HHMI investigator John Kappler is part of a group of biologists that is mastering the art of growing protein crystals, and using the techniques of x-ray crystallography to visualize the structure of proteins.
Photo: William K. Geiger

An HHMI initiative places a powerful research technique in the hands of more scientists.

By Gary Taubes

In the early 1990s, John Kappler, an HHMI investigator and immunologist at the National Jewish Medical and Research Center in Denver, found himself complaining to a fellow molecular biologist about the state of his research. He and his colleagues had been studying various proteins involved in the immunologic actions of T cells, and they had done much of the basic biology. Then they had turned to the problem of working out the actual function of the proteins.

"We slogged away," he said, "making antibodies, purifying proteins and doing genetics, only to have the final answer of the function solved by x-ray crystallographers, who then basically owned the protein and 20 years of research and could reinterpret all our research based on the molecular structure they had in front of them." As Kappler remembers, this colleague, who also happened to be one of the most renowned x-ray crystallographers in the country, said, "Stop complaining and do it yourself!" And so he did.

Now Kappler is one of the converted, a biologist who is mastering the black magic of coaxing large biological molecules to form crystals that can reveal the precise atomic structures of those molecules. Aiding that conversion—some would say dramatically—is a multimillion-dollar device known as a synchrotron, a machine born of physicists' quest to understand the fundamental nature of the universe. The devices, which are the size of football stadiums, emit x-rays as a side effect of accelerating electrons to the high energies that physicists needed for their experiments into the nature of matter. Now biologists have harnessed the synchrotron to explore the inner workings of gigantic biomolecules. The dramatic importance of synchrotrons for x-ray crystallography has been demonstrated repeatedly over the past decade.

"A few years ago the fraction of all reports of protein structures that were solved with synchrotron radiation was around 10 percent. Now that number is closer to 50 percent. And if you look at the structures published in prominent journals like Cell or Nature, it's on the order of 80 percent," says HHMI investigator Wayne Hendrickson, a Columbia University crystallographer.

As biologists sequence one genome after another, and the number of biologically intriguing proteins increases apace, the simple act of getting what's known as "beam time" on a synchrotron has become the bottleneck in the whole process. At the Advanced Light Source (ALS) at Lawrence Berkeley National Laboratory, for example, time for crystallography on the synchrotron is dramatically oversubscribed. "There is 3.5 times more beam time requested than we can currently accommodate, and that number will only go up," says Graham Fleming of the Lawrence Berkeley National Laboratory's physical biosciences division, which manages the ALS protein crystallography program. It's a fact of life that biologists often must ask for synchrotron beam time a year in advance to get the two days they need to unravel a structure.

That bottleneck may ease in the near future. The past decade has witnessed a boom in synchrotron facilities and beamlines that will be dedicated to x-ray crystallography. At the ALS, for example, HHMI is funding two new synchrotron beamlines for use by Hughes investigators and others. These will join an HHMI beamline at the Brookhaven National Laboratory on Long Island that has already proven fantastically productive. "These beamlines will make it possible to understand ever larger and more complicated structures and to understand simpler structures very much more quickly," says Hendrickson.

MHC molecules present antigenic peptides to T cells...
Image: Laboratory of John Kappler

Image: Laboratory of John Kappler

Reprinted from the July 2000 HHMI Bulletin, Vol. 13, No. 1, pp. 12-17.

continued...