Restriction Enzymes
Mala Mani, Dana Farber Cancer Institute, Boston, Massachusetts, USA
Karthikeyan Kandavelou, Johns Hopkins University, Baltimore, Maryland, USA
Srinivasan Chandrasegaran, Johns Hopkins University, Baltimore, Maryland, USA
Type II restriction enzymes are the molecular scissors that catalyse the double-strand
cleavage of deoxyribonucleic acid (DNA) at specific base sequences. They are essential
tools for manipulating DNA including but not limited to cloning, analysis and
sequencing. Recent advances have made it possible to design and engineer chimaeric
nucleases to target specific DNA sites within a genome thus making them useful tools to
carry out gene therapy.
Introduction
The type II restriction endonucleases, also commonly
known as restriction enzymes, are molecular ‘scissors’ that
bind to specific sequences in deoxyribonucleic acid (DNA)
and cut within or adjacent to these sites. Their discovery in
the late 1960s ushered in a recombinant DNA technology
revolution in molecular biology. For their work in restriction
enzymes, three scientists were awarded the Nobel Prize
in Medicine (1978): Werner Arber of the Basel University,
for experiments showing the existence of restriction
endonucleases; Hamilton Smith, from Johns Hopkins
University, for the discovery of a restriction enzyme; and
Daniel Nathans, also of the Johns Hopkins University, for
demonstrating the utility of restriction enzymes for analysing
DNA. See also: Bacterial Restriction–Modification
Systems
Restriction Enzymes: Essential Tools
for Manipulating DNA
Restriction enzymes are endonucleases that recognize specific
DNA sequences and make double-strand cleavages.
Restriction endonucleases are divided into three groups,
type I, type II and type III based on their subunit composition,
cofactor requirement and enzymatic mechanism
(Table 1). They all require Mg2+
as cofactor. Type I and
type III also require adenosine triphosphate (ATP) and
S-adenosyl-L-methionine (AdoMet or SAM) as activators
and their cleavage sites are situated at a distance from their
recognition site. Cleavage sites for type II enzymes are
located at or close to their recognition site. The common
type II enzymes recognize specific sequences with a dyad
axis of symmetry, called palindromes, and cleave within or
adjacent to these sequences (Mani et al., 2003). Some enzymes
cleave at the axis of symmetry to yield ‘flush’ or
‘blunt’ ends, while others make staggered cuts to yield
overhanging single-stranded 3’ or 5’ ends, known as cohesive
termini. The phosphodiester bond cleavage results in
3’ hydroxyl and 5’-phosphate termini. DNA fragments
with compatible (i.e. complementary) ends may be ligated
to each other using DNA ligases to produce recombinant
DNA molecules. There are also numerous enzymes that
recognize an asymmetric sequence and cleave a short
distance from that sequence. These are termed as type
IIs enzymes, where ‘s’ stands for shifted cleavage. These
enzymes do not recognize any specific sequence at the
site cut.
The ability to manipulate DNA in defined ways has led
to many discoveries in molecular biology since the 1970s.
The essential tools of a genetic engineer are the enzymes
that catalyse specific reactions on DNA molecules. The
restriction enzymes that cleave DNA at discrete nucleotide
sequences are critical for carrying out the most important
reactions involved in recombinant DNA technology. The
discovery of restriction enzymes has made it possible to
purify homogeneous DNA fragments of defined length by
molecular cloning. The DNA fragments generated by restriction
enzymes are used as substrates for a wide variety
of other enzymatic manipulations. Furthermore, the specific
cleavage sites provide unique molecular landmarks for
obtaining a physical map of DNA. Thus, restriction enzymes
have proved to be essential tools for analysing and
manipulating DNA.
Cleavage Specificity of Restriction
Enzymes
Over 3000 restriction–modification enzymes have been
identified; these come from widely diverse organisms,
but mostly from bacteria. These enzymes fall into ‘isoschizomer’
(identically cleaving) groups with about 200
sequence specificities. REBASE, the restriction enzyme
database, provides information regarding restriction
Article Contents
Advanced article
. Introduction
. Restriction Enzymes: Essential Tools for Manipulating
DNA
. Cleavage Specificity of Restriction Enzymes
. Biochemical Properties of Restriction Enzymes
. Restriction Enzyme Production by Many Diverse Bacteria
. Restriction–Modification Systems
. Modification Methylases
. Restriction–Modification System in vivo
. Changing the Sequence Specificity of Restriction
Enzymes
. Summary
. Acknowledgement
doi: 10.1002/9780470015902.a0000973.pub3
ENCYCLOPEDIA OF LIFE SCIENCES & 2007, John Wiley & Sons, Ltd. www.els.net 1
enzymes, DNA methyltransferases, isoschizomers, neoschizomers,
recognition sequence, commercial availability
and references. Most restriction enzymes typically recognize
DNA sites that are 4–8 base pairs (bp) in length. Prototypes
of type II and type IIs restriction endonucleases
and modification methylases are shown in Table1. Enzymes
recognizing simple tetranucleotide and hexanucleotide palindromic
sequences are quite prevalent. There are 16 possible
palindromic tetranucleotide sites and 64 possible
hexanucleotide sites. Members of almost all of the possible
sequences of these types have been found. The discovery of
new enzymes involves tedious and time-consuming efforts
that entail extensive screening of bacteria and other microorganisms.
Even when a new enzyme is found, more
often than not it falls into one of the already discovered
isoschizomer groups. It has become increasingly difficult to
discover new specificities by random screening (Table 2).
Restriction enzymes appear to have a key biological
function: to protect cells from infection by foreign DNA
that would otherwise destroy them. A corresponding cognate
methylase protects the host genome from cleavage by
the restriction enzymes. Viral genomes are usually small
and are unlikely to carry the methylation pattern necessary
to render them resistant to cleavage by the restriction enzymes.
Thus it is likely that bacteria select for restriction–
modification systems with small recognition sites (4–6 bp)
because these sites occur more frequently in the viral
genome.
Biochemical Properties of Restriction
Enzymes
The type II enzymes are homodimers with subunits of
about 25–50 kDa. A divalent cation, preferably Mg2+
, is
required for cleavage. Since most type II enzymes recognize
palindromic sequences with 2-fold rotational symmetry, it
is expected that two enzyme subunits arranged symmetrically
bind the recognition sites. Crystal structures of several
enzyme–cognate site complexes have shown this to be the
case. In contrast, the type IIs enzymes that recognize an
asymmetric sequence appear to bind DNA as monomers.
How do the restriction enzymes recognize and cleave
DNA? In vitro studies suggest that restriction enzymes initially
bind DNA nonspecifically and then slide along the
DNA to find their recognition site. Once they locate their
binding sites, conformational changes in the DNA and the
enzyme occur to form the specific DNA–enzyme complex
that triggers DNA cleavage. Catalysis is mediated by a
hydroxyl ion that is formed as a result of abstraction of a
proton from a water molecule or through activation of
a water molecule by complexation with Mg2+
. The restriction
enzymes discriminate their substrate from specific and
nonspecific DNA sequences with several direct and indirect
sequence-recognition mechanisms. Direct recognition is
mediated through specific hydrogen bonds with purine and
pyrimidine bases within the recognition site and van der
ND, not determined; Nn, nonspecific sequence of length n bases
Enzyme Recognition site/cleavage site(↓)
AluI 5′ --- A G ↓ C T ---- 3′
3′ --- T C ↑ G A ---- 5′
BamHI 5′ – G ↓ G A T C C --- 3′
3′ - C C T A G ↑ G --- 5′
EcoRI 5′ --- G ↓ A A T T C --- 3′
3′ --- C T T A A ↑ G --- 5′
PstI 5′ ---- C T G C A ↓ G ---- 3′
3′ ---- G ↑ A C G T C ---- 5′
NotI 5′ ----- G C ↓ G G C C GC ---- 3′ ND
3′ ----- C G C C G G ↑ CG ---- 5′
PacI 5′ ----- T T A A T ↓ T A A ---- 3′ ND
3′ ----- A A T ↑ T A A T T ---- 5′
PleI 5′ ---- G A G T C (N4) ↓ ----- 3′ ND
3′ ---- C T C A G (N5) ↑ ----- 5′
FokI 5′ ----- G G A T G (N9) ↓ ------ 3′
3′ ----- C C T A C (N13) ↑ ------ 5′
Recognition site/methylation site (m)
3′--- C C T A C (N13) --- 5′
m
5′--- G G A T G (N9) --- 3′
m
3′--- G A C G T C --- 5′
m
5′--- C T G C A G ---- 3′
m
3′--- C T T A A G --- 5′
m
5′--- G A A T T C --- 3′
m
3′ - C C T A G G --- 5′
m
5′ – G G A T C C --- 3′
m
3′--- T C G A --- 5′
m
5′--- A G C T --- 3′
m
Table 1 Prototypes of type II and type I is restriction endonucleases and methylases
Restriction Enzymes
ENCYCLOPEDIA OF LIFE SCIENCES & 2007, John Wiley & Sons, Ltd. www.els.net2
Waals interactions with pyrimidines. Indirect interactions
are mediated through sequence-specific hydrogen bonds
to the phosphate backbone of the DNA substrate and
sequence-dependent DNA distortions.
Star activity
Some restriction endonucleases relax or alter their recognition
specificity under sub-optimal reaction conditions.
This altered specificity is termed as ‘star’ activity. Star activity
may be an inherent property of some enzymes. Conditions
that influence star activity are high glycerol
concentrations, low ionic strength, high enzyme concentration,
high pH, trace organic solvents and substitution of
Mg2+
by other divalent metals. EcoRI cleaves at the canonical
site G#AATTC, but at high pH and low ionic
strength it cleaves the sequence N#AATTN. Recent studies
have shown that EcoRI star activity results in the cleavage
of any site that differs from the canonical recognition site
by a single base substitution. Star activity can be avoided
by following the optimal buffer conditions recommended
by the manufacturer.
Restriction Enzyme Production by
Many Diverse Bacteria
Type II restriction enzymes are widespread in nature.
Many thousands of bacterial species have been examined
for the presence of restriction endonucleases, and they
have been found in all genera examined. One-quarter of
all bacterial species examined appear to contain one or
more type II restriction–modification systems. They have
been characterized in 11 of 13 phyla of bacteria and archaea.
Most of them are derived from bacilli or proteobacteria.
The four common sequence specificities, namely
CGCG, GGCC, CCGG and GATC, have been found
throughout the phylogenetic tree, including the archaea
and proteobacteria.
Barany and co-workers have cloned and purified TaqI
endonuclease isoschizomers from Thermus species
obtained from different regions of the globe. These grow
in hot spring temperatures ranging from 53 to 858C.
It appears that, although the protein sequences are generally
conserved and amino acid residues important for catalytic
activity are highly conserved, their thermostability
can vary significantly.
Van Etten and co-workers have shown that Chlorella
viruses that infect certain unicellular, eukaryotic, chlorellalike
green algae encode multiple DNA methyltransferases
and endonucleases. The recognition sites of these virusencoded
systems vary from 2 to 4 bp, in contrast to
bacterial enzymes, which recognize sequences of 4–8
base-pairs. The functional significance of the Chlorella
virus endonucleases remains a mystery.
Restriction–Modification Systems
Bacterial species contain restriction–modification systems
with genes that encode both a restriction endonuclease and
a methyltransferase that recognizes the same sequence. The
host DNA is fully protected from the action of the restriction
enzyme by the methyltransferase. However, invading
or infecting DNA from a plasmid or a phage is not likely to
carry the appropriate methylation and so it is susceptible to
cleavage by the restriction enzyme. If there are one or more
recognition sites on the incoming DNA, a single cut by the
restriction enzyme is likely to incapacitate the invading
DNA. This leads to the destruction of the DNA.
More than 200 restriction–modification systems have
been cloned. Although some of the type II restriction systems
are encoded on the plasmid, others have been found
on the chromosome. The most striking feature of the gene
organization of the restriction–modification systems is that
both the restriction enzyme and the methyltransferase
genes lie close to one another. In some cases, the two genes
are separated by a single, small open reading frame that
appears to control the expression of the system. The proximity
of the methyltransferase and restriction enzyme genes
may be necessary for some kind of coordinate expression
at the transcriptional level or at the translational level, or
this may be important to prevent their separation by
recombination.
Sequences are now known for more than 100 restriction
enzymes and 150 methyltransferase genes. There is no
Table 2 Salient features of restriction enzymes
Type I Type II Type III
Requires Mg2+
, ATP, AdoMet for
DNA cleavage
Requires Mg2+
for DNA cleavage Requires Mg2+
, ATP, AdoMet for
DNA cleavage
Requires Mg2+
and AdoMet for
methylation
Requires AdoMet for methylation Requires Mg2+
and AdoMet for
methylation
Cleaves as far as 1000 bp away from
recognition site
Cleaves at or near recognition site Cleaves 25–27 bp away from
recognition site
Consists of three different subunits
catalysing both restriction and
methylation of DNA
Restriction and methylation carried
out by separate enzymes
Consists of two subunits catalysing
both restriction and methylation of
DNA
Restriction Enzymes
ENCYCLOPEDIA OF LIFE SCIENCES & 2007, John Wiley & Sons, Ltd. www.els.net 3
significant sequence similarity between the restriction endonuclease
gene and the methyltransferase gene of any
cognate systems. Protein sequence alignment of several
methyltransferases indicates that the m5C-methyltransferases
contain several core motifs that are highly conserved,
along with a variable target recognition domain.
While the N6A-methyltransferases and N4C-methyltransferases
also show similarities, they contain only a
few conserved protein motifs.
Other bacterial restriction–modification systems include
type I and type III enzymes. The former comprises three
subunits: the R subunit, which contains the nuclease; the M
subunit, which contains the methylase; and the S subunit,
which contains the DNA sequence-specificity determinant.
The cleavage can occur as far as–1000 bp away from the
recognition site. Adenosine triphosphate (ATP) and AdoMet
are requiredfor the cleavage activity. Type III enzymes
comprise two subunits: a methylation subunit (Mod) and a
restriction subunit (Res). They do not require ATP for
cleavage and the cut site occurs fairly close to the recognition
site.
Modification Methylases
Each type II restriction enzyme has a counterpart modification
methylase. This binds to the same recognition
sequence and methylates one nucleotide within that
sequence (Table 1). The methyltransferase uses AdoMet as
a substrate and transfers the methyl group from AdoMet to
either a cytosine or an adenine residue within the recognition
sequence. This methylation renders the site insensitive
to cleavage by the restriction endonuclease. Three
types of methylations are used to provide protection
against the cognate restriction enzyme: N6-methyladenine,
N4-methylcytosine and 5-methylcytosine. Many laboratory
strains of Escherichia coli contain three site-specific
DNA methylases: Dam methylase, Dcm methylase and
EcoKI methylase.
The crystal structure of the HhaI methyltransferase
complexed with its cognate-binding site has revealed that
the enzyme flips its target base out of the DNA helix. The
structure of the enzyme suggests that it comprises two domains:
a small domain that has evolved the sequencespecific
DNA-binding apparatus and a large domain that
contains the catalytic apparatus. It appears that flipping a
base out of the DNA helix is a common feature employed
by other proteins that chemically modify bases in DNA.
Methylation is found in C residues of CG sequences in
animals and CNG sequences in plant cells. Although
the biological role of methylation in bacteria is fairly clear,
its importance in eukaryotes is still unclear. After eukaryotic
DNA replication, a maintenance methylase ensures
that all of the sites that were methylated in parental DNA
are methylated in daughter DNA. Methylation may be
responsible for tissue-specific inactivation of genes
during development. Bestor and co-workers have suggested
that the primary function of mammalian DNA
methyltransferase may be to suppress the parasitic sequence
elements that are present in the genome, including
endogenous retroviruses and transposable elements.
Restriction–Modification System
in vivo
An important characteristic of restriction enzymes is their
ability to discriminate their recognition sites from all other
sites. This is especially true in vivo. A restriction enzyme
would be lethal if it cleaved DNA readily at any sequence
other than its recognition site. In a cell containing a restriction–modification
system, all the recognition sites on
the chromosome are methylated by the methyltransferases.
Sequences that differ from the recognition site by one base
pair remain unmethylated. Under certain conditions, restriction
enzymes cleave DNA both at their recognition
sites and at a limited number of additional sites that
generally differ from the recognition site by one base pair.
Under optimal conditions, the ratio of activities at cognate
versus noncognate sites is very large. Reactions at noncognate
sites proceed via two successive single-strand
breaks. Under similar conditions, the same enzyme
produces double-strand breaks at the cognate sites. It appears
that the single-strand nicks that arise as a result of
cleavage at noncognate sites are readily sealed by the DNA
ligase that is present in the cell with no physiological
consequences.
A hemimethylated DNA, which has a methyl group on
one strand only, is a preferred substrate for the methylase
but not for the restriction enzymes, which generally cleave
only when the recognition site is unmethylated on both
strands. This is another important aspect of restriction
in vivo. Immediately after replication, there is hemimethylated
DNA within a cell. The genome needs to be resistant
to double-strand cleavage by the restriction endonuclease
that is present in the cell. In some cases, one strand may be
nicked. In these cases, the DNA ligase can readily seal the
nicks. The ability of ligase to repair the nicks can be critical,
especially when bacterial cells express high levels of restriction
endonucleases and the cognate methylase fails to
provide complete protection.
Type II restriction–modification systems are not easily
lost from their host cell. The progeny of cells that lose a
restriction–modification system are unable to modify a
sufficient number of recognition sites in their chromosome
to protect them from lethal attack by the remaining molecules
of restriction enzyme. Kobayashi and co-workers
have hypothesized that this capacity of restriction–modification
systems to act as a selfish genetic element may have
contributed to the spread and maintenance of restriction–
modification systems. This hypothesis remains to be tested.
Although in most restriction–modification systems the
expression of the restriction enzyme and that of methyltransferase
are tightly linked, it is still possible for a strain
Restriction Enzymes
ENCYCLOPEDIA OF LIFE SCIENCES & 2007, John Wiley & Sons, Ltd. www.els.net4
to have lost its restriction endonuclease but have retained
its methyltransferase intact.
Changing the Sequence Specificity of
Restriction Enzymes
Although the type II enzymes are very useful in manipulating
recombinant DNA, they are not suitable for producing
large DNA segments or for genome engineering.
For example, restriction enzymes that recognize six base
pairs result in cuts as often as every 4096 bases. Even rare
cutters (enzymes that recognize 8 bp-long sequences) cut
DNA once every 65 536 bases on average. So far, only a
limited number of restriction enzymes that recognize sequences
longer than six base pairs have been identified. In
many instances, it is preferable to have fewer but longer
DNA strands, especially during genome mapping. Therefore,
a long-term goal in the field of restriction–modification
enzymes has been to generate novel restriction
endonucleases with longer restriction sites by mutating or
engineering existing enzymes.
Before restriction enzymes can cleave DNA, they must
bind to the correct DNA sequence. Thus, they have a dual
function, namely DNA recognition and DNA cleavage. In
the case of type II enzymes, these functions overlap each
other. Several methods, including the bacteriophage P22
challenge-phage system, have been applied for the selection
of mutations that alter sequence specificity in restriction–
modification enzymes. However, attempts by genetic manipulation
of the existing type II enzymes to generate new
specificities, particularly longer recognition sites, have not
been successful. This may simply be because multiple mutations
are needed before a change in specificity can be
achieved. Alternatively, since the DNA recognition and
catalytic functions overlap each other in type II enzymes, it
is possible that attempts to change amino acid residues
within the DNA recognition domain that are responsible
for the sequence specificity may also affect the catalytic
domain. Changes in the DNA-binding domain may alter
the geometry of catalytic site; this is probably accompanied
by a drop in cleavage activity over several orders of magnitude.
Type II enzymes may simply not be suitable subjects
for changing sequence specificity.
Researchers have tried to generate universal restriction
enzymes by combining type IIs enzymes like FokI with
properly designed oligonucleotide adaptors. However, this
method is not as useful as chimaeric nucleases (see later),
since the target needs to be single-stranded DNA.
The type IIs enzymes, for example FokI, have been
shown to contain two domains: one responsible for DNA
sequence recognition and the other for DNA cleavage
(Figure 1). An elegant study of the crystal structures of FokI
with and without DNA by Aggarwal and co-workers has
shown this model to be correct. Based on the modular nature
of type IIs endonucleases, Chandrasegaran and coworkers
postulated that these enzymes probably evolved
by random fusions of the DNA-binding domains to nonspecific
endonucleases. Over time, these fusions were further
refined into sequence-specific type IIs restriction
enzymes by acquiring allosteric interaction between the
recognition domain and the catalytic domain. Recent studies
suggest that the modular architecture of type IIs enzymes
is much more common in nature than previously
thought. I-TevI, an intron-encoded homing endonuclease,
appears to have a similar bipartite structure (Figure 1).
Unlike FokI, in which the recognition domain is at the
N-terminus and the cleavage domain is at the C-terminal
third of the molecule, the homing endonuclease I-TevI
appears to be an enzyme with an N-terminal catalytic domain
and C-terminal DNA-binding domain connected by
a flexible linker. Recent studies suggest that similar multimodular
endonuclease fusions may be much more prevalent.
R2 retrotransposon endonuclease, Drosophila P1
transposase and Rec BCD enzyme involved in recombination
may fall into this category.
The type IIs enzymes appear to be ideal candidates for
changing sequence specificities. The modular structure of
FokI endonuclease suggested that it might be feasible to
construct chimaeric nucleases with novel sequence specificities
by linking other DNA-binding proteins to the cleavage
domain of FokI. Chandrasegaran and co-workers have
created the first chimaeric nucleases by fusing the isolated
nuclease domain of FokI to other sequence-specific DNAbinding
proteins. These include the three common
eukaryotic DNA-binding motifs, namely the helix–turn–
helix motif, the zinc-finger motif and the basic helix–loop–
helix protein (b-HLH) containing a leucine zipper motif
(Figure 2). These hybrid enzymes cleave the DNA at the
binding site preferred by the DNA-binding proteins. Increased
levels of ligase within cells have been utilized for the
production of chimaeric nucleases. Since there are no
counterpart methylases available for the chimaeric nucleases,
production of these enzymes in vivo is lethal to cells.
By increasing the levels of the DNA ligase within the cells,
the clones carrying the chimaeric nucleases are made more
viable.
(a)
R EN
(b)
REN
Figure 1 Representation of multimodular enzymes. (a) FokI (a type IIs
restriction enzyme); (b) I-TevI (a homing endonuclease). R, recognition
domain; EN, endonuclease domain.
Restriction Enzymes
ENCYCLOPEDIA OF LIFE SCIENCES & 2007, John Wiley & Sons, Ltd. www.els.net 5
The most important chimaeric restriction endonucleases
are those based on zinc-finger DNA-binding proteins.
Each individual zinc finger, a peptide of about 30 amino
acids, recognizes three bases along the DNA. These proteins,
like many sequence-specific DNA-binding proteins,
bind to the DNA by inserting an a helix into the major
groove of the double helix. The crystallographic structure
of the three zinc-finger proteins bound to cognate sites reveals
that each finger interacts with a triplet within the
DNA substrate. Each finger, because of variations of certain
key amino acids from one zinc finger to the next, makes
its own unique contribution to DNA-binding affinity and
specificity. Because they appear to bind as independent
modules, the zinc fingers can be linked together in a peptide
designed to bind a predetermined DNA site. In theory, one
can design a zinc finger for each of the 64 possible triplets;
by using a combination of these fingers, one could design a
protein for sequence-specific recognition of any segment of
DNA. Studies attempting to understand the rules relating
to zinc-finger sequences as well as their DNA-binding
preferences and redesigning of DNA-binding specificities
of zinc-finger protein are under way. An alternate
approach to the design of zinc-finger proteins with new
specificities involves the selection of desirable mutants
from a library of randomized zinc fingers displayed on
phage. The ability to design or select zinc-finger proteins
with desired specificity implies that DNA-binding proteins
containing zinc fingers will be made to order. Therefore,
one could design zinc-finger nucleases (ZFNs) that will cut
DNA at any preferred site by making fusions of zinc-finger
proteins to the cleavage domain of FokI endonuclease.
Zinc-finger proteins, because of their modular nature, offer
an attractive framework for chimaeric nucleases with
tailor-made sequence specificities.
Chandrasegaran and co-workers have fused three zinc
fingers to the nuclease domain and achieved cleavage at the
predicted 9-bp recognition site. It is immediately obvious
that, by combining different zinc fingers together, numerous
new DNA-binding specificities and cleavage patterns
could be achieved. Pabo and co-workers have reported the
design of poly zinc-finger proteins that bind to 18-bp DNA
sites with high affinity. These proteins could be converted
into site-specific cleavage enzymes by linking them to the
FokI cleavage domain.
How might these ZFNs be used in genome engineering?
One approach would be to recruit the preexisting cellular
machinery toachievethis goal.Insomaticorvegetativecells
of many different organisms, homologous recombination is
FokI
R EN
Zinc finger
Zn Zn Zn Zn Zn Zn
EN ZF-EN
Homeodomain
EN Ubx-EN
Leucine zipper
EN EN Gal4-EN
Figure 2 Representation of various chimaeric nucleases. ZF-EN, Ubx-EN and Gal4-EN were made by fusing the isolated nuclease domain (EN) of FokI to the zincfinger
motif, helix–turn–helix motif and Gal4, respectively. R, recognition domain; EN, endonuclease domain.
Restriction Enzymes
ENCYCLOPEDIA OF LIFE SCIENCES & 2007, John Wiley & Sons, Ltd. www.els.net6
used to repair DNA damage, especially double-strand
breaks (Porteus et al., 2006). Carroll and co-workers have
shown that making a targeted double-strand break would
greatly stimulate homologous recombination between the
exogenous DNA and a chromosomal sequence in frog
oocytes. Such experiments have been performed using intron-encoded
homing endonucleases in yeast, cultured
mammalian cells and plant cells. In collaboration with
Dana Carroll’s laboratory, we have successfully used ZFNs
to stimulate homologous recombination in frog oocytes.
Carroll’s group has extended this approach to show specific
chromosomal cleavage and mutagenesis of the yellow (Y)
gene by ZFNs in Drosophila melanogaster. More recently,
Porteus and Baltimore have described a gene targeting system
based on the correction of a mutated green fluorescent
protein (GFP) gene to show that chimaeric nucleases can
stimulate gene targeting in human somatic cells. Recent
studies have reliably used ZFNs to introduce chromosomal
double-strand breaks and stimulate homologous recombination
(HR) in C. elegans, cultured human cells (Moehle
et al., 2007; Urnov et al., 2005) and Arabidopsis. Gene targeting
using ZFNs is an emerging new technology. These
developments indicate that ZFNs could be used to correct
human genetic defects in the near future.
Summary
Since their discovery, type II restriction enzymes have
played a crucial role in the development of biotechnology
and the field of molecular biology. The restriction and
modification enzymes, which protect bacteria from phage
or foreign DNA, are important because they provide reagents
for recombinant DNA technology. They are the essential
tools for manipulating DNA. Restriction enzymes
play an important role in the cloning and sequencing of
numerous DNA fragments. Restriction mapping using restriction
enzymes involves the cleavage of DNA at specific
sites followed by the determination of the length of the
DNA fragments by gel electrophoresis. The physical map is
constructed after digesting the DNA of interest with a variety
of enzymes. Thus, restriction enzymes have played a
key role in the assembly of a physical map of the human
genome.
Although the type II enzymes are useful in manipulating
recombinant DNA, they are not suitable for producing
large DNA fragments or for genome engineering. Restriction
enzymes that recognize 15–18 bp sites would be invaluable
in genome engineering experiments. Because the
DNA recognition and catalytic functions of type II enzymes
overlap each other, attempts to change the sequence
specificity of these enzymes have not been successful. The
type IIs enzymes appear to be ideal candidates for change
sequence specificity. The modular nature of FokI restriction
endonuclease has made it possible to construct
chimaeric nucleases by linking other DNA-binding proteins
to the cleavage domain of FokI. The modular nature
of zinc-finger proteins enables the design or selection of
peptides that will bind DNA at any predetermined site. The
convergence of these two areas of research makes it possible
to create artificial nucleases that will cut DNA near a
predetermined site. By using these ZFNs, one can make
targeted double-strand breaks within a chromosome, and
thereby stimulate homologous recombination of exogenous
DNA with a chromosomal sequence.
A silent revolution is taking place in the field of restriction–modification
enzymes. Scientists are gathering important
information about structure, and about the
mechanism of DNA recognition and DNA cleavage by
many restriction endonucleases. There is real excitement
about the possibility of making ‘artificial’ restriction enzymes
that will recognize a particular site within a genome
and cleave near that site. We may be able to generate many
novel enzymes with tailor-made sequence specificities that
are desirable for various applications. Ultimately, we
might be able to target specific genes for cleavage within
cells. In the future, ZFNs are likely to provide the second
generation of the molecular scissors that are such important
diagnostic and therapeutic reagents for the research
community. Also, the complete nucleotide sequence of the
genomes of many organisms, including the human genome,
are now known. The availability of chimaeric nucleases
that target a specific site within a genome should make it
feasible to carry out gene therapy.
Acknowledgement
The work on chimaeric nucleases in Prof Chandrasegaran’s
lab. is funded by a grant from NIH (GM 53923).
References
Mani M, Kandavelou K, Wu J and Chandrasegaran S (2003)
Restriction enzymes. In: Cooper DN (ed.) Nature Encyclopedia
of the Human Genome, vol. 5, pp. 31–35. London: Nature
Publishing Group.
Moehle EA, Rock JM, Lee YL et al. (2007) Targeted gene
addition into a specified location in the human genome using
designed zinc finger nucleases. Proceedings of the National
Academy of Sciences of the USA 104: 3055–3060.
Porteus MH, Connelly JP and Pruett SM (2006) A look to future
directions in gene therapy research for monogenic diseases.
PLoS Genetics 2: e133.
Urnov FD, Miller JC, Lee YL et al. (2005) Highly efficient
endogenous human gene correction using designed zinc-finger
nucleases. Nature 435: 646–651.
Further Reading
Aggarwal AK and Wah DA (1998) Novel site-specific DNA
endonuclease. Current Opinion in Structural Biology 8(1):
19–25.
Berg JM (1995) Zinc finger domains: from prediction to design.
Accounts of Chemical Research 28: 14–19.
Restriction Enzymes
ENCYCLOPEDIA OF LIFE SCIENCES & 2007, John Wiley & Sons, Ltd. www.els.net 7
Bibikova M, Beumer K, Trautman JK and Carroll D (2003)
Enhancing gene targeting with designed zinc finger nucleases.
Science 300: 764.
Bibikova M, Carroll D, Segal DJ et al. (2001) Stimulation of
homologous recombination through targeted cleavage by chimeric
nucleases. Molecular and Cellular Biology 21: 289–297.
Bibikova M, Golic M, Golic KG and Carroll D (2002) Targeted
chromosomal cleavage and mutagenesis in Drosophila using
zinc finger nucleases. Genetics 161: 1169–1175.
Carroll D (1996) Homologous genetic recombination in Xenopus:
mechanisms and implications for gene manipulation. Progress
in Nucleic Acid Research and Molecular Biology 54: 101–125.
Choo Y and Klug A (1997) Physical basis of a protein–DNA recognition
code. Current Opinion in Structural Biology 7: 117–125.
Durai S, Mani M, Kandavelou K et al. (2005) Zinc finger nucleases:
custom-designed molecular scissors for genome engineering
of plant and mammalian cells. Nucleic Acids Research
33: 5978–5990.
Gormley NA, Watson MA and Halford SE (2005) Bacterial
restriction–modification systems, doi.wiley.com/10.1038/npg.
els.0003897.
Kandavelou K, Mani M, Durai S and Chandrasegaran S (2004)
Engineering and applications of chimeric nucleases. In:
Pingoud AM (ed.) Nucleic Acids and Molecular Biology,
vol. 14, pp. 413–434. Berlin: Springer-Verlag.
Kandavelou K, Mani M, Durai S and Chandrasegaran S (2005)
‘‘Magic’’ scissors for genome surgery. Nature Biotechnology 23:
686–687.
Kandavelou K, Mani M, Reddy SPM and Chandrasegaran S
(2005) Ligation: theory and practice, doi.wiley.com/10.1038/
npg.els.0003838.
Kim J-S and Pabo CO (1998) Getting a handhold on DNA: design
of poly zinc finger proteins with femtomolar dissociation
constants. Proceedings of the National Academy of Sciences of
the USA 95: 2812–2817.
Kim Y-G, Cha J and Chandrasegaran S (1996) Hybrid restriction
enzymes: zinc finger fusion to FokI cleavage domain. Proceedings
of the National Academy of Sciences of the USA 93:
1156–1160.
Li L, Wu LP and Chandrasegaran S (1992) Functional domains in
FokI restriction endonuclease. Proceedings of the National
Academy of Sciences of the USA 89: 4275–4279.
New England Biolabs (2000–2001) New England Biolabs Catalog
and Technical Reference. Beverly, MA: New England Biolabs.
Porteus MH andBaltimore D(2003) Chimeric nucleasesstimulate
gene targeting in human cells. Science 300: 763.
Roberts RJ and Cheng X (1998) Base flipping. Annual Review of
Biochemistry 67: 181–198.
Roberts RJ and Halford SE (1993) Type II restriction endonuclease.
In: Linn SM, Lloyd RS and Roberts RJ (eds) Nuclease,
pp. 35–38. Cold Spring Harbor, NY: Cold Spring Harbor
Laboratory Press.
Roberts RJ, Vincze T, Posfai J and Macelis D (2003) REBASE:
restriction enzymes and methyltransferase. Nucleic Acids
Research 31: 418–421(http://rebase.neb.com/rebase/rebase.
html).
Samuelson JC, Morgan RD, Benner JS et al. (2006) Engineering a
rare-cutting restriction enzyme: genetic screening and selection
of NotI variants. Nucleic Acids Research 34: 796–805.
Sistla S and Rao DN (2004) S-Adenosyl-L-methionine-dependent
restriction enzymes. Critical Reviews in Biochemistry and
Molecular Biology 39: 1–19.
Szybalski W, Kim SC, Hasan N and Podhajska AJ (1991) ClassIIs
restriction enzymes – a review. Gene 100: 13–26.
Williams RJ (2001) Restriction endonucleases and their uses.
Methods in Molecular Biology 160: 409–429.
Restriction Enzymes
ENCYCLOPEDIA OF LIFE SCIENCES & 2007, John Wiley & Sons, Ltd. www.els.net8