New technology provides near-atomic level resolution for viewing structures

HOUSTON -- (March 12, 2008) -- The combination of a powerful cryo-electron microscope, super-fast computers and sophisticated new computer algorithms enables researchers at Baylor College of Medicine in Houston to determine the structure of proteins, viruses and particles at a near-atomic level resolution.

In a report in the current issue of the journal Structure, Dr. Steven Ludtke, Dr. Wah Chiu and others in the National Center for Macromolecular Imaging at BCM, along with researchers from The University of Texas Southwestern Medical School in Dallas, describe the structure of a protein called GroEL, which chaperones or helps misfolded protein molecules fold into their correct, native shape within the cell.

Previously, the researchers had determined the structure of GroEL at 11.5 Angstroms (Å) and then at 6.5 Å resolution. The current structure at approximately 4 Å resolution helped researchers better understand the biology of the particle, said Ludtke, associate professor of biochemistry and molecular biology and co-director for the National Center for Macromolecular Imaging at BCM. (An Angstrom is one ten-billionth of a meter.)

At lower resolutions, Ludtke and his colleagues, including Chiu, professor of biochemistry and molecular biology at BCM and director of the National Center for Macromolecular Imaging, found that isolated in solution, the top and bottom rings of GroEL were identical. However, this new, higher resolution structure shows subtle, but biologically important differences between the two rings, said Ludtke. It was previously known that GroEL's symmetry was lost when it bound to its partner molecule, GroES, but now they can see differences between the rings even without GroES.

"The fundamental advance in this study is the resolution," said Ludtke. The group has also recently produced an approximately 4 Å-resolution structure of a virus called epsilon15. However, GroEL is more than 100-fold less massive, and has lower symmetry, making high resolution refinement much more challenging.

In electron cryomicroscopy, to preserve the natural state of the structure as much as possible, researchers embed the molecules in a thin layer of water and freeze it rapidly using liquid ethane. The material cools so rapidly that crystals do not have time to form. Rather, it forms a glassy state known as vitreous ice, which preserves the molecules in a snapshot of their conformation as it would appear in solution. Tens to hundreds of thousands of images of particles in different orientations are then recorded, and processed using computer clusters with hundreds of processors to produce the final three-dimensional reconstructions.

This single-particle analysis technique is particularly valuable in analyzing larger molecules and protein assemblies or other difficult candidates for traditional X-ray crystallographic methods.

Faster computers compress the time needed to process information from the microscope, he said. Solving the structure of GroEL in this study took about 100,000 processor-hours on the latest generation of computers. With the computers available a decade ago, this wouldn't have been feasible. New computer algorithms developed by Ludtke and members of his group and improvements in electron microscope technology were also fundamental in achieving this landmark resolution.

"This is the first time we've reached a resolution where this technique could be used to trace the protein backbone of the structure," said Ludtke. Before, he said, they had to use crystal structures or information obtained from crystal structures to manage this feat. In this work, they did not have to rely on X-ray crystallography (the most common technique for determining protein structures) at all.

"We fully expect that we will extend the resolution even further n the not-too-distant future," said Ludtke. He also hopes to be able to determine not only the static structure of molecules, but also the dynamic movements these molecules undergo in solution as part of their function within the cell. "Dynamics is the next major frontier in single particle analysis," he said.

Others who took part in this work include Matthew L. Baker and Dong-Hua Chen of BCM and Jiu-Li Song and David T. Chuang of UT Southwestern.

Funding this work came from the National Institutes of Health Roadmap for Medical Research, the National Center for Research Resources, the National Institute of Diabetes, Digestive and Kidney Diseases, the National Institute of General Medical Science, the National Science Foundation and the Robert Welch Foundation.

The full article is available at www.structure.org.