Restriction Enzymes: Molecular Scissors

Cutting DNA at a precise, predictable location is the first practical requirement of genetic engineering, and it is solved by a class of bacterial enzymes called restriction endonucleases, or restriction enzymes.

Restriction enzymes were discovered as part of a bacterial defence system. Many bacteria are attacked by bacteriophages that inject their own DNA into the bacterial cell to hijack it. To defend against this, some bacteria produce enzymes that recognise and cut up any 'foreign' DNA that enters the cell, while protecting their own DNA from the same enzymes by chemically modifying (methylating) it at the very sites the enzyme would otherwise cut. In effect, the bacterium restricts the growth of the invading phage by destroying its DNA -- which is why these enzymes are called restriction enzymes, and why the whole defence system is sometimes called a restriction-modification system.

The property that makes restriction enzymes so useful in the laboratory is their sequence specificity. Each restriction enzyme recognises one particular, short DNA sequence -- typically four to eight base pairs long -- and cuts the DNA only at that exact sequence, nowhere else. These recognition sequences are almost always palindromic, meaning the sequence reads the same on both strands when each is read in the 5' to 3' direction. For example, EcoRI recognises the sequence GAATTC on one strand; reading the complementary strand in its own 5' to 3' direction gives the identical sequence GAATTC. This palindromic symmetry is what allows the enzyme, which usually works as a pair of identical protein subunits, to bind the DNA symmetrically and cut both strands in a coordinated way. …