BioWorld International Correspondent

LONDON - New insights into the molecular mechanisms by which the anticancer drug cisplatin operates could one day lead to equally effective drugs with fewer side effects.

A team of researchers working in Germany have shown that the cell's DNA repair mechanisms respond differently to the DNA damage caused by cisplatin than to that caused naturally - as a result of exposure to sunlight, for example.

Patrick Cramer, director of Gene Center, at the Ludwig-Maximilians-University in Munich, Germany, said, "Cisplatin is one of the most widely used anticancer agents, and it works, but nobody really knows why. We are now beginning to elucidate the atomic details of its mechanism of action."

Cramer and his colleagues reported their findings in a paper in the Nov. 4, 2007, issue of Nature Structural and Molecular Biology titled "Mechanism of transcriptional stalling at cisplatin-damaged DNA."

Cells are continually repairing damage to DNA caused by, for example, ultraviolet light or oxidative stress. The enzyme RNA polymerase II, which is the same enzyme that transcribes DNA and produces messenger RNA, "scans" the genome for abnormalities. When it finds such a lesion, it "stalls" and triggers repair of the DNA. That process, which is very fast and efficient, is known as transcription-coupled repair.

Cisplatin is known to work by introducing lesions into the DNA, thus interfering with transcription and DNA replication.

"We think that normal cells," Cramer said, "because they are proliferating only slowly, survive long enough to remove the lesions caused by cisplatin, while cancer cells, because they divide very rapidly, do not have enough time to remove the lesions, and therefore they die."

He and his colleagues set out to characterize the first step in the removal of a cisplatin-induced lesion in the DNA, when RNA polymerase II reaches such a site.

"We found that the polymerase does stall at these cisplatin-induced lesions, as you would expect, but it stalls using a completely different mechanism than when it stalls at a natural lesion, such as one induced by ultraviolet light," Cramer said. "Both types of lesions are removed by transcription-coupled repair, but the mechanism of stalling the polymerase and triggering repair is very different when you look at the chemical details."

In some circumstances, the researchers found, it was possible for RNA polymerase II to bypass a cisplatin-induced lesion completely.

"This is an important finding," Cramer noted, "because people working on developing new drugs for cancer treatment dream of being able to design a cisplatin type of drug that would be less visible to RNA polymerase II. If you had such a drug, then it would be possible to lower the dose of the drug and reduce the side effects associated with it, but the cell would not repair the drug-induced lesion."

Cisplatin can damage DNA in different ways, and the study reported in Nature Structural and Molecular Biology examined the most common type of DNA damage it causes, which involves cross-linking adjacent guanosines in a DNA strand. About one-third of lesions caused by cisplatin, however, involve cross-linking two guanosines that are separated by a third nucleotide.

Cramer said: "Our future studies will involve looking at this other type of cisplatin-induced lesion. These types of studies will give more clues to why cisplatin works, and how it works; and maybe once we have this information, we will have more ideas about which types of lesion are more or less visible to the cell's DNA repair machinery, and therefore how to design new cisplatin-type compounds that induce these particular types of damage."

He emphasized, however, that the goal is not to design a drug that produces DNA lesions that are not visible to the cell's DNA repair machinery at all. "This would be highly toxic and not a drug at all, because the cell would not carry out any repair. What we need to do is to lower the frequency of recognition so that we can use a lower dose," Cramer said.