Why Pentose- And Not Hexose-Nucleic Acids? Purine-Purine Pairing In Homo-Dna: Guanine, Isoguanine, 2,6-Diaminopurine, And Xanthine
Katrin Groebke,J�rg Hunziker,William Fraser,Ling Peng,Ulf Diederichsen,Kaspar Zimmermann,Armin Holzner,Christian Leumann,Albert Eschenmoser
DOI: https://doi.org/10.1002/hlca.19980810302
1998-01-01
Helvetica Chimica Acta
Abstract:This paper concludes the series of reports in this journal [1-4] on the chemistry of homo-DNA, the constitutionally simplified model system of hexopyranosyl-(6' --> 4')-oligonucleotide systems studied in our laboratory as potentially natural-nucleic-acid alternatives in the context of a chemical aetiology of nucleic-acid structure. The report describes the synthesis and pairing properties of homo-DNA oligonucleotides which contain as nucleobases exclusively purines, and gives, together with part III of the series [3], a survey of what we know today about purine-purine pairing in homo-DNA. In addition, the paper discusses those aspects of the chemistry of homo-DNA which, we think, influence the way how some of the structural features of DNA (and RNA) are to be interpreted on a qualitative level. Purine-purine pairing occurs in the homo-DNA domain in great variety. Most prominent is a novel tridentate Watson-Crick pair between guanine and isoguanine, as well as one between 2,6-diaminopurine and xanthine, both giving rise to very stable duplexes containing the all-purine strands in antiparallel orientation. For the guanine-isoguanine pair, constitutional assignment is based on temperature-dependent UV and CD spectroscopy of various guanine-and isoguanine-containing duplexes in comparison with duplexes known to be paired in the reverse-Hoogsteen mode. The assignment is supported by the characteristic changes observed in pairing behavior when guanine is replaced by 7-carbaguanine. Isoguanine and 2,6-diaminopurine also have the capability of self-pairing in the reverse-Hoogsteen mode, as previously observed for adenine and guanine [3]. In this type of pairing, the purine bases that contain an amino group in the 6-position (adenine, a,6-diaminopurine, and isoguanine) behave interchangeably. Fig. 36 provides an overall survey of the relative strength of pairing in all possible purine-purine combinations. Watson-Crick pairing of isoguanine with guanine demands the former to participate in its 3H-tautomeric form; hitherto this specific tautomer had not been considered in the pairing chemistry of isoguanine. Whereas (cumulative) purine-purine pairing in DNA (reverse-Hoogsten or Hoogsteen) seems to occur in triplexes and tetraplexes only, its occurrence in duplexes is a characteristic feature of homo-DNA chemistry. The occurrence of purine-purine Watson-Crick base pairs is probably a consequence of homo-DNA's quasi-linear ladder structure [1][4]. In a double helix, the distance between the two sugar C atoms, on which a base pair is anchored, is expected to be constrained by the dimensions of the helix; in a linear duplex, however, there would be no restrictions with regard to base-pair length. Homo-DNA's ladder-like model also allows one to recognize one of the reasons why nucleic-acid duplexes prefer to pair in antiparallel, rather than parallel strand orientation: in homo-DNA duplexes: (averaged) backbone and base pair axes are strongly inclined toward one another [4]; the stronger this inclination, the higher the preference for antiparallel strand orientation is expected to be (Fiq. 16). In retrospect, homo-DNA turns out to be one of the first artificial oligonucleotide systems (cf. Footnote 6.5) to demonstrate in a comprehensive way that informational base pairing involving purines and pyrimidines is not a capability unique to ribofuranosyl systems. Stability and helical shape of pairing complexes are not necessary conditions of one another; it is the potential for extensive conformational cooperativity of the backbone structure with respect to the constellational demands of base pairing and base stacking that determines whether or not a given type of base-carrying backbone structure is an informational pairing system. From the viewpoint of the chemical aetiology of nucleic-acid structure, which inspired our investigations on hexopyranosyl-(6' --> 4')-oligonucleotide systems in the first place, the work on homo-DNA is only an extensive model study, because homo-DNA is not to be considered a potential natural-nucleic-acid alternative. In retrospect. it seems fortunate that the model study was carried out, because without it we could hardly have comprehended the pairing behavior of the proper nucleic-acid alternatives which we have studied later and which will be discussed in Part VI of this series.