Over the past 10 years an increasing number of proteins that don't exist in nature have been designed "in silico" (in a computer). Scientists use sophisticated protein modeling software that incorporates the relevant laws of physics and chemistry to find amino acid sequences that fold into stable forms and have specific functions.
Imagine making a necklace 10 beads long with beads that come in 20 different colors. There are more than 10 trillion different combinations to choose among. This provides an idea of the complexity involved in designing novel proteins. For a protein of a given size, the modeling software creates millions of versions by putting each amino acid in every position and evaluating the stability of the resulting molecule. This takes a tremendous amount of computing power which skyrockets as the length of the protein increases.
"The current limit of this approach, even using the fastest supercomputers, is about 120 amino acids," said Meiler. The previous record holder contained 106 amino acids. The newly designed protein contains 242 amino acids. The Vanderbilt group got around this limit by modifying the widely used protein engineering platform called ROSETTA so that it can incorporate symmetry in the design process.
Their success provides new support for a controversial theory about protein evolution called the gene duplication and fusion hypothesis. The advantage of small proteins is that they can evolve rapidly in response to changing conditions, but larger proteins can perform more complex functions. Nature found a way to get both advantages by selecting small proteins that can interact with other copies of themselves to form larger proteins, which are called dimers. Once useful dimers have
|Contact: David F. Salisbury|