Tautomeric equilibria of isoguanine and related purine analogs

Nucleobase pairs in DNA match hydrogen-bond donor and acceptor groups on the nucleobases. However, these can adopt more than one tautomeric form, and can consequently pair with nucleobases other than their canonical complements, possibly a source of natural mutation. These issues are now being re-visited by synthetic biologists increasing the number of replicable pairs in DNA by exploiting unnatural hydrogen bonding patterns, where tautomerism can also create mutation. Here, we combine spectroscopic measurements on methylated analogs of isoguanine tautomers and tautomeric mixtures with statistical analyses to a set of isoguanine analogs, the complement of isocytosine, the 5th and 6th "letters" in DNA.

Synthesis and Enzymology of 2'-Deoxy-7-deazaisoguanosine Triphosphate and Its Complement: A Second Generation Pair in an Artificially Expanded Genetic Information System

As with natural nucleic acids, pairing between artificial nucleotides can be influenced by tautomerism, with different placements of protons on the heterocyclic nucleobase changing patterns of hydrogen bonding that determine replication fidelity. For example, the major tautomer of isoguanine presents a hydrogen bonding donor-donor-acceptor pattern complementary to the acceptor-acceptor-donor pattern of 5-methylisocytosine. However, in its minor tautomer, isoguanine presents a hydrogen bond donor-acceptor-donor pattern complementary to thymine. Calculations, crystallography, and physical organic experiments suggest that this tautomeric ambiguity might be "fixed" by replacing the N-7 nitrogen of isoguanine by a CH unit. To test this hypothesis, we prepared the triphosphate of 2'-deoxy-7-deazaiso-guanosine and used it in PCR to estimate an effective tautomeric ratio "seen" by Taq DNA polymerase. With 7-deazaisoguanine, fidelity-per-round was ∼92%. The analogous PCR with isoguanine gave a lower fidelity-per-round of ∼86%. These results confirm the hypothesis with polymerases, and deepen our understanding of the role of minor groove hydrogen bonding and proton tautomerism in both natural and expanded genetic "alphabets", major targets in synthetic biology.

Transcription, reverse transcription, and analysis of RNA containing artificial genetic components

Expanding the synthetic biology of artificially expanded genetic information systems (AEGIS) requires tools to make and analyze RNA molecules having added nucleotide "letters". We report here the development of T7 RNA polymerase and reverse transcriptase to catalyze transcription and reverse transcription of xNA (DNA or RNA) having two complementary AEGIS nucleobases, 6-amino-5-nitropyridin-2-one (trivially, Z) and 2-aminoimidazo[1,2a]-1,3,5-triazin-4(8H)-one (trivially, P). We also report MALDI mass spectrometry and HPLC-based analyses for oligomeric GACUZP six-letter RNA and the use of ribonuclease (RNase) A and T1 RNase as enzymatic tools for the sequence-specific degradation of GACUZP RNA. We then applied these tools to analyze the GACUZP and GACTZP products of polymerases and reverse transcriptases (respectively) made from DNA and RNA templates. In addition to advancing this 6-letter AEGIS toward the biosynthesis of proteins containing additional amino acids, these experiments provided new insights into the biophysics of DNA.

Autonomous assembly of synthetic oligonucleotides built from an expanded DNA alphabet. Total synthesis of a gene encoding kanamycin resistance

BACKGROUND

Many synthetic biologists seek to increase the degree of autonomy in the assembly of long DNA (L-DNA) constructs from short synthetic DNA fragments, which are today quite inexpensive because of automated solid-phase synthesis. However, the low information density of DNA built from just four nucleotide "letters", the presence of strong (G:C) and weak (A:T) nucleobase pairs, the non-canonical folded structures that compete with Watson-Crick pairing, and other features intrinsic to natural DNA, generally prevent the autonomous assembly of short single-stranded oligonucleotides greater than a dozen or so.

RESULTS

We describe a new strategy to autonomously assemble L-DNA constructs from fragments of synthetic single-stranded DNA. This strategy uses an artificially expanded genetic information system (AEGIS) that adds nucleotides to the four (G, A, C, and T) found in standard DNA by shuffling hydrogen-bonding units on the nucleobases, all while retaining the overall Watson-Crick base-pairing geometry. The added information density allows larger numbers of synthetic fragments to self-assemble without off-target hybridization, hairpin formation, and non-canonical folding interactions. The AEGIS pairs are then converted into standard pairs to produce a fully natural L-DNA product. Here, we report the autonomous assembly of a gene encoding kanamycin resistance using this strategy. Synthetic fragments were built from a six-letter alphabet having two AEGIS components, 5-methyl-2'-deoxyisocytidine and 2'-deoxyisoguanosine (respectively S and B), at their overlapping ends. Gaps in the overlapped assembly were then filled in using DNA polymerases, and the nicks were sealed by ligase. The S:B pairs in the ligated construct were then converted to T:A pairs during PCR amplification. When cloned into a plasmid, the product was shown to make Escherichia coli resistant to kanamycin. A parallel study that attempted to assemble similarly sized genes with optimally designed standard nucleotides lacking AEGIS components gave successful assemblies of up to 16 fragments, but generally failed when larger autonomous assemblies were attempted.

CONCLUSION

AEGIS nucleotides, by increasing the information density of DNA, allow larger numbers of DNA fragments to autonomously self-assemble into large DNA constructs. This technology can therefore increase the size of DNA constructs that might be used in synthetic biology.