Bioinformatics and cell-free protein expression for enzyme engineering.

摘要

Bioinformatics and cell-free protein expression have shown themselves to be natural allies to metabolic engineering efforts. Bioinformatic techniques require information in order to tie proteins to each other, to tie sequence to structure and function. Here we demonstrate the use of the Bayesian sequence-based bioinformatics algorithms PROBE and Classifier to mine nature's existing data on protein function. We identify a low-scoring motif, residues 193--208, a strand-turn-strand motif, in the alignment model of the subtilisin superfamily, and make the functional prediction of association with thermal stability without resort to structural information. We validate the hypothesis by engineering increased thermal stability into the non-thermally stable protease subtilisin E.; This technique is most useful when functional information is available for a sample of the members of a protein superfamily. Cell-free protein synthesis provides a quick way to synthesize and screen small quantities of protein in large numbers. We demonstrate the use of cell-free protein synthesis in a non-natural environment, on surface-bound DNA, such as may be used in a microfluidic biochip. The use of cell-free protein synthesis on surface-attached DNA also bodes well for the demonstration of functionality of DNA in other non-natural environments, such as bound to carbon nanotubes.; In addition to utilizing in vitro protein synthesis in a biochip-type environment, we have also demonstrated the synthesis of two metabolic pathway enzymes, malonyl-CoA synthetase and THN synthetase. We show functionality of both enzymes, including of a previous uncharacterized putative malonyl-coA synthetase.