Toward the selection of ribozymes for 1,3-dipolar cycloaddition reactions.

In vitro selection from combinatorial RNA libraries has repeatedly been used to study the catalytic and binding potential of nucleic acids. These selections not only led to RNA sequences catalyzing transformations known from metabolic pathways but also generated novel ribozymes for typical organic reactions. We were interested in 1,3-dipolar cycloaddition reactions, which are important tools for the formation of heterocyclic systems in organic chemistry and might also be found in the hypothetic RNA world. Here we describe our strategy and experiments to isolate RNA molecules catalyzing a 1,3-dipolar cycloaddition between nitrile oxides and an acrylate conjugated to RNA. We used direct selection with linker-coupled reactants, which has previously allowed the generation of true trans-acting catalysts for bimolecular reactions. A photocleavable linker was introduced to provide for a more stringent selection criterion. The 1,3-dipolar cycloaddition reaction was established in aqueous solution using a modified dinucleotide that was tethered to the dipolarophilic substrate. Two selection protocols were established, namely, a low-stringency affinity-based selection protocol, and a high-stringency procedure using the photocleavable moiety. In neither case was an increased activity toward the desired reaction obtained after 15 and 11 selection rounds, respectively. The resulting pools of RNA from several rounds were investigated both in cis and in trans. The limitations of this selection methodology are discussed in comparison with other catalysts for dipolar cycloadditions and, also, with respect to the unconventional substrates used.