Nucleobase tautomerism in codon-anticodon decoding

Abstract

Accurate recognition of base pairs in the processes of Central Dogma is the basis of faithful replication and expression of genetic information. Among the possible sources of errors in this processes, G◦U mismatch recognition during codon-anticodon decoding in translation has the highest error rate. This has been linked to the occurrences of rare enol tautomers of nucleobases, which enable formation of the Watson-Crick (WC) geometry of this mismatch from the wobble (wb) geometry formed with canonical keto tautomers. The WC geometry of G◦U was observed in the environment of the closed ribosomal decoding site in equilibrium structural studies. This observation currently lacks a physicochemical explanation as well as a consistent model to reconcile it with the mechanism of decoding. To address this problem, we studied effects of the decoding site on the wobble↔WC tautomerization reaction in G◦U (wb-WC reaction). Using quantum-mechanical/molecular-mechanical umbrella sampling simulations, we found this reaction to be exoergic in the closed state of the decoding site, but endoergic in the open state. We also calculated the dielectric constant of the decoding site and revealed its decrease in the closed state. Together, these results provide an explanation to the structural observations. To reconcile the stabilization of the WC geometry with the mechanism of base pair recognition in codon-anticodon decoding, we developed a new kinetic model of decoding that incorporates the wb-WC reaction parameters in the open and closed states of the decoding site. In this model, the exoergic wb-WC reaction is kinetically restricted by the decoding rates. This model explains the observations of the WC geometry at equilibrium conditions, thereby uniting structural and kinetic data. Moreover, the model reveals constraints imposed by the exoergic wb-WC reaction on the decoding accuracy: equilibration of the reaction counteracts equilibration of the open-closed transition. Our model can be a step towards a general recognition model for flexible substrates. We applied this model, supported by additional computational studies, to provide a putative mechanism of how a specific U modification in anticodon can facilitate decoding of both A- and G-ending codons.