A kinetic error filtering mechanism for enzyme-free copying of nucleic acid sequences

Accurate copying of nucleic acid sequences is essential for self-replicating systems. Modern cells achieve error ratios as low as 10-9 with sophisticated enzymes capable of kinetic proofreading. In contrast, experiments probing enzyme-free copying of RNA and DNA as potential prebiotic replication processes find error ratios on the order of 10%. Given this low intrinsic copying fidelity, plausible scenarios for the spontaneous emergence of molecular evolution require an accuracy-enhancing mechanism. Here, we study a 'kinetic error filtering' scenario that dramatically boosts the likelihood of producing exact copies of nucleic acid sequences. The mechanism exploits the observation that initial errors in template-directed polymerization of both DNA and RNA are likely to trigger a cascade of consecutive errors and significantly stall downstream extension. We incorporate these characteristics into a mathematical model with experimentally estimated parameters, and leverage this model to probe to what extent accurate and faulty polymerization products can be kinetically discriminated. While limiting the time window for polymerization prevents completion of erroneous strands, resulting in a pool in which full-length products show an enhanced accuracy, this comes at the price of a concomitant reduction in yield. We show that this fidelity-yield trade-off can be circumvented via repeated copying attempts in cyclically varying environments such as the temperature cycles occurring naturally in the vicinity of hydrothermal systems. This setting could produce exact copies of sequences as long as 50mers within their lifetime, facilitating the emergence and maintenance of catalytically active oligonucleotides.