Biochemical and computational exploration of heterocyclic substrate recognition and discrimination by eubacterialtRNA guanine transglycosylase

The eubacterial enzyme tRNA guanine transglycosylase (TGT) catalyzes the base exchange of guanine-34 in tRNAsTyr, His, Asn, and Asp for preQ₁, a precursor of the modified base queuine. Using both biochemical and computational approaches, it was possible to investigate heterocyclic substrate recognition and to probe the discrimination of non-substrates of E. coli TGT. This work was accomplished in three specific aims. The first aim was the determination of the role of aspartic acid 143 in TGT. The second aim was the creation of a computational model of TGT for exploration of molecular recognition and prediction of ligand affinity. The last aim was to probe biochemically and computationally the ability of D143 mutants of TGT to recognize alternate substrates.; It was experimentally determined that aspartic acid 143 in E. coli TGT plays a role in heterocycle recognition. A homology model of E. coli TGT was built to complement these biochemical studies, providing a more thorough investigation of the role of D143. Using the Linear Interaction Energy method, free energies of binding of guanine to wild-type and D143 mutant TGTs were computationally determined. They reproduce the experimental results very well, indicating a reliable and accurate computational model of TGT.; Once the role of D143 in heterocyclic substrate recognition had been elucidated, the next logical step was to determine whether substrate specificity of D143 mutant TGTs was altered. Using biochemical and computational methods, it was determined that the substrate specificity of D143N TGT was indeed altered to prefer xanthine over guanine. It was also possible to extend the use of this model to predict the binding of other compounds: preQ₁, 2-oxo-preQ ₁, and 3-methyl-preQ₁. A complete inversion of specificity was predicted for both alternate substrates; 2-oxo-preQ₁ was predicted to be best recognized by D143N, while 3-methyl-preQ₁ was predicted to be best recognized by D143S.; This work has provided a solid foundation for the use of multiple techniques to investigate heterocyclic substrate recognition. Additionally, since this enzyme has been shown to be partially responsible for the virulence of the disease Shigellosis, it is possible to use these complementary techniques to enhance the efficiency of drug design.