QUALITY CONTROL In its search for damaged guanines, the repair enzyme hOGG1 (blue) flips undamaged guanine (red) out of the DNA helix (orange) to examine it.
	COURTESY OF ERIC SMITH/DANA FARBER CANCER INSTITUTE

"This structure provides a glimpse at how a DNA repair enzyme interrogates and rejects normal bases that are similar to their damaged counterparts," comments biochemist Sheila S. David of the University of Utah, Salt Lake City.

An army of DNA repair enzymes is charged with the huge task of searching vast stretches of genomic DNA for damaged bases. When a repair enzyme finds the specific damaged base it's looking for, the enzyme flips the damaged base out of the DNA helix and into its active site. There, the aberrant base is clipped from the DNA backbone. How these enzymes manage to avoid removing normal bases, which in many cases differ only subtly from their damaged counterparts, has remained unclear.

To shed light on that question, Harvard University crystallographers Gregory L. Verdine and Anirban Banerjee have used a clever strategy to capture a picture of human 8-oxoguanine glycosylase (hOGG1)--the enzyme responsible for finding and removing the mutagenic oxidized base 8-oxoguanine from DNA--while it searches a stretch of undamaged DNA (Nature 2005, 434, 612).

Verdine's lab previously published a structure of inactivated hOGG1 bound to DNA containing an 8-oxoguanine, which differs from guanine by only two atoms. That structure showed that hOGG1 extrudes the damaged base out of the DNA helix and inserts it into a deep active site where it can be clipped from the DNA. The new structure reveals that hOGG1 flips undamaged guanine out of the DNA helix, too. But the undamaged base is denied entrance to the active site and instead gets stuck in a nearby binding pocket before being returned to the DNA helix.

That secondary binding site, Verdine suggests, acts as a gatekeeper to the active site, thereby ensuring that 8-oxoguanine is removed but undamaged guanine is not. In Verdine's previous structure, the only obvious contribution to discrimination between 8-oxoguanine and guanine was a single hydrogen bond in the active site. Now, calculations performed by Verdine's Harvard colleagues Martin Karplus and Wei Yang indicate that the mechanism for discrimination is far more refined: Both favorable and unfavorable interactions lead to preferential binding of guanine over 8-oxoguanine in the secondary binding site and of 8-oxoguanine over guanine in the active site.

James T. Stivers of Johns Hopkins University School of Medicine says that the new structure represents "a late-stage intermediate in the still-mysterious pathway [that] repair enzymes use to flip bases out of the DNA helix." Stivers has used stopped-flow fluorescence and solution and solid-state NMR techniques to detect early-stage intermediates in the process by which a related DNA repair enzyme hunts for its target. He is optimistic that the combination of such methods and Verdine's trapping technique may soon yield a complete picture of how base flipping proceeds