Controlling enzyme inhibition using an expanded set of genetically encoded amino acids

Enzyme inhibition plays an important role in drug development, metabolic pathway regulation, and biocatalysis with product inhibition. When an inhibitor has high structural similarities to the substrate of an enzyme, controlling inhibitor binding without affecting enzyme substrate binding is often challenging and requires fine-tuning of the active site. We hypothesize that an extended set of genetically encoded amino acids can be used to design an enzyme active site that reduces enzyme inhibitor binding without compromising substrate binding. As a model case, we chose murine dihydrofolate reductase (mDHFR), substrate dihydrofolate, and inhibitor methotrexate. Structural models of mDHFR variants containing non-natural amino acids complexed with each ligand were constructed to identify a key residue for inhibitor binding and non-natural amino acids to replace the key residue. Then, we discovered that replacing the key phenylalanine residue with two phenylalanine analogs (p-bromophenylalanine (pBrF) and L-2-naphthylalanine (2Nal)) enhances binding affinity toward the substrate dihydrofolate over the inhibitor by 4.0 and 5.8-fold, respectively. Such an enhanced selectivity is mainly due to a reduced inhibitor binding affinity by 2.1 and 4.3-fold, respectively. The catalytic efficiency of the mDHFR variant containing pBrF is comparable to that of wild-type mDHFR, whereas the mDHFR variant containing 2Nal exhibits a moderate decrease in the catalytic efficiency. The work described here clearly demonstrates the feasibility of selectively controlling enzyme inhibition using an expanded set of genetically encoded amino acids.