Proteins that copy and edit DNA help protect cells from attack by removing viral DNA insertions. They are thought to reach their target sequence by "sliding", "hopping" and "jumping" along DNA strands. Observing this behavior has proved hard, however, due to its speed of action. We discuss a recent research project that used a high-speed camera to observe - for the first time - the sliding and jumping of EcoRV, a restriction enzyme, along DNA molecules.

Understanding how cells maintain their integrity and protect themselves from viral attack by removing invasive DNA has motivated much research over the last thirty years [e.g. 2-4]. Researchers have been particularly interested in discovering how proteins, such as type II restriction enzymes, can rapidly find the short sequences of viral DNA they must cut. Such studies are also useful to researchers using restriction enzymes for genetic engineering.

DNA Hopscotch

The mechanism that these enzymes use to find their DNA target site is interesting because target location happens faster than is possible through chance interactions alone. It is now widely accepted that these enzymes move along nonspecific DNA to the specific target site but, until recently, how they move was a matter for debate. The main process was thought to be ‘sliding' along the DNA double strand [5]. This is a linear movement where the enzyme searches in one-dimension - along the strand - for the target site. There are several studies that showed that certain proteins can slide along DNA [e.g. 6]. Sliding can be faster than locating the specific site through random collisions between the DNA and enzyme, which involves a three-dimensional search as they float in solution.
More recent research has suggested that some enzymes can "jump", as well as "slide", along DNA strands to find their target [7]. During a "jump", the enzyme leaves the DNA strand, moves in three dimensions and then returns to the same DNA strand. Jumping is a faster mechanism for finding targets than sliding over long distances [8].

Too Fast to Catch

Single enzymes have been observed sliding, but it has proved hard to observe 3D jumping because this process is orders of magnitude faster than sliding [1].

A jump can be completed in less than milliseconds. How important sliding is compared to jumping was therefore uncertain, in particular for EcoRV - a type II restriction enzyme that is commonly studied because it is easy to obtain and purify. EcoRV also probably acts in a similar way to many other enzymes, making it a good model for their behavior. Some studies carried out on EcoRV have supported sliding as being the main mechanism for locating the target site. For example, the way EcoRV interacts with non-target DNA, revealed by the crystallographic structure of the EcoRV-DNA complex, made it likely that there will be linear movement along the strands [9]. Other studies, such as those involving DNA with two target sites close together, suggested jumping was the most important mechanism [10].

Catching Enzymes in the Act

The mystery of how EcoRV moves along DNA was tackled by directly observing single molecules tagged with a fluorescent label using Total Internal Reflection Fluorescence Microscopy (TIRFM). The DNA molecules used were T7 bacteriophage DNA, which does not have target sites for EcoRV. This meant the experiment could focus entirely on observing EcoRV movement rather than cutting of the DNA.
The fluorescent light captured by the microscope was detected using a very fast, sensitive camera (Andor iXonEM+ 860 back-illuminated EMCCD). Speed and sensitivity are often mutually exclusive in a camera because, to gather enough light from fluorescing molecules for single molecule detection, the exposure time needs to be long (more than a few tens of milliseconds). This Electron-Multiplying CCD camera is one of only a few in the world sensitive enough to image individual fluorescent Cy3 labels at speeds of more than 500 frames per second. The Andor camera made it possible to measure the length and frequency of jumps made by individual EcoRV molecules.
The experimental apparatus consisted of an inverted microscope with a 60X oil-immersion objective lens, the Andor camera and a simple flow cell. A cover slip coated with streptavidin was attached to a microscope slide in which inlet and outlet holes had been drilled to form a simple flow cell. DNA molecules were extended and bound to the surface of the slide by their biotinylated ends.
The flow cell was placed on the inverted microscope and the fluorescent enzymes were excited with a 532nm laser in a TIRFM configuration. Fluorescent light was collected with a dichroic mirror together with a long-pass filter and detected by the EMCCD camera.