Nature Structural Biology pp 126 - 130

Restriction enzymes are proteins that are made in bacteria and cut DNA at specific sites. These enzymes can be isolated from bacterial cells and used in the laboratory to shuffle pieces of DNA around. This allows researchers to isolate pieces of DNA that contain genes and combine them with other DNA molecules, that is, to clone genes.

More than 3,000 type II restriction enzymes (there are three classes in all) have been identified to date. Each recognizes a particular DNA sequence that is four to eight base pairs in length. These enzymes cut the DNA by catalyzing the hydrolysis (the splitting of a bond by the addition of water) of DNA.

How these enzymes specifically recognize and cut their DNA targets has been a long-standing question in biology. By comparing the three dimensional X-ray crystal structures of these enzymes alone and bound to DNA one can get some idea of what needs to happen in order for cutting of the DNA to occur.

Now, Aneel Aggarwal and coworkers at The Mount Sinai School of Medicine in New York City, have shown that one restriction enzyme, BglII, opens by a 'scissor-like' motion to capture the DNA and encircle it. This kind of dramatic motion has not been observed for other restriction enzymes and so expands our understanding of how proteins specifically recognize DNA.

Nature Structural Biology pp 121 - 125

People often think of DNA as a fixed repository of genetic information. In some sense, that is true. But that does not mean that DNA is never altered inside the cell. On the contrary, there are many enzymes that act upon DNA—for example, to duplicate (or replicate) it during cell division, or to modify it in some way.

Modifications of the A, G, C, and T DNA bases occur often and can sometimes be very useful—for example, specific modifications can help to indicate to repair proteins how to fix mistakes generated by the replication machinery. To understand how DNA is maintained faithfully as the genetic code, it is important to understand exactly how such enzymes carry out their reactions.

Enzymes called 'DNA methyltransferases' (MTases, for short) are one such set of proteins that act on DNA. They take a methyl group (made up of a single carbon attached to three hydrogen atoms) from a certain chemical substrate and attach it to a particular position on a DNA base.

One predicament that DNA MTases must overcome is that the DNA base to be methylated is usually tightly positioned in the DNA helix, making it difficult to imagine how the enzyme, upon binding to the DNA, could maneuver the base into the active site where the methylation reaction takes place.

Previous work a class of MTases that modify cytosine (C) bases showed one dramatic way to overcome this problem. The target C base is 'flipped out' of the helix and positioned close to the active site of the enzyme—the cytosine is rotated to expose the proper position for methylation. The rest of the DNA surrounding the target base is not greatly altered.

Now, Elmar Weinhold, of the Institut f�r Organische Chemie der RWTH Aachen in Germany, and his colleagues have solved the structure of another kind of MTase, one that modifies adenine (A) bases. This work shows that the adenine base is also rotated out of the helix in this complex, suggesting that 'base flipping' is a universal mechanism used by MTases.

Robert Blumenthal and Xiaodong Cheng discuss these results in an associated News and Views report.