Overview and history
Experiments of O. Avery (1942) and A. Hershey (1952) demonstrated that DNA is the principle genetic material. DNA became an important subject of intensive research after its structure was determined by F. Crick and J. Watson (1953). Discovery of the restriction endonucleases allows researchers to cleave DNA at specific sites, a great advantage over physical or chemical cleavage that results in random fragmentation of DNA. Restriction endonucleases have become a new tool for DNA manipulation, laying a foundation for modern biology that led to the development of genetic engineering, molecular biology, and genomics.
Analogous to the immune response in mammals, bacteria such as E. coli possess a defense system against infection of bacteriophages. S. Luria (1952) and G. Bertani (1953) discovered that efficiency of plating (EOP) of a bacteriophage could be markedly different, depending on the E. coli strains used. For example, EOP was close to 1 when bacteriophages prepared from type-B E. coli strain were infected to the same type-B strain. When they were used to infect type-K E. coli strain, however, EOP was greatly reduced to approximately 10-4. On the other hand, EOP was again close to 1 when bacteriophages from type-K E. coli infected the same type-K strain. However, infection of type-B strain with bacteriophages from type-K E. coli reduced EOP to approximately 10-4. This phenomenon was referred to as restriction, because multiplication of bacteriophages obtained from one strain of E. coli was restricted in the other E. coli strain.
The bacteriophages originated from type-B strain could still infect type-K strain. With a low EOP (10-4), this was due to the fact that some bacteriophages from type-B avoided restriction and could infect the type-K strain. The genetic materials of the surviving ones were modified so that they were no longer restricted by the new host strain being infected. Thus, bacteriophages produced from type-K strain easily infect type-K strain, while their infection of type-B strain is much more difficult due to restriction/modification - a defense system of bacteria against bacteriophages.
Discovery, purification, and application
The early history of restriction endonucleases was written several decades ago primarily by three scientists Werner Arber, Hamilton Smith, and Daniel Nathans who earned a Nobel Prize in Physiology or Medicine in 1978.
W. Arber and D. Dussoix (1962) proposed a model that restriction/modification phenomena depended on two types of enzyme activities. One is a methylase that transfers a methyl group to DNA, and the other is a restriction nuclease that cleaves unmethylated DNA. "Suppression of bacteriophage growth" occurs via the cleavage of unmethylated DNA (foreign bacteriophage DNA) by restriction enzymes present in bacteria. In contrast, methylation of DNA renders its own DNA resistant to cleavage by restriction nucleases. However, foreign bacteriophage DNA can get methylated with a low frequency before restriction enzymes act, leading to low EOP (10-4) in a new host cell (modification phenomenon). Complete modifications of phage DNA through several generations of cultivation in the new host strain lead to loss of its ability to survive in the original host strain.
W. Arber and S. Linn (1968) demonstrated that restriction endonucleases from E. coli possessed the ability to degrade DNA, and M. Meselson and R. Yuan (1968) purified a type l restriction endonucleases for the first time. However, this enzyme cleaved DNA non-specifically to the disappointment of many biologists. In 1970, H. Smith and his colleagues finally purified a type ll restriction endonucleases (Hind II) from Haemophilus which cleaves at a specific recognition sequence. D. Nathan and K. Danna used restriction endonucleases to cleave SV40 DNA undergoing replication and analyzed its cleavage products, expanding the applications of restriction endonucleases. This accomplishment provided P. Berg with a revolutionary idea to create recombinant DNA for the first time in 1972. Following the advent of recombinant DNA technology, discovery of many more restriction endonucleases and their use have led to rapid growth of molecular biology and biotechnology.
So far, approximately 3,000 restriction endonucleases have been found that recognize 230 different recognition sequences. Restriction endonucleases are found mostly in bacteria. However, their presence has been confirmed in viruses, archebacteria, and even in eukaryotes. Approximately 25% of bacteria contain at least one restriction endonuclease and some bacterial strains contain as many as seven different restriction endonucleases.
Typically restriction endonucleases are found together with one or two modification enzymes (methylase or DNA-methyltransferase). Modification enzymes, like restriction endonucleases, recognize the same sequences. However, they protect DNA from cleavage by adding a methyl group to one base in the recognition sequence. Hence the modified sequences go unrecognized by restriction endonucleases. A methyl group sticks out to the major groove in the double helix, interrupting the access of the restriction endonucleases. After DNA replication, DNA exits in a hemimethylated state, in which only one strand is methylated. Hemimethylation also disables action of restriction endonucleases. This partnership of restriction and modification enzymes is referred to as the "restriction/modification system." In some bacteria, the two enzymes exist separately, whereas in the other bacteria they are present together as a complex enzyme or as distinct domains of a single polypeptide.
Restriction endonucleases have traditionally been categorized into three groups. This classification depends on the configurations of the constituent subunits, cleavage position, cleavage sequence specificity, and requirements of cofactor. However, it is difficult to group all restriction endonucleases into only three types due to the high degree of divergence in amino acid sequences in restriction endonucleases
Restriction and methyl transferase enzymes exist together in one complex. The recognition sequence and cleavage site are different. In other words, they recognize a specific sequence, but cleave DNA non-specifically approximately 1,000 base pairs away from the recognition sequence. In addition, ATP, AdoMet (S-adenosyl methionine, SAM), and Mg2+ are required for enzyme activity.
Although these enzymes provided a good model for studies of restriction phenomenon observed in
bacteria, they are not suitable for molecular biology experiments.
Examples : EcoK l, EcoB l, CfrA l, StyLT ll
Type ll and Type ll subclasses
Cleavage occurs within or near the recognition sequence. Thus, this type of enzyme is well suited to DNA analysis or cloning in laboratories. Type ll restriction endonucleases include enzymes of vast diversity and thus are divided into many subgroups. For example, amino acid sequences of Type ll restriction endonucleases vary greatly, indicating that convergent evolution occurred from many distinct ancestral proteins.
Orthodox type ll
This is the most common type of restriction endonucleases, which cleaves DNA within the recognition sequence. This type of restriction endonucleases exists separately from its cognate modifying enzyme. Most of the restriction endonucleases commercially available belong to this category. Since these enzymes bind DNA as homodimers, their recognition sites contain palindromic sequences of 4-8 base pairs, and they cleave both strands of DNA. As a result, a 5' overhang and a 3' overhang or a blunt ended structure can be produced. Cleavage requires Mg2+, and the sizes of these enzymes are small, typically with 200-350 amino acids.
Examples : EcoR l (Cat.# R002), BamH l (Cat.# R003), Hind lll (Cat.# R008), Kpn l (Cat.# R014),
Not l (Cat.# R001), Pst l (Cat.# R019), Sma l (Cat.# R015), Xho l (Cat.# R007)
These enzymes exist separately from modification enzymes like the orthodox type II, enzymes,
but they cleave DNA outside the recognition sequence. The recognition sequences are
continuous and asymmetric. Even though they consist of a single polypeptide, their DNA
binding and cleavage activities are present in a distinct domain. DNA binding of this type enzyme
on its recognition sequence requires a single enzyme molecule, but cleavage of DNA requires
interaction with the other enzyme molecule. For this reason, these enzymes and more active on
DNA molecules that contain multiple recognition sites. Cleavage activity requires Mg2+, and the
sizes of these enzymes are intermediate with about 400-650 amino acids.
Examples : Fok l (Cat.# R088), Alw26 (Cat.# R066) l, Bbv l, Bsr l, Ear l, Hph l (Cat.# R053), Mbo l (Cat.# R050), SfaN l, Tth111 l (Cat.# R038)
These enzymes react with two recognition sequences. One is cut by the enzyme, while the
other bound to the enzyme functions as an allosteric effector. Marked site preference can be
observed in this type of restriction endonucleases.
Examples : Nae l (Cat.# R035)
These enzymes react with two recognition sequences like the Type llE enzymes, but they cleave
DNA at both sites.
Examples : NgoM lV (Cat.# R095)
These enzymes such as Bpu10 l and Bsl l exist as a complex enzyme containing two different
subunits with restriction and modification activities respectively. Bpu10 l is a heterodimeric
enzyme, in which the two subunits form a single active site that recognizes asymmetric
sequences. Bsl l is a heterotetramer which recognizes palindromic sequences.
This type of restriction endonucleases requires AdoMet for activity like Type ll B (see below;
Type llB). Both restriction and modification activities reside in a single polypeptide.
Examples: Eco57 l
This type of enzyme recognizes and cleaves methylated DNA.
Examples: Dpn l (Cat.# R054)
This subclass can also be categorized into Type lV. These enzymes contain three distinct
domains for restriction, modification, and recognition activity in a single polypeptide, and they are
active in the form of heterodimer or heterotrimer. The recognition sequences can be symmetric
(Bpl l) or asymmetric. They require Mg2+ and AdoMet for restriction and modification activities
and have a larger structure with 850-1,250 amino acids. The recognition sequences can be
continuous or discontinuous (two halves of the recognition sequence are separated). They cleave
10-12 base pairs away from the recognition sequence and produce a 3’ overhang.
Examples: Bcg l, Bpl l, Bsp24 l, Bae l, Cje l
Similarly to the Type l restriction endonucleases, restriction and modification enzymes exist in one
complex. ATP is necessary for the enzymes to move along DNA. Cleavage occurs approximately 25
base pairs away from the recognition sequence. Complete digestion is almost never obtained. This
is because cleavage of one recognition sequence needs one additional intact recognition sequence
on the same DNA running in the opposite direction. Thus, this type of enzymes has no laboratory
value and thus is not availablecommercially.
Examples : EcoP l, Hinf lll, StyLT l
These are DNA cleaving enzymes which contain an intron or intein. Intron is a region present in
precursor RNAs that is removed by a process called splicing, and intein is a region in precursor
protein that is removed during protein maturation. The prefix " l -" or "P l -" is used to indicate the
presence of intron or intein, respectively, in the enzyme. These enzymes bind relatively longer
sequences (12-40 base pairs) and cleave inside the recognition sequences. The recognition
sequences are rare to find even within a very large genome because they are long.
This enzyme can catalyze cleavage of the sequences with some base replacements.
Examples : l -Ceul , PI-Psp l
Nicking endonucleases are genetically modified enzymes. Natural restriction endonucleases are
genetically engineered to cleave only one strand of DNA double helix. This can be achieved by
disabling dimerization, or suppressing the cleavage activity of one of the two monomers. It can
be used to convert supercoiled DNA into open circular DNA.
Examples : N-Alw l , N-BstNB l
The name of restriction endonucleases is determined by the genus, species, and strain of a
bacterium in which a restriction endonuclease is found, and finally the order of discovery. For
example, EcoR l was discovered from Escherichia coli RY13; "E" (capital) from Escherichia (genus),
“co” (lower case, first two letters) from coli (species), “R” (capital) for RY13 (strain), “ l ” (Roman
numeral) for the enzyme which was discovered first.
Table 1. Classification of type II restriction endonucleases