20 In Vitro Selection of Functional RNA Sequences

Abstract

A critical but seldom articulated assumption common to models of the evolution of an RNA world is that prebiotically synthesized polynucleotides of random sequence were the source of the first self-replicating molecules. However, the probability of finding functional molecules of any sort in a mixture of random sequences is unknown. A high probability would favor the origin of life, whereas a very low probability would make the origin of life more difficult.

To address this problem, we developed an iterative method for the isolation of nucleic acid sequences with specific ligand-binding or catalytic properties from very large pools of random sequences. The key to this method is the use of cycles of in vitro selection to enrich the pool for RNA species with the desired properties, followed by amplification of the selected molecules. This cycle of in vitro selection and amplification can be repeated as many times as necessary to isolate a population of molecules composed entirely of binding species, which can then be cloned and characterized individually. We used cycles of affinity chromatography followed by polymerase chain reaction (PCR) amplification to isolate ligand-binding RNAs and DNAs. We found that roughly 10⁻¹⁰ of the original sequences in our random pools were able to bind to organic dyes and other small ligands, including amino acids. If these results can be extrapolated to the binding of the transition states of chemical transformations, it is likely that a wide range of RNA catalysts might be found in pools of random sequence RNA molecules...