Contact

Duke Health News

919-660-1306

DURHAM, N.C. -- Researchers at Duke University Medical Center have solved the three-dimensional structure of an enzyme crucial to the process that turns some normal cells cancerous, and have determined how it works on a molecular level.

The protein enzyme, called farnesyltransferase (FTase), is being intensely studied by university scientists and pharmaceutical company researchers because it appears to be a true molecular Achilles' heel in cancer cells. The FTase structure shows the key regions in FTase that are crucial for its function, and identifies target sites that are important for designing drugs to stop the development of cancerous cells.

Enzymes are the molecular workhorses of the cell, fostering the myriad of chemical reactions necessary for the cell to live and reproduce.

The FTase enzyme structure, published in the March 21 issue of the journal Science, could have important clinical implications, said biochemist Lorena Beese, who led the research team that determined the crystal structure.

"We hope that by understanding how this relatively newly discovered class of proteins works on a detailed, molecular level, we will gain a foothold into understanding the signaling process and, importantly, launch an entirely new class of anticancer drugs," Beese said. According to Beese, understanding the enzyme could allow the first new class of cancer-fighting compounds in almost 40 years.

The research was a collaborative effort among Beese, an X-ray crystallographer; Pat Casey, a protein biochemist; postdoctoral fellow Hee-Won Park and technicians Sobha Boduluri and John Moomaw. Beese and Casey are investigators in Duke's Comprehensive Cancer Center.

The research was funded by the National Institutes of Health, the Searle Scholar Foundation, Schering-Plough Research Institute and the Council for Tobacco Research.

FTase is one of the larger proteins whose molecular structure has been determined to high resolution, taking nearly four years to complete. It is also the first protein structure to be solved in its class of enzymes -- those that add fatty prenyl groups to other proteins to activate them.

In 1989, Casey and his collaborators, along with scientists at other institutions, demonstrated the important role of FTase in activating a critical cancer-causing molecular switch in the cell -- the enzyme Ras.

FTase adds the fat molecule "farnesyl" to Ras and other proteins in the cell. Casey showed that this addition step is necessary to activate the cancer-causing form of Ras. When the scientists blocked FTase activity, Ras could no longer form tumors. Shortly thereafter, the search for drugs to block FTase began.

Knowledge of FTase's molecular structure is invaluable, the Duke researchers say, because, although several drugs that block the FTase protein are showing promise against cancer in animals, they may have side effects when tested in people.

Information about the protein's structure will allow researchers to narrow down which FTase blockers are likely to stop cancer cells without causing unwanted side effects, Casey said. Such knowledge could help speed anticancer drugs from this entirely new class of chemotherapy agents to market.

The scientists determined the FTase protein's structure using X-ray crystallography. The technique works by shining X-rays through a crystallized form of the protein and deducing the molecule's structure from the pattern of diffracted spots that results. The Duke researchers used the technique to identify the location of the more than 7,000 atoms that comprise the FTase enzyme.

"The main question we wanted to answer was, where is the active site?" said Beese. "By looking at the structure we identified two clefts that intersect at a zinc ion, which is required for activity."

One cleft forms a snug pocket where the Ras protein docks with the enzyme, Beese said. The other cleft has an "oily" lining that repels water but which is a perfect fit for the fatty farnesyl molecule. The location of the two pockets puts the Ras and the farnesyl in just the right docking position to link the two together.

By attaching the fatty farnesyl group, the FTase flags the Ras enzyme for transport to the cell's outer membrane, where it goes to work transmitting outside signals from hormones and growth factors into the cell to tell it to divide.

Normally, the Ras protein signals the cell to divide only in response to such external growth factors and then shuts off.

But a mutant cancer-causing Ras never turns off, continuously telling the cell to divide over and over again to produce cancerous growth.

This uncontrolled signal from Ras has been shown to be involved in a quarter of all cancers in people. Defective Ras enzymes cause up to 90 percent of pancreatic cancers, half of all colon cancers, and a quarter of all lung cancers, some of the most difficult cancers to cure.

"Shutting down Ras has been a major focus of cancer research for over a decade," Casey said. "Now we are coming close to success, and this crystal structure of FTase should help speed things even further."

Animal studies of FTase inhibitors have shown their potential as anticancer drugs. In the August 1995 issue of Nature Medicine, scientists at Merck Co. showed that one such drug caused breast tumors in mice to shrink, and in some cases, disappear completely, without causing any detectable side effects.

"There are at least 50 issued patents and many more pending on inhibitors to this FTase protein," said Casey. "These inhibitors fall into several distinct classes of compounds. But it is unclear how any of these will fare as chemotherapy agents in the clinic. As least six pharmaceutical companies have active programs in this area. And taking any one of these drugs to clinical trials is very expensive. That is why this structural information is so vital. Now that the structure is known, compounds can be refined and tested to a much greater degree before they are sent to clinical trials in patients."

While no pharmaceutical companies have begun testing FTase inhibitors against cancers in people, such tests could begin as soon as late 1997, Casey said.

Beese and Casey are now determining the crystal structure of the FTase protein with some of the drug inhibitors attached.

"The protein may change shape when an inhibitor is attached," Beese said. "We need to determine the crystal structure with inhibitor bound to have confidence that our molecular models are correct."