De novo design of a library of beta-sheet protein

De novo design tests the ability of a particular amino acid sequence to fold into a pre-determined three dimensional structure. The structure of the designed proteins can give insight into the characteristics ofthe primary sequence which are important for protein folding. An important feature of a proteins primary sequence is that amino acid side chains are either polar or nonpolar. The folded native structure of proteins possess well defined interiors that are comprised mainly of nonpolar residues from which solvent is excluded. It has been hypothesized that the burial of nonpolar residues is the main force that causes a protein to collapse from an unfolded random coil to a folded native state. If this hypothesis is correct, it should be possible to design proteins which fold into predetermined structures by designing interior amino acids to be nonpolar and exterior amino acids to be polar. This "binary code" strategy has previously been used to design a library of $alpha$-helical proteins.;This thesis describes the de novo design of a library of $beta$-sheet proteins using the binary code strategy. The library was created by constructing synthetic genes that used degenerate codons to encode the nonpolar (Phe, Leu, Ile and Met) and polar (Glu, Gln, Asp, Asn, His and Lys) positions. This produced a collection of proteins in which each sequence is unique, but all have the identical pattern of polar and nonpolar amino acids. Several of these proteins were expressed in E. coli and purified. Further characterization demonstrates that some of these proteins have significant $beta$-sheet secondary structure. Thus a simple binary code of polar and nonpolar amino acids arranged in the appropriate order can drive polypeptide chains to collapse into globular $beta$-sheet folds.