# Restriction Enzyme Definition: How They Cut DNA

A restriction enzyme is a protein that recognizes a specific, short DNA sequence and cleaves the phosphodiester bonds within or near that sequence. These enzymes are also called restriction endonucleases because they cut DNA from within the molecule, rather than chewing from the ends like exonucleases. Their discovery in the late 1960s and early 1970s transformed [molecular biology](/blog/careers/molecular-biology) by providing scientists with precise molecular scissors—tools that could reproducibly cut DNA at defined positions. Today, restriction enzymes are foundational to [molecular cloning](/knowledge/molecular-biology/molecular-cloning-a-laboratory-manual), DNA mapping, and genetic engineering.

## What Is a Restriction Enzyme?

A restriction enzyme is a bacterial protein that binds to a specific double-stranded DNA sequence, typically 4–8 base pairs long, and hydrolyzes the phosphodiester backbone at defined positions. The term "restriction" comes from the enzyme's natural role: restricting the replication of foreign DNA, such as bacteriophage genomes, inside a bacterial cell. When a phage injects its DNA into a bacterium, restriction enzymes recognize the foreign DNA—which lacks the bacterium's protective methylation marks—and cut it into fragments, disabling the phage.

The recognition sequences are usually palindromic, meaning the sequence reads the same forward on one strand and backward on the complementary strand. For example, the enzyme *Eco*RI recognizes the sequence 5'-GAATTC-3' on one strand and 3'-CTTAAG-5' on the other. Reading the top strand left to right (GAATTC) and the bottom strand left to right (CTTAAG) gives the same sequence when read in the 5'-to-3' direction on each strand. This symmetry is critical because the enzyme must bind both strands and cut both strands at equivalent positions.

### Natural Function in Bacteria

Restriction enzymes are part of a bacterial defense system called the restriction-modification system. The bacterium protects its own DNA by adding methyl groups to adenine or cytosine residues within the recognition sequence. This methylation is performed by a companion enzyme called a methyltransferase. The restriction enzyme will not cut DNA that carries these methylation marks, so the bacterial genome remains intact. Foreign DNA, which is unmethylated, is recognized as non-self and destroyed.

This system is analogous to a lock and key: the restriction enzyme is the lock, the recognition sequence is the keyhole, and methylation is the protective cover that prevents the key from fitting. More than 3,000 restriction enzymes have been characterized from various bacterial species, each recognizing a distinct sequence. Some recognize the same sequence but cut at different positions; these are called neoschizomers. Others recognize the same sequence and cut at the same position but come from different organisms; these are isoschizomers.

### Recognition Sequences

Recognition sequences range from 4 to 8 base pairs. The frequency of a recognition site in a random DNA sequence depends on its length. A 4-base-pair cutter, such as *Mbo*I (GATC), will occur approximately once every 256 base pairs (4⁴). A 6-base-pair cutter, such as *Eco*RI (GAATTC), occurs roughly once every 4,096 base pairs (4⁶). An 8-base-pair cutter, such as *Not*I (GCGGCCGC), occurs about once every 65,536 base pairs (4⁸). This frequency calculation assumes random base composition, which is rarely exact, but it provides a useful estimate for planning digests.

The human genome is approximately 3.1 billion base pairs. A 6-base-pair cutter would be expected to cut it roughly 750,000 times, producing fragments too small to be useful for whole-genome analysis. An 8-base-pair cutter would cut approximately 47,000 times, producing large fragments that can be resolved by [pulsed-field gel electrophoresis](/knowledge/diagnostics/molecular/pulsed-field-gel-electrophoresis). This is why rare cutters like *Not*I are used for generating large DNA fragments for genome mapping.

## How Restriction Enzymes Work

Restriction enzymes recognize their target sequence through direct contacts between amino acid residues in the enzyme's DNA-binding domain and the functional groups exposed in the major and minor grooves of the DNA double helix. These contacts include hydrogen bonds, van der Waals interactions, and electrostatic interactions with the phosphate backbone. The enzyme scans the DNA by a combination of three-dimensional diffusion and one-dimensional sliding along the helix until it encounters a matching sequence.

Once bound, the enzyme undergoes a conformational change that positions two catalytic residues—typically glutamic acid and lysine—near the scissile phosphodiester bonds. These residues coordinate a magnesium ion (Mg²⁺) that is absolutely required for catalysis. The magnesium ion activates a water molecule, which performs a nucleophilic attack on the phosphorus atom of the phosphate group. This breaks the phosphodiester bond, leaving a 5'-phosphate group on one fragment and a 3'-hydroxyl group on the other.

### Sticky Ends vs. Blunt Ends

[Restriction enzymes cut DNA](/knowledge/molecular-biology/restriction-enzymes-cut-dna) in two patterns: staggered cuts that produce sticky ends (also called cohesive ends) and straight cuts that produce blunt ends.

A sticky-end cut occurs when the enzyme cleaves the two strands at different positions, leaving short single-stranded overhangs. For example, *Eco*RI cuts the sequence GAATTC between the G and the A on both strands, but the cuts are offset by four bases. The resulting fragments have 5'-overhangs of four bases: AATT. These overhangs are complementary to each other, so fragments cut with the same enzyme can anneal through base pairing. This complementarity is exploited in cloning to join DNA fragments from different sources.

A blunt-end cut occurs when the enzyme cleaves both strands at the same position, leaving no overhang. For example, *Sma*I recognizes CCCGGG and cuts straight through the middle, producing flush ends. Blunt ends are less efficient for ligation because they lack the stabilizing base-pairing of sticky ends, but they are useful when no suitable sticky-end enzyme is available.

The table below summarizes the key differences:

| Feature | Sticky Ends | Blunt Ends |
|---|---|---|
| Cut pattern | Staggered, with 1–4 base overhangs | Straight through both strands |
| Example enzyme | *Eco*RI (5'-G^AATTC-3') | *Sma*I (CCC^GGG) |
| Ligation efficiency | High due to complementary overhangs | Lower, requires higher DNA concentration |
| Compatibility | Only with fragments cut by the same enzyme or a compatible enzyme | Any blunt-ended fragment can be ligated to any other |
| Use in cloning | Preferred for directional cloning | Used when no suitable sticky-end site exists |

### Role of Magnesium Ions

Magnesium ions are essential cofactors for essentially all Type II restriction enzymes. The enzyme binds two Mg²⁺ ions per active site, although some enzymes require only one. These ions serve two functions: they stabilize the transition state of the phosphodiester bond cleavage, and they position the water molecule for nucleophilic attack. Without magnesium in the reaction buffer, the enzyme binds DNA but cannot cut it. Most commercial restriction enzyme buffers contain 10 mM magnesium acetate or magnesium chloride.

The requirement for magnesium also explains why EDTA (ethylenediaminetetraacetic acid) stops restriction digests. EDTA chelates divalent cations, sequestering the magnesium and making it unavailable for catalysis. Adding EDTA to a final concentration of 10–20 mM is a standard method to terminate a restriction digest.

## Types of Restriction Enzymes

Restriction enzymes are classified into three main types—I, II, and III—based on their subunit composition, cofactor requirements, and cleavage position relative to the recognition sequence. A fourth type, Type IV, targets methylated DNA and is less commonly used in cloning.

### Type II Enzymes

Type II restriction enzymes are the workhorses of [molecular biology](/blog/careers/molecular-biology). They are simple enzymes composed of a single protein that recognizes a specific sequence and cuts within or immediately adjacent to that sequence. They require only magnesium as a cofactor and do not need ATP. Because their cleavage position is predictable and precise, Type II enzymes are the ones used in virtually all cloning applications.

Type II enzymes are further subdivided based on the symmetry of their recognition sequence and the position of the cut. Type IIP enzymes recognize palindromic sequences and cut symmetrically; *Eco*RI and *Hind*III are examples. Type IIS enzymes recognize asymmetric sequences and cut at a defined distance away from the recognition site. For example, *Bsa*I recognizes GGTCTC and cuts 1 base downstream on the top strand and 5 bases downstream on the bottom strand. Type IIS enzymes are valuable for generating custom overhangs in techniques like Golden Gate assembly.

### Other Types

Type I enzymes recognize a specific sequence but cut at a random position up to 1,000 base pairs away. They require ATP and S-adenosylmethionine as cofactors and function as large multi-subunit complexes. Because their cleavage position is unpredictable, Type I enzymes are not useful for cloning.

Type III enzymes recognize a specific sequence and cut approximately 25–27 base pairs downstream. They require ATP but not for cleavage itself; ATP is needed for DNA translocation. Type III enzymes produce partial digests and are rarely used in molecular biology.

Type IV enzymes recognize methylated, hydroxymethylated, or glucosyl-hydroxymethylated DNA. They are involved in restriction of foreign DNA that carries methylation marks, but they are not used in standard cloning.

## Discovery and Historical Significance

The discovery of restriction enzymes is a story of incremental insight spanning two decades. In the 1950s, Salvador Luria and Giuseppe Bertani observed that bacteriophages grown in one strain of *Escherichia coli* grew poorly when transferred to another strain. This phenomenon was called host-controlled restriction. In the 1960s, Werner Arber showed that this restriction was caused by enzymatic cleavage of the phage DNA and that the bacterium's own DNA was protected by methylation.

In 1970, Hamilton Smith isolated the first Type II restriction enzyme, *Hin*dII, from *Haemophilus influenzae*. He demonstrated that it cut DNA at a specific sequence, making it the first enzyme known to recognize a defined DNA sequence. Shortly thereafter, Daniel Nathans used *Hin*dII to cut the genome of the tumor virus SV40 into specific fragments, producing the first restriction map. Arber, Smith, and Nathans shared the 1978 Nobel Prize in Physiology or Medicine for their work.

The practical impact was immediate. In 1973, Stanley Cohen and Herbert Boyer used a restriction enzyme to cut a plasmid and a foreign DNA fragment, then ligated them together to create the first recombinant DNA molecule. This experiment launched the field of genetic engineering. The ability to cut DNA at precise locations made it possible to isolate genes, amplify them in bacteria, and produce proteins like insulin and growth hormone recombinantly. For a deeper look at how plasmids serve as the vehicles for this work, see the [Plasmid Definition](/knowledge/molecular-biology/plasmid-definition).

## Naming and Classification

Restriction enzymes are named according to a systematic convention established by Smith and Nathans. The name consists of three italicized letters derived from the bacterial genus and species, followed by a strain designation and a Roman numeral indicating the order of discovery from that strain.

The first letter of the genus name and the first two letters of the species name form the three-letter core. For example, *Eco*RI comes from *Escherichia coli*: E for *Escherichia*, co for *coli*. The strain designation follows: R for the RY13 strain. The Roman numeral I indicates it was the first enzyme isolated from that strain. Thus, *Eco*RI is the first restriction enzyme isolated from *E. coli* strain RY13.

Other examples: *Hind*III is the third enzyme from *Haemophilus influenzae* strain d. *Bam*HI is the first enzyme from *Bacillus amyloliquefaciens* strain H. *Not*I is the first enzyme from *Nocardia otitidis-caviarum*.

Enzymes are classified by their recognition sequence and cut position. The recognition sequence is written 5' to 3' on the top strand. The cut position is indicated by a caret (^) or by numbering the bases. For example, *Eco*RI cuts as 5'-G^AATTC-3', meaning the cut occurs between the G and the A on the top strand. The bottom strand cut is offset by four bases in the 5' direction, producing the AATT overhang.

## Applications in Molecular Cloning

Restriction enzymes are indispensable for constructing recombinant DNA molecules. The general workflow involves cutting a plasmid vector and an insert DNA with the same restriction enzyme (or enzymes), purifying the digested fragments, and ligating them together. For a comprehensive treatment of how these enzymes fit into the broader toolkit, see the [Restriction Enzymes Overview](/knowledge/molecular-biology/restriction-enzymes-overview).

### Creating Recombinant DNA

The process of creating a recombinant plasmid follows a defined sequence of steps:

1. **Select restriction enzymes.** Choose one or two enzymes that cut the [multiple cloning site](/knowledge/diagnostics/molecular/multiple-cloning-site-plasmids-structure-function) (MCS) of the plasmid and that also cut at the ends of your insert DNA without cutting inside the insert. Using two different enzymes with incompatible overhangs ensures directional cloning—the insert can only ligate in one orientation.

2. **Digest the vector and insert.** Set up separate reactions containing the DNA, the restriction enzyme(s), and the appropriate buffer. A typical reaction contains 1 µg of DNA, 10 units of enzyme, and 1× buffer in a total volume of 50 µL. Incubate at the enzyme's optimal temperature, usually 37°C, for 1–2 hours.

3. **Purify the digested DNA.** Remove the enzymes and buffer components using a spin column or gel extraction. This step is essential because residual enzyme activity can interfere with ligation.

4. **Ligate the insert into the vector.** Mix the digested vector and insert at a molar ratio of approximately 1:3 (vector:insert). Add T4 DNA ligase and ATP, and incubate at 16°C for several hours or overnight. The ligase seals the nicks in the phosphodiester backbone.

5. **Transform into competent bacteria.** Introduce the ligated plasmid into *E. coli* cells by heat shock or electroporation. Plate on selective media containing an antibiotic such as ampicillin or kanamycin. Only cells that have taken up the plasmid will survive.

6. **Screen for correct clones.** Pick individual colonies, grow them in liquid culture, and isolate the plasmid DNA. Verify the presence of the insert by restriction digestion followed by agarose gel electrophoresis or by DNA sequencing.

The choice of restriction enzyme determines the compatibility of the ends. If the vector and insert are both cut with *Eco*RI, they will have complementary AATT overhangs and can anneal. If the vector is cut with *Eco*RI and the insert has blunt ends, the ligation will be inefficient because the sticky end cannot base-pair with a blunt end. In such cases, the ends can be modified—for example, by filling in the overhang with a DNA polymerase to create blunt ends—but this reduces efficiency.

### Restriction Mapping

Restriction enzymes are also used to create restriction maps, which are diagrams showing the positions of restriction sites along a DNA molecule. To construct a map, you digest the DNA with one enzyme at a time and with combinations of enzymes, then separate the fragments by agarose gel electrophoresis. The fragment sizes are determined by comparison to a DNA ladder of known sizes. By analyzing the fragment patterns from single and double digests, you can deduce the relative positions of the restriction sites.

Restriction mapping is used to verify the structure of cloned DNA, to compare related plasmids, and to identify mutations that create or destroy restriction sites. It is also used in forensic [DNA fingerprinting](/blog/guides/dna-fingerprinting), where variations in restriction fragment lengths between individuals are detected by Southern blotting.

## Factors Affecting Enzyme Activity

Restriction enzymes are sensitive to reaction conditions. The manufacturer's specifications for buffer, temperature, and incubation time should be followed precisely. Deviations can lead to reduced activity, non-specific cleavage, or complete failure of the digest.

The standard reaction buffer for most Type II enzymes contains Tris-HCl (pH 7.5–8.5), sodium chloride or potassium acetate (at concentrations ranging from 0 to 100 mM), magnesium chloride (10 mM), and sometimes bovine serum albumin (BSA) to stabilize the enzyme. The salt concentration is particularly important because it affects the enzyme's DNA-binding affinity. Some enzymes, like *Sal*I, require high salt (100 mM NaCl), while others, like *Sma*I, work best at low salt (0 mM).

The optimal temperature is usually 37°C, but there are exceptions. *Sma*I works best at 25°C, and *Taq*I (from *Thermus aquaticus*) has an optimal temperature of 65°C. Incubating at the wrong temperature can reduce activity or cause star activity (discussed below).

### Star Activity

Star activity is the relaxation of a restriction enzyme's specificity under non-optimal conditions. Under normal conditions, *Eco*RI cuts only GAATTC. Under star conditions, it will also cut GAATTA, GAGTTC, and other related sequences. Star activity is caused by conditions that weaken the enzyme's discrimination between the correct and incorrect sequences, including:

- Low salt concentration (below 25 mM for enzymes that normally require high salt)
- High glycerol concentration (above 5% v/v in the final reaction)
- High pH (above 8.0)
- The presence of organic solvents like ethanol or DMSO
- Excess enzyme (more than 100 units per µg of DNA)
- Prolonged incubation (more than 16 hours)

To avoid star activity, use the buffer recommended by the manufacturer, keep the glycerol concentration below 5% by ensuring the enzyme volume is no more than 10% of the total reaction volume, and do not exceed the recommended enzyme-to-DNA ratio.

### Methylation Sensitivity

Many restriction enzymes are sensitive to DNA methylation. If the recognition sequence contains a methylated cytosine or adenine, the enzyme will not cut. This is a natural consequence of the restriction-modification system: the enzyme must not cut the bacterium's own methylated DNA.

In the laboratory, methylation sensitivity can be a problem when working with DNA isolated from *E. coli*, which methylates adenine at GATC sites (via the Dam methylase) and cytosine at CCAGG and CCTGG sites (via the Dcm methylase). For example, *Mbo*I recognizes GATC and will not cut Dam-methylated DNA, while its isoschizomer *Sau*3AI cuts both methylated and unmethylated DNA. If you need to cut DNA that may be methylated, choose an enzyme that is not methylation-sensitive or use a Dam⁻ Dcm⁻ *E. coli* strain to propagate the plasmid.

Methylation sensitivity is also relevant in the study of [Epigenetics Definition](/knowledge/molecular-biology/epigenetics-definition), where restriction enzymes that discriminate between methylated and unmethylated DNA are used to assay methylation status at specific loci.

## Common Mistakes and How to Avoid Them

Even experienced researchers make errors when working with restriction enzymes. The following are the most common failure modes and how to prevent them.

**Choosing the wrong enzyme.** The most frequent mistake is selecting an enzyme that cuts inside the insert DNA. Before setting up a digest, check the insert sequence for the presence of the recognition site. Use a sequence analysis tool or simply scan the sequence manually. If the enzyme cuts inside the insert, the insert will be fragmented and cannot be cloned intact.

**Using the wrong buffer.** Each enzyme has a specific optimal buffer, usually supplied as a 10× concentrate by the manufacturer. Using the wrong buffer can reduce activity or cause star activity. Always check the compatibility chart if you are using multiple enzymes in a single reaction. Some buffers are compatible across multiple enzymes; others are not. If two enzymes require different buffers, digest sequentially: perform the first digest, purify the DNA, then perform the second digest in the appropriate buffer.

**Incorrect incubation temperature.** Most enzymes work at 37°C, but not all. Check the product manual for the optimal temperature. *Sma*I at 37°C will show reduced activity; *Taq*I at 37°C will be nearly inactive.

**Contamination with nucleases.** Restriction digests are ruined by contaminating nucleases that degrade the DNA non-specifically. Use sterile pipette tips and tubes, and avoid touching the inside of tube lids with your fingers. If your DNA preparation contains residual nucleases, the digest may show a smear instead of distinct bands on a gel. Purify the DNA with a spin column before digestion.

**Not checking for multiple sites.** If your plasmid or insert contains more than one recognition site for the chosen enzyme, you will get more fragments than expected. This is not necessarily a problem—it may be exactly what you want—but it must be accounted for. Always map the restriction sites in your DNA before designing the experiment.

**Insufficient enzyme or DNA.** Using too little enzyme results in a partial digest. Using too much enzyme can cause star activity. The general guideline is 10 units of enzyme per µg of DNA, but this should be adjusted based on the enzyme's specific activity and the quality of the DNA preparation.

**Forgetting to heat-inactivate.** After the digest, you may want to heat-inactivate the enzyme to stop the reaction. Most enzymes are inactivated by heating at 65°C or 80°C for 20 minutes. However, some enzymes are not heat-labile and cannot be inactivated this way. In that case, purify the DNA by spin column or gel extraction before proceeding to the next step.

**Ligating without purification.** If you do not remove the restriction enzyme before ligation, the enzyme can re-cut the ligated product, reducing cloning efficiency. Always purify the digested DNA before adding ligase.

## Frequently Asked Questions

### What is a restriction enzyme simple definition?

A restriction enzyme is a protein that cuts DNA at a specific, short sequence of nucleotides. It is also called a [restriction endonuclease](/knowledge/molecular-biology/restriction-endonuclease). Each enzyme recognizes a particular sequence, usually 4–8 base pairs long, and cleaves the DNA backbone at defined positions within or near that sequence.

### What is the function of restriction enzymes in bacteria?

In bacteria, restriction enzymes serve as a defense mechanism against foreign DNA, such as bacteriophage genomes. They recognize unmethylated DNA and cut it into fragments, preventing the phage from replicating. The bacterium's own DNA is protected by methylation at the same recognition sequences, which blocks the enzyme from cutting.

### How do restriction enzymes recognize specific DNA sequences?

Restriction enzymes recognize their target sequences through direct molecular contacts between amino acid residues in the enzyme's DNA-binding domain and the chemical groups exposed on the DNA double helix. The enzyme scans the DNA by sliding along the helix and diffusing in three dimensions until it encounters a matching sequence. The recognition sequence is typically palindromic, which allows the enzyme to bind symmetrically to both strands.

### What are sticky ends and blunt ends?

Sticky ends are short, single-stranded overhangs produced when a restriction enzyme cuts the two DNA strands at different positions. These overhangs are complementary to each other, allowing fragments cut with the same enzyme to anneal. Blunt ends are produced when the enzyme cuts both strands at the same position, leaving no overhang. Sticky ends ligate more efficiently than blunt ends.

### Why are restriction enzymes important in genetic engineering?

Restriction enzymes allow scientists to cut DNA at precise locations, making it possible to isolate specific genes, insert them into plasmids, and create recombinant DNA molecules. They are essential for molecular cloning, restriction mapping, and the production of recombinant proteins. For more on the downstream applications, see the [Recombinant Protein Definition](/knowledge/molecular-biology/recombinant-protein-definition).

### What is star activity in restriction enzymes?

Star activity is the loss of specificity that occurs when a restriction enzyme cuts sequences similar to, but not identical with, its canonical recognition sequence. It is caused by non-optimal reaction conditions such as low salt, high glycerol, high pH, excess enzyme, or prolonged incubation. Star activity produces extra, unwanted fragments.

### How are restriction enzymes named?

Restriction enzymes are named using a three-letter abbreviation derived from the bacterial genus and species, followed by a strain designation and a Roman numeral. For example, *Eco*RI comes from *Escherichia coli* strain RY13 and was the first enzyme isolated from that strain. The first letter is the genus, the next two letters are the species, and the Roman numeral indicates the order of discovery.

## Key Takeaways

- Restriction enzymes are bacterial proteins that cut DNA at specific recognition sequences, typically 4–8 base pairs long, and are also called restriction endonucleases.
- They function in bacteria as a defense against foreign DNA, cutting unmethylated phage genomes while the bacterium's own methylated DNA is protected.
- Type II restriction enzymes are the most useful for molecular biology because they cut at precise, predictable positions and require only magnesium as a cofactor.
- Sticky ends (cohesive overhangs) ligate efficiently and enable directional cloning; blunt ends ligate less efficiently but are compatible with any other blunt end.
- Restriction enzymes require specific buffer conditions, including the correct salt concentration, pH, and temperature, and are sensitive to DNA methylation.
- Star activity—non-specific cutting—occurs under suboptimal conditions such as low salt, high glycerol, or excess enzyme, and must be avoided for reliable results.
- Restriction enzymes are named systematically based on the bacterial species and strain of origin, and they are classified by their recognition sequence and cut position.
- These enzymes enabled the first recombinant DNA experiments and remain essential tools for cloning, restriction mapping, and genetic engineering. For the underlying mechanism of DNA cutting, see [Restriction Enzymes Cut DNA](/knowledge/molecular-biology/restriction-enzymes-cut-dna), and for the enzyme class itself, see [Restriction Endonuclease](/knowledge/molecular-biology/restriction-endonuclease).

## Related Clinical & Scientific Guides

* [MAPK Pathway: Mechanism, Function, and Clinical Relevance](/knowledge/molecular-biology/mapk-pathway)
* [Mammalian Cell Culture Bioreactors: A Practical Guide](/knowledge/molecular-biology/mammalian-cell-culture-bioreactor)
* [Nucleotide Formation: Biosynthesis and Assembly of DNA/RNA Building Blocks](/knowledge/molecular-biology/nucleotide-formation)