyer J, Bachi B, Stalhammar-Carlemalm M, Schrickel S, Bickle TA, Arber W. 1983. DNA restriction–modification genes of phage P1 and plasmid p15B. Structure and in vitro transcription. J Mol Biol 165: 1–18. 10.1016/S0022-2836(83)80239-3

Jack WE, Terry BJ, Modrich P. 1982. Involvement of outside DNA sequences in the major kinetic path by which EcoRI endonuclease locates and leaves its recognition sequence. Proc Natl Acad Sci 79: 4010–4014. 10.1073/pnas.79.13.4010

Jack WE, Greenough L, Dorner LF, Xu SY, Strzelecka T, Aggarwal AK, Schildkraut I. 1991. Overexpression, purification and crystallization of BamHI endonuclease. Nucleic Acids Res 19: 1825–1829. 10.1093/nar/19.8.1825

Janscak P, MacWilliams MP, Sandmeier U, Nagaraja V, Bickle TA. 1999. DNA translocation blockage, a general mechanism of cleavage site selection by type I restriction enzymes. EMBO J 18: 2638–2647. 10.1093/emboj/18.9.2638

Jeltsch A, Alves J, Oelgeschläger T, Wolfes H, Maass G, Pingoud A. 1993. Mutational analysis of the function of Gln115 in the EcoRI restriction endonuclease, a critical amino acid for recognition of the inner thymidine residue in the sequence -GAATTC- and for coupling specific DNA binding to catalysis. J Mol Biol 229: 221–234. 10.1006/jmbi.1993.1019

Kannan P, Cowan GM, Daniel AS, Gann AA, Murray NE. 1989. Conservation of organization in the specificity polypeptides of two families of type I restriction enzymes. J Mol Biol 209: 335–344. 10.1016/0022-2836(89)90001-6

Kasarjian JK, Hidaka M, Horiuchi T, Iida M, Ryu J. 2004. The recognition and modification sites for the bacterial type I restriction systems KpnAI, StySEAI, StySENI and StySGI. Nucleic Acids Res 32: e82. 10.1093/nar/gnh079

Kauc L, Piekarowicz A. 1978. Purification and properties of a new restriction endonuclease from Haemophilus influenzae Rf. Eur J Biochem/FEBS 92: 417–426. 10.1111/j.1432-1033.1978.tb12762.x

Kelleher JE, Daniel AS, Murray NE. 1991. Mutations that confer de novo activity upon a maintenance methyltransferase. J Mol Biol 221: 431–440. 10.1016/0022-2836(91)80064-2

Kelly TJ Jr, Smith HO. 1970. A restriction enzyme from Hemophilus influenzae II. J Mol Biol 51: 393–409. 10.1016/0022-2836(70)90150-6

Kim YC, Grable JC, Love R, Greene PJ, Rosenberg JM. 1990. Refinement of Eco RI endonuclease crystal structure: a revised protein chain tracing. Science 249: 1307–1309. 10.1126/science.2399465

Kim YG, Smith J, Durgesha M, Chandrasegaran S. 1998. Chimeric restriction enzyme: Gal4 fusion to FokI cleavage domain. Biol Chem 379: 489–495. 10.1515/bchm.1998.379.4-5.489

King G, Murray NE. 1995. Restriction alleviation and modification enhancement by the Rac prophage of Escherichia coli K-12. Mol Microbiol 16: 769–777. 10.1111/j.1365-2958.1995.tb02438.x

King K, Benkovic SJ, Modrich P. 1989. Glu-111 is required for activation of the DNA cleavage center of EcoRI endonuclease. J Biol Chem 264: 11807–11815.

Kiss A, Posfai G, Keller CC, Venetianer P, Roberts RJ. 1985. Nucleotide sequence of the BsuRI restriction-modification system. Nucleic Acids Res 13: 6403–6421. 10.1093/nar/13.18.6403

Klimasauskas S, Timinskas A, Menkevicius S, Butkiene D, Butkus V, Janulaitis A. 1989. Sequence motifs characteristic of DNA[cytosine-N4]methyltransferases: similarity to adenine and cytosine-C5 DNA-methylases. Nucleic Acids Res 17: 9823–9832. 10.1093/nar/17.23.9823

Kong H, Morgan RD, Maunus RE, Schildkraut I. 1993. A unique restriction endonuclease, BcgI, from Bacillus coagulans. Nucleic Acids Res 21: 987–991. 10.1093/nar/21.4.987

Kosykh VG, Buryanov YI, Bayev AA. 1980. Molecular cloning of EcoRII endonuclease and methylase genes. Mol Gen Genet MGG 178: 717–718. 10.1007/BF00337884

Kosykh VG, Repik AV, Kaliman AV, Bur'ianov Ia I, Baev AA. 1989. [Primary structure of the gene of restriction endonuclease EcoRII]. Dokl Akad Nauk SSSR 308: 1497–1499.

Krüger DH, Bickle TA. 1983. Bacteriophage survival: multiple mechanisms for avoiding the deoxyribonucleic acid restriction systems of their hosts. Microbiol Rev 47: 345–360.

Krüger DH, Barcak GJ, Reuter M, Smith HO. 1988. EcoRII can be activated to cleave refractory DNA recognition sites. Nucleic Acids Res 16: 3997–4008. 10.1093/nar/16.9.3997

Kusiak M, Price C, Rice D, Hornby DP. 1992. The HsdS polypeptide of the type IC restriction enzyme EcoR124 is a sequence-specific DNA-binding protein. Mol Microbiol 6: 3251–3256. 10.1111/j.1365-2958.1992.tb01779.x

Lauster R, Kriebardis A, Guschlbauer W. 1987. The GATATC-modification enzyme EcoRV is closely related to the GATC-recognizing methyltransferases DpnII and dam from E. coli and phage T4. FEBS Lett 220: 167–176. 10.1016/0014-5793(87)80897-9

Lauster R, Trautner TA, Noyer-Weidner M. 1989. Cytosine-specific type II DNA methyltransferases. A conserved enzyme core with variable target-recognizing domains. J Mol Biol 206: 305–312. 10.1016/0022-2836(89)90480-4

Lehman IR. 1974. DNA ligase: structure, mechanism, and function. Science 186: 790–797. 10.1126/science.186.4166.790

Lesser DR, Kurpiewski MR, Jen-Jacobson L. 1990. The energetic basis of specificity in the Eco RI endonuclease–DNA interaction. Science 250: 776–786. 10.1126/science.2237428

Lesser DR, Grajkowski A, Kurpiewski MR, Koziolkiewicz M, Stec WJ, Jen-Jacobson L. 1992. Stereoselective interaction with chiral phosphorothioates at the central DNA kink of the EcoRI endonuclease-GAATTC complex. J Biol Chem 267: 24810–24818.

Lesser DR, Kurpiewski MR, Waters T, Connolly BA, Jen-Jacobson L. 1993. Facilitated distortion of the DNA site enhances EcoRI endonuclease-DNA recognition. Proc Natl Acad Sci 90: 7548–7552. 10.1073/pnas.90.16.7548

Li L, Chandrasegaran S. 1993. Alteration of the cleavage distance of Fok I restriction endonuclease by insertion mutagenesis. Proc Natl Acad Sci 90: 2764–2768. 10.1073/pnas.90.7.2764

Li L, Wu LP, Clarke R, Chandrasegaran S. 1993. C-terminal deletion mutants of the FokI restriction endonuclease. Gene 133: 79–84. 10.1016/0378-1119(93)90227-T

Li T, Huang S, Jiang WZ, Wright D, Spalding MH, Weeks DP, Yang B. 2011. TAL nucleases (TALNs): hybrid proteins composed of TAL effectors and FokI DNA-cleavage domain. Nucleic Acids Res 39: 359–372. 10.1093/nar/gkq704

Loenen WAM. 2003. Tracking EcoKI and DNA fifty years on: a golden story full of surprises. Nucleic Acids Res 31: 7059–7069. 10.1093/nar/gkg944

Loenen WA, Murray NE. 1986. Modification enhancement by the restriction alleviation protein (Ral) of bacteriophage λ. J Mol Biol 190: 11–22. 10.1016/0022-2836(86)90071-9

Loenen WA, Daniel AS, Braymer HD, Murray NE. 1987. Organization and sequence of the hsd genes of Escherichia coli K-12. J Mol Biol 198: 159–170. 10.1016/0022-2836(87)90303-2

Looney MC, Moran LS, Jack WE, Feehery GR, Benner JS, Slatko BE, Wilson GG. 1989. Nucleotide sequence of the FokI restriction-modification system: separate strand-specificity domains in the methyltransferase. Gene 80: 193–208. 10.1016/0378-1119(89)90284-9

Lu A-L, Jack WE, Modrich P. 1983. DNA determinants important in sequence recognition by EcoRI endonuclease. J Biol Chem 256: 13200–13206.

Maxwell A, Halford SE. 1982. The SalGI restriction endonuclease. Mechanism of DNA cleavage. Biochem J 203: 85–92. 10.1042/bj2030085

McClarin JA, Frederick CA, Wang BC, Greene P, Boyer HW, Grable J, Rosenberg JM. 1986. Structure of the DNA-Eco RI endonuclease recognition complex at 3 Å resolution. Science 234: 1526–1541. 10.1126/science.3024321

McLaughlin LW, Bensler F, Graeser E, Pile N, Scholtissek S. 1987. Effects of functional group changes in the EcoRI recognition site on the cleavage reaction catalysed by the endonuclease. Biochemistry 26: 7238–7245. 10.1021/bi00397a007

Meisel A, Krüger DH, Bickle TA. 1991. M.EcoP15 methylates the second adenine in its recognition sequence. Nucleic Acids Res 19: 3997. 10.1093/nar/19.14.3997

Meisel A, Bickle TA, Krüger DH, Schroeder C. 1992. Type III restriction enzymes need two inversely oriented recognition sites for DNA cleavage. Nature 355: 467–469. 10.1038/355467a0

Meister J, MacWilliams M, Hubner P, Jutte H, Skrzypek E, Piekarowicz A, Bickle TA. 1993. Macroevolution by transposition: drastic modification of DNA recognition by a type I restriction enzyme following Tn5 transposition. EMBO J 12: 4585–4591. 10.1002/j.1460-2075.1993.tb06147.x

Morgan RD, Bhatia TK, Lovasco L, Davis TB. 2008. MmeI: a minimal Type II restriction-modification system that only modifies one DNA strand for host protection. Nucleic Acids Res 36: 6558–6570. 10.1093/nar/gkn711

Morgan RD, Dwinell EA, Bhatia TK, Lang EM, Luyten YA. 2009. The MmeI family: type II restriction-modification enzymes that employ single-strand modification for host protection. Nucleic Acids Res 37: 5208–5221. 10.1093/nar/gkp534

Moses PB, Horiuchi K. 1979. Specific recombination in vitro promoted by the restriction endonuclease HgaI. J Mol Biol 135: 517–524. 10.1016/0022-2836(79)90452-2

Murray NE. 2002. 2001 Fred Griffith review lecture. Immigration control of DNA in bacteria: self versus non-self. Microbiology 148: 3–20. 10.1099/00221287-148-1-3

Murray NE, Batten PL, Murray K. 1973. Restriction of bacteriophage λ by Escherichia coli K. J Mol Biol 81: 395–407. 10.1016/0022-2836(73)90149-6

Murray NE, Gough JA, Suri B, Bickle TA. 1982. Structural homologies among type I restriction-modification systems. EMBO J 1: 535–539. 10.1002/j.1460-2075.1982.tb01205.x

Nagaraja V, Shepherd JC, Bickle TA. 1985a. A hybrid recognition sequence in a recombinant restriction enzyme and the evolution of DNA sequence specificity. Nature 316: 371–372. 10.1038/316371a0

Nagaraja V, Shepherd JC, Pripfl T, Bickle TA. 1985b. Two type I restriction enzymes from Salmonella species. Purification and DNA recognition sequences. J Mol Biol 182: 579–587. 10.1016/0022-2836(85)90243-8

Nardone G, George J, Chirikjian JG. 1986. Differences in the kinetic properties of BamHI endonuclease and methylase with linear DNA substrates. J Biol Chem 261: 12128–12133.

Narva KE, Van Etten JL, Slatko BE, Benner JS. 1988. The amino acid sequence of the eukaryotic DNA [N6-adenine]methyltransferase, M·CviBIII, has regions of similarity with the prokaryotic isoschizomer M·TaqI and other DNA [N6-adenine] methyltransferases. Gene 74: 253–259. 10.1016/0378-1119(88)90298-3

Nelson M, Raschke E, McClelland M. 1993. Effect of site-specific methylation on restriction endonucleases and DNA modification methyltransferases. Nucleic Acids Res 21: 3139–3154. 10.1093/nar/21.13.3139

Newman PC, Williams DM, Cosstick R, Seela F, Connolly BA. 1990. Interaction of the EcoRV restriction endonuclease with the deoxyadenosine and thymidine bases in its recognition hexamer d(GATATC). Biochemistry 29: 9902–9910. 10.1021/bi00494a021

Newman M, Strzelecka T, Dorner LF, Schildkraut I, Aggarwal AK. 1994. Structure of restriction endonuclease bamhi phased at 1.95 A resolution by MAD analysis. Structure 2: 439–452. 10.1016/S0969-2126(00)00045-9

Oller AR, Vanden Broek W, Conrad M, Topal MD. 1991. Ability of DNA and spermidine to affect the activity of restriction endonucleases from several bacterial species. Biochemistry 30: 2543–2549. 10.1021/bi00223a035

Olsen DB, Kotzorek G, Eckstein F. 1990. Investigation of the inhibitory role of phosphorothioate internucleotidic linkages on the catalytic activity of the restriction endonuclease EcoRV. Biochemistry 29: 9546–9551. 10.1021/bi00493a008

Orekhov AV, Rebentish BA, Debabov VG. 1982. [New site-specific endonuclease from Streptomyces: SGRII]. Dokl Akad Nauk SSSR 263: 217–220.

Pein CD, Reuter M, Meisel A, Cech D, Krüger DH. 1991. Activation of restriction endonuclease EcoRII does not depend on the cleavage of stimulator DNA. Nucleic Acids Res 19: 5139–5142. 10.1093/nar/19.19.5139

Piekarowicz A, Kalinowska J. 1974. Host specificity of DNA in Haemophilus influenzae: similarity between host-specificity types of Haemophilus influenzae Re and Rf. J Gen Microbiol 81: 405–411. 10.1099/00221287-81-2-405

Piekarowicz A, Bickle TA, Shepherd JC, Ineichen K. 1981. The DNA sequence recognised by the HinfIII restriction endonuclease. J Mol Biol 146: 167–172. 10.1016/0022-2836(81)90372-7

Piekarowicz A, Goguen JD, Skrzypek E. 1985. The EcoDXX1 restriction and modification system of Escherichia coli ET7. Purification, subunit structure and properties of the restriction endonuclease. Eur J Biochem/FEBS 152: 387–393. 10.1111/j.1432-1033.1985.tb09209.x

Pingoud A, Jeltsch A. 2001. Structure and function of type II restriction endonucleases. Nucleic Acids Res 29: 3705–3727. 10.1093/nar/29.18.3705

Pingoud A, Fuxreiter M, Pingoud V, Wende W. 2005. Type II restriction endonucleases: structure and mechanism. Cell Mol Life Sci 62: 685–707. 10.1007/s00018-004-4513-1

Pingoud A, Wilson GG, Wende W. 2014. Type II restriction endonucleases—a historical perspective and more. Nucleic Acids Res 42: 7489–7527. 10.1093/nar/gku447

Polisky B, Greene P, Garfin DE, McCarthy BJ, Goodman HM, Boyer HW. 1975. Specificity of substrate recognition by the EcoRI restriction endonuclease. Proc Natl Acad Sci 72: 3310–3314. 10.1073/pnas.72.9.3310

Pósfai J, Bhagwat AS, Posfai G, Roberts RJ. 1989. Predictive motifs derived from cytosine methyltransferases. Nucleic Acids Res 17: 2421–2435. 10.1093/nar/17.7.2421

Potter BV, Eckstein F. 1984. Cleavage of phosphorothioate-substituted DNA by restriction endonucleases. J Biol Chem 259: 14243–14248.

Price C, Bickle TA. 1986. A possible role for DNA restriction in bacterial evolution. Microbiol Sci 3: 296–299.

Price C, Pripfl T, Bickle TA. 1987a. EcoR124 and EcoR124/3: the first members of a new family of type I restriction and modification systems. Eur J Biochem/FEBS 167: 111–115. 10.1111/j.1432-1033.1987.tb13310.x

Price C, Shepherd JC, Bickle TA. 1987b. DNA recognition by a new family of type I restriction enzymes: a unique relationship between two different DNA specificities. EMBO J 6: 1493–1497. 10.1002/j.1460-2075.1987.tb02391.x

Price C, Lingner J, Bickle TA, Firman K, Glover SW. 1989. Basis for changes in DNA recognition by the EcoR124 and EcoR124/3 type I DNA restriction and modification enzymes. J Mol Biol 205: 115–125. 10.1016/0022-2836(89)90369-0

Rao DN, Eberle H, Bickle TA. 1989a. Characterization of mutations of the bacteriophage P1 mod gene encoding the recognition subunit of the EcoP1 restriction and modification system. J Bacteriol 171: 2347–2352. 10.1128/jb.171.5.2347-2352.1989

Rao DN, Page MG, Bickle TA. 1989b. Cloning, over-expression and the catalytic properties of the EcoP15 modification methylase from Escherichia coli. J Mol Biol 209: 599–606. 10.1016/0022-2836(89)90597-4

Reiser J, Yuan R. 1977. Purification and properties of the P15 specific restriction endonuclease from Escherichia coli. J Biol Chem 252: 451–456.

Reuter M, Schneider-Mergener J, Kupper D, Meisel A, Mackeldanz P, Krüger DH, Schroeder C. 1999. Regions of endonuclease EcoRII involved in DNA target recognition identified by membrane-bound peptide repertoires. J Biol Chem 274: 5213–5221. 10.1074/jbc.274.8.5213

Risser R, Hopkins N, Davis RW, Delius H, Mulder C. 1974. Action of Escherichia coli P1 restriction endonuclease on simian virus 40 DNA. J Mol Biol 89: 517–544. 10.1016/0022-2836(74)90034-5

Roberts RJ, Halford S.E. 1993. Type II restriction enzymes. In Nucleases (2nd ed.) (ed. Linn SM, Lloyd RS, Roberts RJ), pp. 35–88. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Roberts RJ, Macelis D. 1993a. REBASE—restriction enzymes and methylases. Nucleic Acids Res 21: 3125–3137. 10.1093/nar/21.13.3125

Roberts RJ, Macelis D. 1993b. Appendix A. The restriction enzymes. In Nucleases 2nd ed. (ed. Linn S, Lloyd RS, Roberts RJ), pp. 439–444.

Roberts RJ, Belfort M, Bestor T, Bhagwat AS, Bickle TA, Bitinaite J, Blumenthal RM, Degtyarev S, Dryden DT, Dybvig K, et al. 2003. A nomenclature for restriction enzymes, DNA methyltransferases, homing endonucleases and their genes. Nucleic Acids Res 31: 1805–1812. 10.1093/nar/gkg274

Rosamond J, Endlich B, Linn S. 1979. Electron microscopic studies of the mechanism of action of the restriction endonuclease of Escherichia coli B. J Mol Biol 129: 619–635. 10.1016/0022-2836(79)90472-8

Rosenberg JM. 1991. Structure and function of restriction endonucleases. Curr Opin Struct Biol 1: 104–113. 10.1016/0959-440X(91)90018-O

Rosner JL. 1973. Modification-deficient mutants of bacteriophage P1. I. Restriction by P1 cryptic lysogens. Virology 52: 213–222. 10.1016/0042-6822(73)90410-8

Sain B, Murray NE. 1980. The hsd (host specificity) genes of E. coli K 12. Mol Gen Genet 180: 35–46. 10.1007/BF00267350

Sanders KL, Catto LE, Bellamy SR, Halford SE. 2009. Targeting individual subunits of the FokI restriction endonuclease to specific DNA strands. Nucleic Acids Res 37: 2105–2115. 10.1093/nar/gkp046

Schroeder C, Jurkschat H, Meisel A, Reich JG, Krüger D. 1986. Unusual occurrence of EcoP1 and EcoP15 recognition sites and counterselection of type II methylation and restriction sequences in bacteriophage T7 DNA. Gene 45: 77–86. 10.1016/0378-1119(86)90134-4

Seeman NC, Rosenberg JM, Rich A. 1976. Sequence-specific recognition of double helical nucleic acids by proteins. Proc Natl Acad Sci 73: 804–808. 10.1073/pnas.73.3.804

Selent U, Ruter T, Kohler E, Liedtke M, Thielking V, Alves J, Oelgeschläger T, Wolfes H, Peters F, Pingoud A. 1992. A site-directed mutagenesis study to identify amino acid residues involved in the catalytic function of the restriction endonuclease EcoRV. Biochemistry 31: 4808–4815. 10.1021/bi00135a010

Sharp PM, Kelleher JE, Daniel AS, Cowan GM, Murray NE. 1992. Roles of selection and recombination in the evolution of type I restriction-modification systems in enterobacteria. Proc Natl Acad Sci 89: 9836–9840. 10.1073/pnas.89.20.9836

Sharrocks AD, Hornby DP. 1991. Transcriptional analysis of the restriction and modification genes of bacteriophage P1. Mol Microbiol 5: 685–694. 10.1111/j.1365-2958.1991.tb00740.x

Skrzypek E, Piekarowicz A. 1989. The EcoDXX1 restriction and modification system: cloning the genes and homology to type I restriction and modification systems. Plasmid 21: 195–204. 10.1016/0147-619X(89)90043-7

Smith HO, Annau TM, Chandrasegaran S. 1990. Finding sequence motifs in groups of functionally related proteins. Proc Natl Acad Sci 87: 826–830. 10.1073/pnas.87.2.826

Smith DW, Crowder SW, Reich NO. 1992. In vivo specificity of EcoRI DNA methyltransferase. Nucleic Acids Res 20: 6091–6096. 10.1093/nar/20.22.6091

Som S, Bhagwat AS, Friedman S. 1987. Nucleotide sequence and expression of the gene encoding the EcoRII modification enzyme. Nucleic Acids Res 15: 313–332. 10.1093/nar/15.1.313

Stephenson FH, Ballard BT, Boyer HW, Rosenberg JM, Greene PJ. 1989. Comparison of the nucleotide and amino acid sequences of the RsrI and EcoRI restriction endonucleases. Gene 85: 1–13. 10.1016/0378-1119(89)90458-7

Strzelecka T, Dorner L, Schildkraut I, Agarwal A. 1990. Structural studies of the BamHI restriction enzyme. Biophys J 57: 68.

Strzelecka T, Newman M, Dorner LF, Knott R, Schildkraut I, Aggarwal AK. 1994. Crystallization and preliminary X-ray analysis of restriction endonuclease BamHI-DNA complex. J Mol Biol 239: 430–432. 10.1006/jmbi.1994.1383

Studier FW, Bandyopadhyay PK. 1988. Model for how type I restriction enzymes select cleavage sites in DNA. Proc Natl Acad Sci 85: 4677–4681. 10.1073/pnas.85.13.4677

Sugisaki H, Kanazawa S. 1981. New restriction endonucleases from Flavobacterium okeanokoites (FokI) and Micrococcus luteus (MluI). Gene 16: 73–78. 10.1016/0378-1119(81)90062-7

Sugisaki H, Kita K, Takanami M. 1989. The FokI restriction-modification system. II. Presence of two domains in FokI methylase responsible for modification of different DNA strands. J Biol Chem 264: 5757–5761.

Suri B, Bickle TA. 1985. EcoA: the first member of a new family of type I restriction modification systems. Gene organization and enzymatic activities. J Mol Biol 186: 77–85. 10.1016/0022-2836(85)90258-X

Suri B, Nagaraja V, Bickle TA. 1984a. Bacterial DNA modification. Curr Top Microbiol Immunol 108: 1–9.

Suri B, Shepherd JC, Bickle TA. 1984b. The EcoA restriction and modification system of Escherichia coli 15T-: enzyme structure and DNA recognition sequence. EMBO J 3: 575–579. 10.1002/j.1460-2075.1984.tb01850.x

Szybalski W, Kim SC, Hasan N, Podhajska AJ. 1991. Class-IIS restriction enzymes—a review. Gene 100: 13–26. 10.1016/0378-1119(91)90345-C

Tao T, Blumenthal RM. 1992. Sequence and characterization of pvuIIR, the PvuII endonuclease gene, and of pvuIIC, its regulatory gene. J Bacteriol 174: 3395–3398. 10.1128/jb.174.10.3395-3398.1992

Taylor JD, Halford SE. 1989. Discrimination between DNA sequences by the EcoRV restriction endonuclease. Biochemistry 28: 6198–6207. 10.1021/bi00441a011

Taylor JD, Halford SE. 1992. The activity of the EcoRV restriction endonuclease is influenced by flanking DNA sequences both inside and outside the DNA–protein complex. Biochemistry 31: 90–97. 10.1021/bi00116a014

Taylor JD, Goodall AJ, Vermote CL, Halford SE. 1990. Fidelity of DNA recognition by the EcoRV restriction/modification system in vivo. Biochemistry 29: 10727–10733. 10.1021/bi00500a003

Taylor JD, Badcoe IG, Clarke AR, Halford SE. 1991. EcoRV restriction endonuclease binds all DNA sequences with equal affinity. Biochemistry 30: 8743–8753. 10.1021/bi00100a005

Taylor I, Patel J, Firman K, Kneale G. 1992. Purification and biochemical characterisation of the EcoR124 type I modification methylase. Nucleic Acids Res 20: 179–186. 10.1093/nar/20.2.179

Terry BJ, Jack WE, Rubin RA, Modrich P. 1983. Thermodynamic parameters governing interaction of EcoRI endonuclease with specific and nonspecific DNA sequences. J Biol Chem 258: 9820–9825.

Terry BJ, Jack WE, Modrich P. 1985. Facilitated diffusion during catalysis by EcoRI endonuclease. Nonspecific interactions in EcoRI catalysis. J Biol Chem 260: 13130–13137.

Terry BJ, Jack WE, Modrich P. 1987. Mechanism of specific site location and DNA cleavage by EcoR I endonuclease. Gene Amplif Anal 5: 103–118.

Thielking V, Alves J, Fliess A, Maass G, Pingoud A. 1990. Accuracy of the EcoRI restriction endonuclease: binding and cleavage studies with oligodeoxynucleotide substrates containing degenerate recognition sequences. Biochemistry 29: 4682–4691. 10.1021/bi00471a024

Thielking V, Selent U, Kohler E, Wolfes H, Pieper U, Geiger R, Urbanke C, Winkler FK, Pingoud A. 1991. Site-directed mutagenesis studies with EcoRV restriction endonuclease to identify regions involved in recognition and catalysis. Biochemistry 30: 6416–6422. 10.1021/bi00240a011

Titheradge AJ, King J, Ryu J, Murray NE. 2001. Families of restriction enzymes: an analysis prompted by molecular and genetic data for type ID restriction and modification systems. Nucleic Acids Res 29: 4195–4205. 10.1093/nar/29.20.4195

Tolstoshev CM, Blakesley RW. 1982. RSITE: a computer program to predict the recognition sequence of a restriction enzyme. Nucleic Acids Res 10: 1–17. 10.1093/nar/10.1.1

Topal MD, Thresher RJ, Conrad M, Griffith J. 1991. NaeI endonuclease binding to pBR322 DNA induces looping. Biochemistry 30: 2006–2010. 10.1021/bi00221a038

Ueno T, Ito H, Kimizuka F, Kotani H, Nakajima K. 1993. Gene structure and expression of the MboI restriction–modification system. Nucleic Acids Res 21: 2309–2313. 10.1093/nar/21.10.2309

Van Cott EM, Wilson GG. 1988. Cloning the FnuDI, NaeI, NcoI and XbaI restriction-modification systems. Gene 74: 55–59. 10.1016/0378-1119(88)90251-X

Vermote CL, Halford SE. 1992. EcoRV restriction endonuclease: communication between catalytic metal ions and DNA recognition. Biochemistry 31: 6082–6089. 10.1021/bi00141a018

Vipond IB, Halford SE. 1993. Structure-function correlation for the EcoRV restriction enzyme: from non-specific binding to specific DNA cleavage. Mol Microbiol 9: 225–231. 10.1111/j.1365-2958.1993.tb01685.x

Vipond IB, Halford SE. 1995. Specific DNA recognition by EcoRV restriction endonuclease induced by calcium ions. Biochemistry 34: 1113–1119. 10.1021/bi00004a002

von Hippel PH, Berg OG. 1989. Facilitated target location in biological systems. J Biol Chem 264: 675–678.

Waters TR, Connolly BA. 1992. Continuous spectrophotometric assay for restriction endonucleases using synthetic oligodeoxynucleotides and based on the hyperchromic effect. Anal Biochem 204: 204–209. 10.1016/0003-2697(92)90162-Z

Waters TR, Connolly BA. 1994. Interaction of the restriction endonuclease EcoRV with the deoxyguanosine and deoxycytidine bases in its recognition sequence. Biochemistry 33: 1812–1819. 10.1021/bi00173a026

Wilson GG. 1991. Organization of restriction-modification systems. Nucleic Acids Res 19: 2539–2566. 10.1093/nar/19.10.2539

Wilson GG, Murray NE. 1991. Restriction and modification systems. Ann Rev Genet 25: 585–627. 10.1146/annurev.ge.25.120191.003101

Winkler FK. 1992. Structure and function of restriction endonucleases. Curr Opin Struct Biol 2: 93–99. 10.1016/0959-440X(92)90183-8

Winkler FK, Banner DW, Oefner C, Tsernoglou D, Brown RS, Heathman SP, Bryan RK, Martin PD, Petratos K, Wilson KS. 1993. The crystal structure of EcoRV endonuclease and of its complexes with cognate and non-cognate DNA fragments. EMBO J 12: 1781–1795. 10.1002/j.1460-2075.1993.tb05826.x

Winkler FK, Prota AE. 2004. Structure and function of EcoRV endonuclease. In Restriction endonucleases (ed. Pingoud A), pp. 179–214. Springer, Berlin.

Winter M. 1997. “Investigation of de novo methylation activity in mutants of the EcoKI methyltransferase.” PhD thesis, University of Edinburgh, UK.

Xu SY, Schildkraut I. 1991. Isolation of BamHI variants with reduced cleavage activities. J Biol Chem 266: 4425–4429.

Yang CC, Topal MD. 1992. Nonidentical DNA-binding sites of endonuclease NaeI recognize different families of sequences flanking the recognition site. Biochemistry 31: 9657–9664. 10.1021/bi00155a019

Yanofsky SD, Love R, McClarin JA, Rosenberg JM, Boyer HW, Greene PJ. 1987. Clustering of null mutations in the EcoRI endonuclease. Proteins 2: 273–282. 10.1002/prot.340020403

Yuan R, Reiser J. 1978. Steps in the reaction mechanism of the Escherichia coli plasmid P15-specific restriction endonuclease. J Mol Biol 122: 433–445. 10.1016/0022-2836(78)90420-5

Yuan R, Bickle TA, Ebbers W, Brack C. 1975. Multiple steps in DNA recognition by restriction endonuclease from E. coli K. Nature 256: 556–560. 10.1038/256556a0

Yuan R, Hamilton DL, Burckhardt J. 1980a. DNA translocation by the restriction enzyme from E. coli K. Cell 20: 237–244. 10.1016/0092-8674(80)90251-2

Yuan R, Hamilton DL, Hadi SM, Bickle TA. 1980b. Role of ATP in the cleavage mechanism of the EcoP15 restriction endonuclease. J Mol Biol 144: 501–519. 10.1016/0022-2836(80)90334-4

Zabeau M, Friedman S, Van Montagu M, Schell J. 1980. The ral gene of phage λ. I. Identification of a non-essential gene that modulates restriction and modification in E. coli. Mol Gen Genet MGG 179: 63–73. 10.1007/BF00268447

Zebala JA, Choi J, Barany F. 1992a. Characterization of steady state, single-turnover, and binding kinetics of the TaqI restriction endonuclease. J Biol Chem 267: 8097–8105.

Zebala JA, Choi J, Trainor GL, Barany F. 1992b. DNA recognition of base analogue and chemically modified substrates by the TaqI restriction endonuclease. J Biol Chem 267: 8106–8116.


http://library.cshl.edu/Meetings/restriction-enzymes/v-Aggarwal.php Aggarwal A. 2013.

http://library.cshl.edu/Meetings/restriction-enzymes/Halford.php Halford SE. 2013.

http://library.cshl.edu/Meetings/restriction-enzymes/v-Morgan.php Morgan RD. 2013.

http://library.cshl.edu/Meetings/restriction-enzymes/v-Roberts.php Roberts RJ. 2013.

http://library.cshl.edu/Meetings/restriction-enzymes/v-Rosenberg.php Rosenberg JM. 2013.

http://library.cshl.edu/Meetings/restriction-enzymes/v-Studier.php Studier FW. 2013.

1 Adapted from Roberts (1993a).

Get the full book:
PDF / ePub / mobi

Buy the Print Book