| Protein binding to small molecules is a critical component of biological function. The manipulation of these interactions is therefore important for basic biochemical studies and has numerous biotechnological applications, including the construction of biosensors for arbitrary ligands, as well as the production of enzymes and signal transduction pathways with novel function. Computational design techniques offer generality for the engineering of protein function, and can address many more potential protein sequences than combinatorial methods. Improvements in the in silico representation of biomolecular structure and function, as well as algorithms for the systematic search of sequence space, have facilitated the construction of a computational method (the ReceptorDesign algorithm) that can drastically redesign protein ligand-binding specificities, with the preservation of overall protein architecture and stability. The ReceptorDesign algorithm has been used to recapitulate wild-type protein sequences for a family of bacterial periplasmic binding proteins, as well as to design novel protein binding partners for trinitrotoluene (TNT), L-lactate, and serotonin. The designed receptors have been shown to bind their target ligands with high affinity and specificity, and were incorporated into synthetic bacterial signal transduction pathways, regulating gene expression in response to extracellular TNT or lactate. The use of diverse ligands and proteins demonstrates the generality of the computational method for the redesign or de novo design of ligand-binding functionality in proteins and enzymes. |
