Overcoming nucleotide bias in the nonenzymatic copying of RNA templates

The RNA World hypothesis posits that RNA was the molecule of both heredity and function during the emergence of life. This hypothesis implies that RNA templates can be copied, and ultimately replicated, without the catalytic aid of evolved enzymes. A major problem with nonenzymatic templated polymerization has been the very poor copying of sequences containing rA and rU. Here we overcome that problem by using a prebiotically plausible mixture of RNA mononucleotides and random-sequence oligonucleotides, all activated by methyl isocyanide chemistry, that direct the uniform copying of arbitrary-sequence templates, including those harboring rA and rU. We further show that the use of this mixture in copying reactions suppresses copying errors while also generating a more uniform distribution of mismatches than observed for simpler systems. We find that oligonucleotide competition for template binding sites, oligonucleotide ligation, and the template binding properties of reactant intermediates work together to reduce product sequence bias and errors. Finally, we show that iterative cycling of the activation chemistry and templated polymerization improves the yield of random-sequence products. These results for random-sequence template copying are a significant advance in the pursuit of nonenzymatic RNA replication.