Computer-Aided Protein Design Wins AAAS Newcomb Cleveland Prize

A report on the process whereby a computer-designed protein was synthesized and confirmed to match the original plan earned the coveted 2003-2004 AAAS Newcomb Cleveland Prize.

The prize—the oldest award conferred by AAAS—was established in 1923 with funds donated by Newcomb Cleveland of New York City to recognize outstanding Science articles. The award currently is supported by Affymetrix of Santa Clara, Calif.

Architects turned long ago to computer-aided design, and scientists have now followed suit to design and build a never-before-seen protein, according to the prize-winning article that was published in the 21 November 2003 issue of Science. The finding will allow "exploration of the large regions of the protein universe not yet observed in nature," say Brian Kuhlman and colleagues.

The AAAS Newcomb Cleveland Prize was awarded to researchers from Howard Hughes Medical Institute in Seattle, Washington; the University of Washington in Seattle; the University of North Carolina at Chapel Hill; the Infectious Disease Research Institute in Seattle; and the Fred Hutchinson Cancer Research Center, also in Seattle.

"The paper presents a remarkable accomplishment," said Science Editor-in-Chief Donald Kennedy. "In a work that will be of interest to a number of disciplines, the authors applied a novel technique of iterative computational design to a problem that many had seen as insoluble: the design—with atomic-level accuracy—of very large molecules with molecular masses greater than 10,000. This opens up new possibilities for studying the channeling problem of protein-folding energetics. It will now be possible to design proteins for therapy or for molecular machines that are not limited to the structures that have been generated by the evolutionary process."

The "tertiary" folded structure of proteins is vitally important in biochemistry and cell biology; this configuration may determine the active site of enzymes, the shape of globular proteins in blood or connective tissue, and the higher-order structure of membrane receptors.

Thus, it would be desirable to be able to choose a particular three-dimensional structure, specify the amino acid sequences appropriate for generating it, determine the predicted result by computation, and finally, show that the protein has the correct topology at the atomic level by x-ray crystallography. That is the achievement reported in this path-breaking report, authored by Kuhlman, Gautam Dantas, Gregory C. Ireton, Gabriele Varani, Barry L. Stoddard and David Baker.

The team took the opposite approach from many in the protein-design field, who try to look at a string of amino acids and predict the appearance of the folded protein. Instead, Kuhlman's group designed a large, complex protein on the computer, synthesized it, and then confirmed that its three-dimensional structure closely matched their original plan.

Their automated design procedure cycles back and forth between assembling the linear protein chain and predicting the three-dimensional structure that would arise from it.

Throughout the year, Science readers may nominate papers appearing in the journal's research articles, reports, or reviews sections.

— Jessica Lawrence-Hurt

16 February 2005