Researchers Design and Build First Artificial Protein | HHMI
Skip to main content
Computers can predict the detailed structure of small proteins nearly as well as experimental methods, at least some of the time, according to new studies by HHMI researchers. The findings provide a glimmer of hope that scientists eventually may be able to determine the structure of proteins from their genomic sequences, a problem that has seemed insurmountable.
Investigator, University of Washington
For media inquires please contact [email protected].
Computers can predict the detailed structure of small proteins nearly as well as experimental methods, at least some of the time, according to new studies by HHMI researchers. The findings provide a glimmer of hope that scientists eventually may be able to determine the structure of proteins from their genomic sequences, a problem that has seemed insurmountable.
Using sophisticated computer algorithms running on standard desktop computers, researchers have designed and constructed a novel functional protein that is not found in nature. The achievement should enable researchers to explore larger questions about how proteins evolved and why nature “chose” certain protein folds over others.
The ability to specify and design artificial proteins also opens the way for researchers to engineer artificial protein enzymes for use as medicines or industrial catalysts, said the study’s lead author, Howard Hughes Medical Institute investigator David Baker at the University of Washington.
Baker and colleagues Brian Kuhlman, who is now at the University of North Carolina, Chapel Hill, and graduate student Gautam Dantas at the University of Washington, published their studies in the November 21, 2003, issue of the journal Science. The scientists collaborated on the studies with other researchers at the University of Washington and the Fred Hutchinson Cancer Research Center in Seattle.
Proteins are initially synthesized as long chains of amino acids and they cannot function properly until they fold into intricate globular structures. Understanding and predicting the rules that govern this complex folding process—involving the folding of the main backbone and the packing of the molecular side chains of the amino acids—is one of the central problems of biology.
According to Baker, the ability to specify a desired folded protein structure and then to create that protein offers powerful scientific and practical benefits. “First, specifying a protein fold and then designing that protein is a very stringent test of our current understanding of the forces and energetics of macromolecular systems,” he said. “Because designing something that’s completely new means you can’t copy any aspect from nature.
“Secondly, if one can design completely new structures, one can potentially design novel molecular machines—proteins for carrying out new functions as therapeutics, catalysts, etc. And finally, there’s the evolutionary question of whether the folds that are sampled in nature are the limit of what’s possible; or whether there are quite different folds that are also possible,” he said. “Basically, we want to understand whether nature only sampled a subset of what’s possible,” said Baker.
The challenges of designing an amino acid sequence to fold into a new structure were considerable, said Baker. “If you draw on the back of an envelope some arbitrary protein structure, it might be that there is simply no amino acid sequence that will fold up to that structure. We had to develop methods to computationally sample possible structures similar to the one drawn on the back of the envelope, searching for a conformation for which there exists a very low energy amino acid sequence,” he said.
Baker and his colleagues took advantage of methods for sampling alternative protein structures that they have been developing for some time as part of the Rosetta ab initio protein structure prediction methodology. “Indeed, the integration of protein design algorithms (to identify low energy amino acid sequences for a fixed protein structure) with protein structure-prediction algorithms (which identify low energy protein structures for a fixed amino acid sequence) was a key ingredient of our success,” Baker said. He likened the problem to the three-dimensional version of attempting to create a specified outline for a jigsaw puzzle, given only a certain number of pieces—the equivalent of the 20 known amino acids in nature. In addition, he said, these amino acids can rotate into a number of different conformations.
“At each position, you can have one of the twenty amino acids, and for each of those amino acids you can have on the order of ten different shapes,” he said. “So, you have two hundred different possible shapes for each piece. With those restrictions, it may be that there are some outlines to this jigsaw puzzle that you just cannot achieve. So you need to have a way of changing the boundary to find a...