Engineering and Assay of Enzymes for Expanded Genetic Alphabet Replication

Use of noncanonical nucleotides for nucleic acid replication provides a critical way to understand how and why natural nucleotides were chosen by evolution and provides potential applications in biotechnology, diagnostics, therapeutics, and DNA computation. However, monitoring noncanonical nucleotide incorporation can be difficult, especially for high-throughput applications. Therefore, we have utilized a 6-letter genetic alphabet (G, A, T, C, Z, and P) for toehold-mediated DNA strand exchange assays. Toehold hybridization and branch migration were found to proceed readily through the formation of multiple consecutive non-natural base pairs between hydrogen bonding synthetic nucleotides dZ (6-amino-5-nitro-2(1H)-pyridone) and dP (2-amino-imidazo[1,2-a]-1,3,5-triazin-4(8H)one). Unlike other mismatches, a dZ:dG mis-pair was minimally disruptive to strand exchange. Overall, however, positional and numerary effects on mismatch tolerance could be leveraged to distinguish both dP and dZ mismatches. The success of the assay allowed a rapid assessment of the semisynthetic DNA replication by Escherichia coli DNA polymerases I, II, and IIIα and the identification of E. coli and Geobacillus stearothermophilus DNA Pol I mutants with improved fidelity. Given the difficulty of adapting organisms for expanded genetic alphabets, the ability to rapidly proof sequences and enzymes in vitro should lead to new parts and circuits that can be modularly introduced to improve the incorporation of more and more different noncanonical nucleotides.