Knowledge of how small molecules interact with proteins can be used to design the proteins to better interact with those small molecules. Our lab uses the Rosetta modeling package to computationally predict which protein mutants would improve binding to particular small molecules. The approach can even be used to improve enzyme performance, by modeling the affinity of the mutant proteins to a computational model of the transition state. Subsequent experimental testing and experiments such as directed evolution can futher refine these binding interfaces.
References:
[1]
Moretti R, Bender BJ, Allison B, Meiler J. Rosetta and the Design of Ligand Binding Sites. Methods Mol Biol. 2016;1414:47-62. doi: 10.1007/978-1-4939-3569-7_4. PMID: 27094285; PMCID: PMC5511788.
[2]
Allison B, Combs S, DeLuca S, Lemmon G, Mizoue L, Meiler J. Computational design of protein-small molecule interfaces. J Struct Biol. 2014 Feb;185(2):193-202. doi: 10.1016/j.jsb.2013.08.003. Epub 2013 Aug 17. PMID: 23962892; PMCID: PMC3946393.
[3]
Nannemann DP, Kaufmann KW, Meiler J, Bachmann BO. Design and directed evolution of a dideoxy purine nucleoside phosphorylase. Protein Eng Des Sel. 2010 Aug;23(8):607-16. doi: 10.1093/protein/gzq033. Epub 2010 Jun 4. PMID: 20525731; PMCID: PMC2898500.