Application of the Knob Socket Model to predict changes in alpha helical structure

The Knob Socket (KS) model is a 4 amino acid motif that describes the way a protein will fold and pack its residues to form tertiary structure. The model includes a one amino acid residue knob from one secondary structure that packs into a three amino acid residue socket of another secondary structure. The α-helical sockets can be placed into three different categories: (1) free, unpacked and favoring intra-helical interactions, (2) filled, packed with a knob, and favoring inter-helical interactions, and (3) non, unpacked and disfavoring α-helical structure. Data within the Protein Data Bank was used to develop propensity libraries for each type of secondary structures. An α-helical propensity library was used to determine the relative frequency in which specific amino acid composition of sockets were free or filled. From this library and use of the KS model, a novel anti-parallel α-helical homodimer, KSα1.1, was designed. A single point mutation in the KSα1.1 sequence was incorporated in order to change the propensities of the surrounding six sockets and named a ‘hexagon.’ Values calculated from the difference between the total socket propensities for KSα1.1 and its corresponding mutated versions were used to predict changes in alpha helical content. Negative values corresponded to a predicted decrease in alpha helical content whereas positive values corresponded to a predicted increase in alpha helical content. Point mutations were made in the KSα1.1 sequence through the use of site-directed mutagenesis. To obtain high amounts of the desired mutated protein, plasmid vectors containing the specific point mutations in KSα1.1 sequence were transformed and expressed in E.coli. The transformed cells were induced for protein expression with Isopropyl β-D-1-thiogalactopyranoside (IPTG) and purified via column chromatography. The mutated versions of KSα1.1 were analyzed via circular dichroism (CD) spectroscopy to confirm predictions made using the KS model and propensity libraries. Deconvolutions were used to analyze the CD graphs and determine the percent content of alpha helix, beta sheet, and random coil structures. Mutant KSα1.1 proteins were compared to wild-type KSα1.1 protein in order to analyze changes in higher ordered protein packing and alpha helical structure