[A stereochemical model of codon-anticodon interactions, in which bases bind via water bridges].

Analysis of available experimental data has allowed us to conclude that upon codon-anticodon interaction there is a functional asymmetry between the third nucleotide of the codon and the first nucleotide of the anticodon. We suggest that this phenomenon would be due to the differential codon and anticodon conformation mobilities, which are somehow restricted. Two postulates are proposed to restrict these mobilities respectively: (1) In the codon-anticodon complex the codon is in the fixed A-RNA conformation, and the rearrangement of the sugar-phosphate backbone, that is necessary for any wobble-pair formation occurs only in the anticodon. (2) This rearrangement does not alter the specific system of contacts between the universal uridine and other nucleotides in the anticodon loop, observed in tRNA crystals. Stereochemical analysis revealed that these restrictions make the construction of any Crick's wobble-pair sterically impossible. To resolve this contradiction we propose that two bases in the third position of the codon-anticodon complex can mate not only according to the Crick's scheme, but also with the help of water bridges. Consequently, besides the standard pairs, the following pairs become permitted: UG, UI, UU, CU, GU and AI (it should be noted that GU and AI are weaker that the other ones). All other pairs appear sterically impossible. These conclusions are in good agreement with the literature date.