Rational design of aminoacyl-tRNA synthetase specific for p-acetyl-l-phenylalanine

Abstract

The Methanococcus jannaschii tRNATyr/tyrosyl-tRNA synthetase pair has been engineered to incorporate unnatural amino acids into proteins in Escherichia coli site-specifically. In order to add other unnatural amino acids into proteins by this approach, the amino acid binding site of M. jannaschii tyrosyl-tRNA synthetase need to be mutated. The crystal structures of M. jannaschii tyrosyl-tRNA synthetase and its mutations were determined, which provided an opportunity to design aminoacyl-tRNA synthetases specific for other unnatural amino acids. In our study, we attempted to design aminoacyl-tRNA synthetases being able to deliver p-acetyl-l-phenylalanine into proteins. p-Acetyl-l-phenylalanine was superimposed on tyrosyl in M. jannaschii tyrosyl-tRNA synthetase–tyrosine complex. Tyr32 needed to be changed to non-polar amino acid with shorter side chain, Val, Leu, Ile, Gly or Ala, in order to reduce steric clash and provide hydrophobic environment to acetyl on p-acetyl-l-phenylalanine. Asp158 and Ile159 would be changed to specific amino acids for the same reason. So we designed 60 aminoacyl-tRNA synthetases. Binding of these aminoacyl-tRNA synthetases with p-acetyl-l-phenylalanine indicated that only 15 of them turned out to be able to bind p-acetyl-l-phenylalanine with reasonable poses. Binding affinity computation proved that the mutation of Tyr32Leu and Asp158Gly benefited p-acetyl-l-phenylalanine binding. And two of the designed aminoacyl-tRNA synthetases had considerable binding affinities. They seemed to be very promising to be able to incorporate p-acetyl-l-phenylalanine into proteins in E. coli. The results show that the combination of homology modeling and molecular docking is a feasible method to filter inappropriate mutations in molecular design and point out beneficial mutations.