Selective Nonenzymatic Formation of Biologically Common RNA Hairpins

The prebiotic formation of RNA building blocks is well-supported experimentally, yet the emergence of sequence- and structure-specific RNA oligomers is generally attributed to biological selection via Darwinian evolution rather than prebiotic chemical selectivity. In this study, we used deep sequencing to investigate the partitioning of randomized RNA overhangs into ligated products by either splinted ligation or loop-closing ligation. Comprehensive sequence-reactivity profiles revealed that loop-closing ligation preferentially yields hairpin structures with loop sequences UNNG, CNNG, and GNNA (where N represents A, C, G, or U) under competing conditions. In contrast, splinted ligation products tended to be GC rich. Notably, the overhang sequences that preferentially partition to loop-closing ligation significantly overlap with the most common biological tetraloops, whereas the overhangs favoring splinted ligation exhibit an inverse correlation with biological tetraloops. Applying these sequence rules enables the high-efficiency assembly of functional ribozymes from short RNAs without template inhibition. Our findings suggest that the RNA tetraloop structures that are common in biology may have been predisposed and prevalent in the prebiotic pool of RNAs, prior to the advent of Darwinian evolution. We suggest that the one-step prebiotic chemical process of loop-closing ligation could have favored the emergence of the first RNA functions.