Evolution of the substrate specificity of an RNA ligase ribozyme from phosphorimidazole to triphosphate activation

The acquisition of new RNA functions through evolutionary processes was essential for the diversification of RNA-based primordial biology and its subsequent transition to modern biology. However, the mechanisms by which RNAs access new functions remain unclear. Do RNA enzymes need completely new folds to support new but related functions, or is reoptimization of the active site sufficient? What are the roles of neutral and adaptive mutations in evolutionary innovation? Here, we address these questions experimentally by focusing on the evolution of substrate specificity in RNA-catalyzed RNA assembly. We use directed in vitro evolution to show that a ligase ribozyme that uses prebiotically relevant 5'-phosphorimidazole-activated substrates can be evolved to catalyze ligation with substrates that are 5'-activated with the biologically relevant triphosphate group. Interestingly, despite catalyzing a related reaction, the new ribozyme folds into a completely new structure and exhibits promiscuity by catalyzing RNA ligation with both triphosphate and phosphorimidazole-activated substrates. Although distinct in sequence and structure, the parent phosphorimidazolide ligase and the evolved triphosphate ligase ribozymes can be connected by a series of point mutations where the intermediate sequences retain at least some ligase activity. The existence of a quasi-neutral pathway between these distinct ligase ribozymes suggests that neutral drift is sufficient to enable the acquisition of new substrate specificity, thereby providing opportunities for subsequent adaptive optimization. The transition from RNA-catalyzed RNA assembly using phosphorimidazole-activated substrates to triphosphate-activated substrates may have foreshadowed the later evolution of the protein enzymes that use monomeric triphosphates (nucleoside triphosphates, NTPs) for RNA synthesis.