Light Shed on Photolyases

DNA photolyase enzymes harness light energy to repair DNA damaged by ultraviolet radiation. The first X-ray structure of one of these enzymes bound to its damaged DNA substrate was recently reported by German crystallographers. In addition to confirming many suspicions about how these enzymes recognize and repair light-induced DNA damage, this long-awaited structure paints a picture of the enzyme midway through its work.

Photolyases are thought to have been one of nature's earliest solutions to the dangers caused by life under the sun. The sun's UV rays play havoc with genomic DNA, causing a variety of toxic DNA lesions. The most common of these is the cis-syn cyclobutane pyrimidine dimer (CPD), formed by [2 + 2] cycloaddition of two adjacent pyrimidine bases (usually a pair of thymines.) Such CPD lesions bring polymerases to a standstill, eventually leading to cell death.

Photolyases--which are found in modern-day prokaryotes; plants; and a variety of animals including frogs, fish, and snakes--use blue or near-UV sunlight to drive the cleavage of CPD lesions' cyclobutane ring. (Humans use another set of enzymes to repair CPD lesions.) Decades' worth of biochemical data, computer modeling, and structural studies have been invested in figuring out how these enzymes carry out CPD repair. But despite many attempts, no one had been able to obtain a structural picture of a photolyase bound to its damaged DNA substrate. A German team led by chemistry professors Thomas Carell of Ludwig Maximilians University, in Munich, and Lars-Oliver Essen of Philipps University, in Marburg, has now provided this long-awaited picture [Science, 306, 1789 (2004)].

"This structure confirms predictions that the damaged dinucleotide would be flipped out of the DNA stack and into the enzyme's active site," comments biochemist Aziz Sancar of the University of North Carolina School of Medicine, Chapel Hill.

"We knew that repair proteins can flip a single damaged base out of the helix for repair," Carell says. "But can a repair enzyme flip two bases out of the helix?" No one had been able to prove that this could happen, he says. "Our DNA photolyase structure tells us that this is indeed possible."

THE STRUCTURE also "reveals the extensive network of interactions the damaged dinucleotide makes with the enzyme's active site," Sancar says. When combined with the wealth of biochemical data that's been gathered for the photolyases, these interactions explain how these enzymes recognize CPD lesions and hint at how they catalyze the repair of such lesions, he adds.

Previous work has suggested that photolyases' so-called antenna pigment (deazaflavin or methenyltetrahydrofolate) absorbs a photon of the appropriate wavelength and passes the energy to the enzyme's catalytic flavin cofactor. It's thought that this energy drives the transfer of an electron from the activated flavin cofactor to the CPD lesion, splitting its cyclobutane ring. The German team's 1.8-Å structure shows that the enzyme flips the CPD lesion into an active-site cavity right next to the catalytic flavin cofactor. This proximity suggests that the activated flavin passes its electron directly to the lesion, Carell says, adding that his group has unpublished calculations that strongly support direct electron transfer. He and Essen are now working to further clarify the electron transfer pathway.

Perhaps the most surprising aspect of the structure is the fact that the initially intact cyclobutane dimer is split during X-ray structure determination. Although it's well known that synchrotron radiation can produce solvated electrons in protein and DNA crystals, previous crystal structures of CPD lesions have captured the intact cyclobutane ring, suggesting that synchrotron-generated electrons aren't themselves capable of repairing CPD lesions. But in the presence of photolyase, "the enzyme binds the dimer lesion in a way that makes it more vulnerable to cleavage" by a synchrotron-generated electron, Carell says. He suggests that his team's structure mimics an intermediate during light-driven DNA repair, where the cleaved thymines have not yet been flipped back into the DNA helix.

So why did this team succeed where others failed? Carell speculates that it's because they used a simple, more easily synthesized analog of the CPD lesion. The analog retains the stereochemistry of the real thing but replaces the real lesion's intradimer phosphate linkage with a more sturdy formacetal linkage. This small change allowed production of the pure DNA lesion in large quantities, he says. "When it comes to repair research, making the DNA lesion is always the limiting step."