Supercomputing Institute Research Bulletin Fall 1997

Unraveling the Properties of Protein Structures

The hexameric helicase from the bacteriophage T7 (yellow ring) has been reconstructed in three dimensions from electorn micrographs. The model shows how this ring binds to DNA, with one strand running through the central channel and the other strand displaced outside of the ring. Large-scale sequence analysis of bacterial, yeast and human genomes is generating incredibly vast amounts of data. For example, we now know the entire genetic code for organisms such as bacteria and yeast. But the actual machines that make living cells function are, in most cases, proteins. The collection of genetic data far outpaces our ability to analyze in the laboratory the properties of the many proteins that are encoded by this DNA. Beneath the complexity of the millions of sequences there appears to be a certain degree of conservation, showing that many different proteins evolved from common ancestors. In the case of protein structures there appears to be an even greater degree of underlying simplicity. Structures are more highly conserved across evolution than sequences. As a result, two sequences that have diverged across evolution so strongly as to have no recognizable similarity may both encode the same highly conserved structure. Cell biology and neuroanatomy professor Edward Egelman and his research group are interested in the higher-order structures of proteins, since many of the proteins that exist, whether in bacteria or humans, occur not as monomers but as a part of a higher-order assembly. In particular, they have been interested in two different types of macromolecular complexes: protein-DNA complexes that play a role in DNA recombination and replication, and the helical protein filament formed by actin, which exists in muscle as well as most non-muscle cells. They are finding that higher-order structures, such as helical filaments and hexameric rings, can be even more highly conserved than the structures of the protein subunits that comprise these assemblies. The group’s primary tools are electron microscopy and computed image analysis. Electron microscopy occupies a unique place in structural biology because, under the best of conditions, it has the resolution to solve the atomic structure of folded proteins, as well as the ability to determine the organization of large complexes, such as viruses and helical polymers. In Egelman’s laboratory, Xiong Yu is investigating protein-DNA complexes and Albina Orlova is investigating actin. The computational requirements of this work are large, as the greatest advances in electron microscopy have depended upon computer-based averaging and three-dimensional reconstruction from two-dimensional images. The main platforms that they use are SGI and DEC Alpha workstations. Yu and Egelman’s most recent work has shown that the bacterial RecA protein, which forms a helical filament on DNA during the process of genetic recombination, also forms a hexameric ring that is a structural homolog of ring helicases. Helicases are proteins that use the energy of ATP hydrolysis to open up double-stranded DNA into two single strands. RecA has been studied for more than ten years in Egelman’s laboratory because of its central role in DNA recombination and repair. Remarkably, the group has also been studying helicases for several years without realizing that they are homologs of the RecA protein. The new insights gained from understanding this structural homology will reveal much about the evolution of these different families of proteins, as well as about the central function of helicases and RecA-like proteins in human DNA repair, replication, and cancer.

In This Issue: 1997 Research Scholars LCPC Workshop T3E Upgrade Computing Applications in Neuroscience > Unraveling Protein Structures Silicon Nanocrystals Research Reports