UC Riverside

The exquisite structures and properties of nucleic acids enable them to function in information storage and transfer. Efforts to extend encoding and copy-ing of genetic information to unnatural nucleobases have been made over the years for potential usage in molecular engineering, in addition to expansion of the genetic alphabet. We have explored alternative base pairing through metal com-plexation in the DNA double helix. Synthetic nucleobase deoxyribosides of purine-6-carboxylate (PurC) and purine-2,6-dicarboxylate (PurDC) deoxyribosides were found to pair with their complementary nucleobases, PurC, 6-(2'-pyridyl)-purine (PurP) and 3-pyridyl (3-Py) deoxyribosides, through preferential binding to Cu2+. All of the metallo base-pairs exhibited high thermal stability in DNA du-plexes, and mismatched base-pairs between unnatural and natural nucleobases were highly destabilizing.

The DNA stabilizing forces of the metal-mediated base-pairs were utilized to guide polymerization of unnatural nucleotides in the presence of DNA poly-merases. Incorporation of PurC or PurDC mononucleotide against its complemen-tary base in the template was Cu2+ concentration dependent, suggesting that nu-cleotide insertion occurred via metallo base-pair formation and that polymerases can recognize this new metal-mediated base pairing. Although no enzymatic se-lectivity for incorporation of unnatural nucleotides over natural nucleotides was observed, sequential incorporation of PurDC and natural nucleotides allowed for the incorporation and subsequent extension of metallo base-pairs by DNA poly-merases. These results are significant in that metal-mediated unnatural base-pair involving replication is unprecedented.

Artificial nucleic acids based on simplified backbones are of great interest as a potential evolutionary progenitor of RNA. While glycol nucleic acid (GNA), which has an acyclic three-carbon backbone and follows the Watson-Crick base pairing rules with higher thermal stability than DNA or RNA has been considered as a candidate progenitor of RNA, the evolutionary relationship between GNA and RNA is unclear due to limited ability of GNA to form heteroduplexes with RNA. We investigated nonenzymatic oligomerization of RNA on GNA templates. However, GNA templates failed to direct oligomerization of activated ribonucleo-tide monomers. Also, cross-pairing experiments with inosine GNA, a guanosine analog, and cytidine in RNA did not show an improvement of the cross-pairing properties between GNA and RNA.