Where would modern-day molecular biology research be without restriction enzymes? These workhorses of the lab have been behind many of the advances in basic biological research and commercial applications for over 40 years. Restriction enzymes (or restriction endonucleases) were first identified in bacteria but have been subsequently found in some archaea. In general, restriction enzymes cleave double-stranded DNA. Each restriction enzyme recognizes specific DNA sequences, and cleavage can occur within the recognition sequence or some distance away, depending on the enzyme. The recognition sequences are generally 4 to 8 base pairs (bp) in length, and cleavage can produce sticky ends (5′ or 3′ protruding ends) or blunt ends (Figure 1).

Illustration of the protruding or blunt ends produced by specific restriction enzymes
Figure 1. Illustration of the sticky or protruding ends (5′ or 3′) or blunt ends produced by specific restriction enzymes.

Today about 4,000 restriction enzymes have been characterized, and over 600 of those are commercially available. REBASE is a useful, browsable resource for comprehensive and up-to-date information about restriction enzymes, including specificity, sensitivity, and commercial sources [1].

History of restriction enzymes

In the early 1950s, a number of research teams observed differences in the efficiency of bacteriophage infection on different bacterial host strains of the same species [2,3]. This was described by Grasso and Paigen: when phage λ propagated in one strain of bacteria (e.g., E. coli C) was used to infect another strain of the same species of bacteria (e.g., E. coli K), a marked decrease in the rate of infection was noted compared to re-infection of the host strain (E. coli C). The new host (E. coli K) seemed to select against or “restrict” the incoming phage. The researchers also noted this was not a hereditary phenomenon, because the phage that did grow on the new strain could infect that strain at more typical rates after one round of infection. The observed phenomenon was defined as “host control variation” and became an area of intense research to discover the underlying mechanisms [4].

It was not until the 1960s that mechanisms underlying host control variation were determined to involve enzymatic cleavage of the phage DNA, which led to the discovery and isolation of restriction enzymes. In the early 1960s, Werner Arber observed that the host-range determinant resided on the phage DNA, and subsequent experiments showed that methionine was involved in host protection [5]. These findings ultimately led to the proposal of a restriction-modification (R-M) system, in which a restriction enzyme and a methylase from the host work together to cleave foreign viral (non-methylated) DNA while keeping the host DNA protected through methylation [6].

Interestingly, most of the early work on R-M systems was on Type I and III groups of restriction enzymes, classified based on aspects of their structure and function (see Restriction enzyme classification). However, the complete utility of restriction enzymes did not become apparent until Kent Wilcox and Hamilton Smith discovered HindII, the first restriction enzyme of the Type II class [7]. HindII recognizes a specific symmetrical DNA sequence and cleaves in a defined manner within that recognition sequence. This feature, found in most early Type II restriction enzymes, led Kathleen Danna and Daniel Nathans to use HindII in the physical mapping of simian virus 40 DNA [8], a process known as restriction enzyme mapping.

For their pioneering work with restriction enzymes, Daniel Nathans, Hamilton Smith, and Werner Arber were awarded the 1978 Nobel Prize in Physiology or Medicine. With the discovery of DNA ligase, in combination with the growing family of site-specific cutting restriction enzymes, recombinant DNA technology was born.

Restriction enzyme naming convention

The naming convention takes into account three characteristics of the enzyme’s organismal origin—genus, species, and strain or serotype—to develop a shortened name followed by roman numerals to represent multiple restriction enzymes from the same strain [9]. For example, the enzyme HindIII (or Hind III in earlier nomenclature) represents:

  • “H” for Haemophilus
  • “in” for influenzae
  • “d” for serotype d
  • “III” to distinguish from other restriction enzymes from Haemophilus influenza serotype d (e.g., HindII vs. HindIII)

Restriction enzyme classification

Restriction enzymes are categorized into four classes, based on their structural complexity, recognition sequence, cleavage site position, and cofactor requirement. Table 1 summarizes the distinguishing characteristics of these classes.

Table 1. Restriction enzyme classes and characteristics.

Enzyme class Characteristics
Type I
  • Multi-subunit protein with both restriction and methylation activities
  • ATP requirement
  • Cleavage site a variable distance from recognition site
Type II
  • Specific recognition sequence
  • Cleavage site within or close to recognition sequence
  • Generates 5′ phosphate and 3′ hydroxyl termini at cleavage site
  • Mg2+ requirement for most
Type III
  • Two-part recognition sequence in inverse orientation
  • Cleavage site a specific distance away from one of the recognition sequences
  • ATP requirement
Type IV
  • Cleavage of only methylated DNA
  • Cleavage site approximately 30 base pairs away from recognition site

Because of the specific characteristics of Type II restriction enzymes, these have become the most commonly used in many research applications such as cloning and forensic DNA analysis. The specific cutting pattern of these enzymes led to their use in restriction fragment length polymorphism (RFLP) analysis, which is a basis of forensic studies. Ligase, which enzymatically joins 5′ phosphates and 3′ hydroxyls at DNA termini, enables rearrangement and connecting of DNA molecules with 5′-phosphate and 3′-hydroxyl termini generated by restriction enzymes—the fundamental principle of recombinant DNA cloning technology.

Due to their usefulness in molecular biology research, the Type II restriction enzymes are the most studied class of enzymes and comprise the largest group. Over 3,500 Type II restriction enzymes have been characterized and subcategorized further into groups such as Type IIP, IIA, IIB, IIC, IIS, etc., where Type IIP enzymes, which recognize palindromic (symmetric) target sequences, are the most prevalent among commercially available restriction enzymes [10].

Recognition site and cleavage specificity

Another important way to classify and compare restriction enzymes is isoschizomers and neoschizomers.

  • Isoschizomers are restriction enzymes that have the same recognition sequence and the same specificity. For instance, AgeI and BshT1 recognize and cleave 5′-A↓CCGGT-3′ in the same pattern. Nevertheless, a set of isoschizomers may differ in site preferences, reaction conditions, methylation sensitivity, and star activity.
  • Neoschizomers recognize the same nucleotide sequence but cleave DNA at different positions. Examples of neoschizomers are SmaI (5′-CCC↓GGG-3′) and XmaI (5′-C↓CCGGG-3′), which both recognize 5′-CCCGGG-3′ but cleave them differently and thus generate different types of ends (in this case, blunt ends for SmaI and 5′ protruding ends for XmaI).

The availability of different specificities in both recognition sequence and cleavage pattern has made restriction enzymes an extremely flexible and powerful set of tools for characterizing and manipulating genetic material.


Related products