The UCLA scientists spent several frustrating years pursuing initially promising technologies that ultimately led nowhere. "We tried a whole bunch of tools using traditional x-ray methods as well as other methods," says Eisenberg, who is also director of the UCLA-DOE Institute for Genomics and Proteomics. "But when you have to shoot a bunch of crystals rather than a single one, not only are the data degraded, but there's too much data, which adds background noise. So we weren't getting a clear enough picture."

All that began to change in July 2003, when Eisenberg attended a conference in Crete. Over lunch with Swedish scientist Carl-Ivar Brändén, he spoke of his dilemma. "There's one person in the world who can help you," Brändén immediately responded, "Christian Riekel."

A crystallographer at the European Synchrotron Radiation Facility in Grenoble, France, Riekel had invented a highly focused x-ray camera capable of analyzing crystals as small as one micrometer in diameter—about 1/100th the width of a human hair—a technology 100 times more powerful than the best available in the United States. Another large plus was that a graduate student at the lab, Anders Madsen, had developed an effective way of manipulating microcrystals.

The UCLA team took their microcrystals to Riekel's lab, where they recorded accurate x-ray diffraction patterns from an individual microcrystal. With this diffraction pattern in hand, the researchers looked for the position of the zinc atom in the structure, knowing that microcrystals grow only in the presence of zinc. "When I found the zinc atom, finding the others was easy," says Michael R. Sawaya, a research scientist at UCLA who contributed to the project. With the knowledge of the position of all the atoms, you can see the structure of the molecule. "Recognizing the features of the peptide was like seeing the familiar face of an old friend," adds Sawaya.

There was much dancing around the lab when, after 8 years of hard work, the three-dimensional structure was known. And the structure suggested to the researchers how the fibrils accumulate in brain tissue and why they are so infectious. At last they had all the puzzle pieces. The peptide molecules in the microcrystal assemble into a structure that resembles a tightly bound zipper that latches on to an identical structure, much in the manner of Velcro, which explains their great stability. This suggests how the amyloid fibrils keep growing: They latch on to either end of the structure, forming an almost indestructible spine-like chain. "This discovery is the pinnacle of years of hard work," says UCSF's Safar, "and is a significant contribution in our understanding at the atomic level of the way amyloids form."

The UCLA team hopes that unraveling the atomic structure will suggest ways to cap the growth of amyloids and pave the way toward formulating treatments that can inhibit this process.

	PAGE 2 OF 2	Back